| |
Last updated on August 8, 2023. This conference program is tentative and subject to change
Technical Program for Saturday August 19, 2023
|
SaPLPL |
Room D |
Plenary Talk: Sergej Fatikow |
Plenary Sessions |
Chair: Wang, Zhidong | Chiba Institute of Technology |
|
09:10-09:55, Paper SaPLPL.1 | |
Assembly, Fabrication, and Characterization at Nanoscale Using High-Precision Robotics |
|
Fatikow, Sergej | University of Oldenburg |
Keywords: Micro/Nano Robotics
Abstract: Current research activities in AMiR focus on high-precision microrobotics for industrial applications and automation of nanomaterial handling at nanoscale. The main focus of the talk is on current research projects in AMiR and related industrial applications. They include automated assembly of nanophotonic structures, nanorobotic handling of graphene, mechanical characterization of nanomaterials, electrical characterization of nanowires, surface adhesion mapping by nanowire probes, fabrication of nanogranular tunneling resistors, manipulation of nanodroplets, and others.
|
|
SaKNKN |
Room D |
Keynote Talk: Ben M. Chen |
Invited Sessions |
Chair: Chen, Fei | The Chinese University of Hong Kong |
|
09:55-10:30, Paper SaKNKN.1 | |
AI and UAS Technologies for Building Inspection and Information Management |
|
Chen, Ben M. | Chinese University of Hong Kong |
Keywords: Artificial Intelligence
Abstract: In this talk, we aim to present a fully autonomous and fully functional infrastructure inspection and information management system with advanced AI and multiple unmanned aerial systems (UAS) technologies. The system includes sophisticated unmanned aerial hardware platforms and software systems for automatic flight control, task and motion planning, artificial intelligent algorithms and software platform for image and infrared data processing, i.e., crack, spalling, delamination and other defect detections, and building information modeling (BIM) and management system integrated with detailed geographical information systems (GIS) and digital twin (DT) technologies. Compared with the manual inspection, the system that we have developed has the advantages of being more economical, safer, flexible and efficient. It can also be adopted for other industrial applications, including smart ocean and smart cities.
|
|
SaS1A |
Room D |
Award Session 1 |
Regular Sessions |
Chair: Su, Baiquan | Beijing University of Posts and Telecommunications |
Co-Chair: Chen, Fei | The Chinese University of Hong Kong |
|
10:50-11:02, Paper SaS1A.1 | |
Fusing Vision and Force : A Framework of Reinforcement Learning for Elastic Peg-In-Hole Assembly |
|
Dang, Renjun | Tsinghua University |
Hou, Zhimin | National University of Singapore |
Yang, Wenhao | Tsinghua University |
Chen, Rui | Tsinghua University |
Xu, Jing | Tsinghua University |
Keywords: Grasping and Manipulation, Intelligent Control and Systems, Intelligent Control
Abstract: Elastic Peg-in-Hole assembly has a wide range of applications in both industrial and home environments. However, accurately representing the infinite continuum of states and obtaining an accurate deformation model of the elastic deformable peg remains a challenge. Reinforcement learning (RL) has demonstrated its ability to learn manipulation skills from interactive experiences without the need for an exact physical model. Nonetheless, current RL methods rely on complex multimodal representations obtained using neural networks and lack clear physical interpretations. This article proposes a practical framework based on RL to interpret vision and force feedback through a concise, reformulated action space. The vision and force-based guidance are fused in the action space, and any continuous RL method can be adopted to learn the fusion pattern. Several experiments were conducted on an elastic peg-in-hole assembly platform to validate the effectiveness of the learned fusion scheme and its comparison to existing control and RL methods.
|
|
11:02-11:14, Paper SaS1A.2 | |
A Constant Force Tracking Strategy for Complex Surface Robots Fused with NURBS Speed Planning and Teaching/Learning Mode |
|
Duan, Jinjun | Nanjing University of Aeronautics and Astronaut |
Cui, Kunkun | Nanjing University of Aeronautics and Astronautics |
Guo, An | Nanjing University of Aeronautics and Astronautics |
Wang, Lingyu | Nanjing University of Aeronautics and Astronautics |
Bin, YiMing | Nanjing University of Aeronautics and Astronautics |
Wan, Minhong | Zhejiang Lab |
Huang, Qiulan | ZHEJIANG LAB |
Keywords: Industrial Robotics and Factory Automation, Path and Motion Planning
Abstract: Compliance control is particularly important in contact operations such as grinding and polishing, surface tracking, and medical diagnosis. However, due to the unknown environmental stiffness/position and dynamic changes, there are some problems, such as poor position/force tracking accuracy and a low degree of complex surface fit. To solve the above problems, a piecewise trajectory planning algorithm based on speed controllable NURBS and a teaching and learning mode constant force tracking strategy based on position/force coordination planning are proposed in this paper. First, a NURBS segmented trajectory planning algorithm with controllable speed is proposed by combining the NURBS trajectory planning algorithm with the T-type speed interpolation algorithm, and constant force tracking control based on the FCPressNURBS instruction set is realized by combining the adaptive admittance control algorithm. Second, based on the concept of teaching and learning modes, the position/force tracking parameters are obtained by the compliant control method, and the reinforcement learning algorithm is used to optimize the constant force tracking trajectory of the robot. Finally, the algorithm is tested on the xMate robot,and the results show that the proposed algorithm improves the constant force tracking accuracy of the robot to the complex contact surface in the heterogeneous environment and can realize the adaptive constant force tracking effect of the robot to the unknown environment.
|
|
11:14-11:26, Paper SaS1A.3 | |
Modeling and Parameter Identification for Human-Robot Coupled Systems in Powered Lower Limb Prostheses (I) |
|
Huang, Yongshan | Zhejiang Lab |
Wang, Xin | Zhejiang Lab |
Liang, Dingkun | Research Institute of Interdisciplinary Innovation, Zhejiang Lab |
Xiong, Jiaming | Peking University |
Xie, Anhuan | Zhejiang University |
Gu, Jason | Dalhousie University |
Keywords: Rehabilitation and Assistive Robotics, Human-Robot Interaction and Cooperation, Human-Machine Interface
Abstract: Good trajectory tracking results are the basis for the implementation of other upper-level algorithms in powered lower limb prostheses, which rely on dynamics modeling and good parameter identification. However, powered lower limb prosthesis are human-robot coupled systems with human-in-the-loop, and it has been a challenge to model them as well as to perform parameter identification. This paper proposes a modeling method that can realize the decoupling of human-machine, and by constructing the linearized models of human-machine coupled system and prosthetic subsystem about the physical parameter sets respectively, and finally obtaining the relationship of their corresponding physical parameter sets, thus the method of indirectly identifying the parameters of human-machine coupled system through the prosthetic sub-system can be realized. The experimental results verify the feasibility and effectiveness of the method proposed in this paper. And the experimental results also shows that prosthetic controller should take advantage of dynamic of hip motion in order to make the wearer more comfortable and fluid when swinging the leg, and will also make the prosthesis more energy efficient.
|
|
SaS1B |
Room E |
Control Methods and Mechatronic Designs towards Higher Level Autonomous
Medical Robots 1 |
Invited Sessions |
Chair: Su, Baiquan | Beijing University of Posts and Telecommunications |
Co-Chair: Liu, Wenyong | Beihang University |
Organizer: Su, Baiquan | Beijing University of Posts and Telecommunications |
|
10:50-11:00, Paper SaS1B.1 | |
A New Method for WFOV Camera Distortion Correction (I) |
|
Chen, Qian | BeiHang University |
Wei, Zhaoxiang | BeiHang University |
Tan, XinYuan | Beihang University |
Shen, Hongbin | BeiHang University |
Meng, Cai | Beihang University |
Keywords: Medical Robotics, Robot Vision and Computer Vision, Industrial Robotics and Factory Automation
Abstract: In laparoscopic surgery, the laparoscope has the advantage of a small volume and a large field of view, but their obvious distortion affect the doctor's judgment. Aimed at this limitation, we proposed a novel method to correct the image distortion while preserving the majority of the peripheral information. At first, a box-shaped surface is introduced before the lens to effectively capture the incident light from the fisheye image. Then, all the image pixels are mapped onto this box-shaped surface. Next, the image plane and the focal point are adjusted, moving them backward, and the pixels on the box-shaped surface are then re-projected onto the final focal plane to generate the corrected image. To verify the feasibility, the proposed method is compared with several classical approaches in terms of two designed comprehensive evaluation metrics, namely the signal retention rate (SRR) and the Offset of Hough Space Histogram (OHSH). The results demonstrate that the proposed method surpasses other classical methods both in SRR and OHSH.
|
|
11:00-11:10, Paper SaS1B.2 | |
Robotic Positioning in Percutaneous Transforaminal Endoscopic Discectomy Based on X-Ray Image and DLT Algorithm (I) |
|
Zhang, Tianyang | Beihang University |
Ji, Xuquan | Beihang University |
Zhu, Yuanyuan | Beihang University |
Zhang, Jian | Beihang University |
Geng, Baoduo | BeiHang University |
Xue, Yumei | Beihang University |
Hu, Lei | Beihang University |
Liu, Wenyong | Beihang University |
Keywords: Medical Robotics, Path and Motion Planning, Robot Vision and Computer Vision
Abstract: Percutaneous transforaminal endoscopic discectomy (PTED) is a decompression surgery on patients with lumbar disc herniation and spinal canal stenosis in a minimally invasive environment, which greatly shorten the rehabilitation cycle of patients. However, the puncture of traditional PTED is performed under non-direct vision, which relies on the surgeon's clinical experience heavily and can easily cause collateral damage. In this paper, we propose a robotic positioning method in PTED based on X-ray image and DLT algorithm. The end-effector with three rings (EETR) for PTED positioning has been specially designed. During the operation, the 3D-2D transformation matrix is calculated by direct linear transform (DLT) algorithm. Finally, the motion parameters are calculated to control the robot. Based on our method, only one X-ray image is needed to complete the positioning, which also works well in a large range of deflection angle between EETR and target puncture channel. Thus, it greatly simplifies the surgical process and reduces the radiation exposure time. By conducting a series of comparative experiments on the positioning of model bones, the translation error is less than 1.1mm and rotation error is less than 0.8°, when the deflection angle is less than 30°. And the translation error is 1.61mm and rotation error is 1.98°, when the angle reached 60°. It meets the requirements for PTED positioning, and has improved the positioning accuracy.
|
|
11:10-11:20, Paper SaS1B.3 | |
Manual Continuum Snake-Like Tube Tool Platform: A Low-Cost Solution with Intuitive Operation (I) |
|
Zhang, Haonan | Beihang University |
Pei, Xu | Beihang University |
Xiao, Feng | Beihang University |
Li, Wenhao | Beihang University |
Keywords: Industrial Robotics and Factory Automation, Medical Robotics, Rescue Robotics
Abstract: Working in narrow and complex spaces is a critical area of focus in industries and medical field. The development of tool for complex confined spaces has primarily concentrated on high-value sectors such as healthcare. Existing solutions with high flexibility and intuitive operation are generally not well-suited for occasional demands. This paper proposes a low-cost multi-purpose work platform for this pain point. It utilizes a motorless manual cable-driven structure, offering higher operational flexibility and intuitiveness compared to traditional segmented control methods used in continuum robots. An intuitive controller is designed based on the instantaneous center concept, and employed a self-decoupled orthogonal layout. The internal and external continuum tubes can switch between rigid and limp modes using the conjugate locking device, enabling the platform to encase nonlinear paths in 3D space with alternating stiffness. To compensate for inevitable small variations in cable lengths within the same cable set, an adjustable passive segmented force control mechanism is designed, which can enhance the safety of platform movements. The proposed work platform can be integrated with existing tools such as endoscopes, offering a versatile, scalable and low-cost solution for complex confined space operations.
|
|
11:20-11:30, Paper SaS1B.4 | |
Design and Analysis of an End-Effector with Dual RCM Mechanism for Robotic Extracorporeal Shock Wave Therapy in Orthopedics (I) |
|
Wang, Boyang | Beihang University |
Guo, Anyi | Beijing Jishuitan Hospital |
Liu, Yan | Beihang University |
Hou, Kedong | Beijing Pinggu District Hospitail |
Liu, Yajun | Beijing Jishuitan Hospital |
Liu, Wenyong | Beihang University |
Keywords: Medical Robotics, Emerging Technologies and Applications, Path and Motion Planning
Abstract: Robotic assistance in extracorporeal shock wave therapy (ESWT) has drawn attention in orthopedics for its automated operation instead of the time-lasting manual operation of physician, which dramatically alleviates the operation intensity of medical staff. It should be noticed that the dexterous operation in the small working space around the treatment site is an essential step for robotic ESWT. Considering the inverted cone shape of the working space around the treatment site in ESWT, this paper proposes and simulates a 3-DOFs robotic end-effector scheme with dual RCMs. A mechanism with PRR-type is determined which working space of 200*200*100 mm3 is verified through forward and inverse kinematic analysis. A Π-shaped structure is designed for rigidly connecting two RCM units which realizes compact assembly of end-effector. The Lagrange equation-based dynamic model of the mechanism is established, and the dynamic simulation demonstrated that the mechanism needs no torque in probe’s vertical position and larger torques are required when the probe moves towards the two ends. The proposed end-effector can be installed at the end of the robotic arm and provides a reference scheme for dexterously local operation with robot navigation in orthopedic ESWT.
|
|
SaS2A |
Room D |
Award Session 2 |
Regular Sessions |
Chair: Tao, Yong | Beijing University of Aeronautics and Astronautics |
|
11:30-11:42, Paper SaS2A.1 | |
Acoustic 3D Shape Reconstruction of Zygomatic Bone for Non-Invasive Spatial Registration |
|
Han, Runzhe | Beihang University |
Yuan, Mengning | Peking University School and Hospital of Stomatology |
Jie, Bimeng | Peking University |
He, Yang | Peking University School and Hospital of Stomatology |
Zhu, Jiayu | Beihang University |
Zhang, Runshi | Beihang University |
Wang, Junchen | Beihang University |
Keywords: Deep Learning, Machine Learning, Artificial Intelligence
Abstract: Image-to-patient registration is an essential step in the overall image-guided surgery and determines the accuracy of precise surgery. Currently used registration methods, such as implanting titanium nails in the patient's face, using occlusal splints or surface registration techniques based on soft tissues, etc., have their own problems such as invasiveness, high cost, or poor stability. Ultrasound imaging has the advantages of non-invasiveness and low cost, and facial bones are usually considered to have minimal deformation. Therefore, we hope to use ultrasound to non-invasively obtain facial bone surface features to complete the preoperative image-patient registration process. Therefore, we propose a technology for cranio-maxillofacial surgery that segments and reconstructs 3D images from 2D ultrasound images obtained by ultrasound scanning of the zygomatic bone surface. In the segmentation module, we use nnU-Net with improved loss function and learning rate strategy to segment the zygomatic bone, achieving a Dice score of 0.8405. In the 3D reconstruction module, we propose a PNN algorithm with image downsampling that can quickly process data and a DSBezier algorithm that adds Bezier curves to fill voxels. The reconstruction results are registered with the head CBCT data, and the average residual can reach within 0.6 mm. The experimental results show that this system can accurately segment and reconstruct the zygomatic bone in three dimensions from ultrasound images.
|
|
11:42-11:54, Paper SaS2A.2 | |
Application of Compliant Control in Pin-Based Shape Display for Compliant Physical Interaction between Human and Machine |
|
Huang, Yizhou | Zhejiang University |
Xiao, Songjie | Zhejiang University |
Shi, Yunze | Zhejiang University |
Li, Haoyu | Zhejiang University |
Yang, Liangjing | Zhejiang University |
Keywords: Human-Robot Interaction and Cooperation, Sensing, Haptic System, Robot Design
Abstract: Compliant control is widely used in the human-robot interaction (HRI) field. However, very few research pays attention on applying compliant control to the pin-based shaped display which is a type of HRI device widely researched as tangible user interface for the enhancement of the HRI quality. Admittance control as one of the compliant controls is performing well under soft environment which is ideal as a candidate control method for pin-based shape display for the reason that the human body surface is generally soft. In this paper, a pin-based shape display with admittance control will be prototyped. The general control scheme as well as the design methodology will be explained and evaluated. A variable compliant control will be proposed and simulated to achieve even normal force distribution on its surface. Different compliant control parameters will be chosen to evaluate the device under experiment.
|
|
11:54-12:06, Paper SaS2A.3 | |
Integrated Task Scheduling, Action Planning and Control for Autonomous Disinfection by a Mobile Robotic Manipulator |
|
Ye, Jiajie | The University of Hong Kong |
Sheng, Yongji | The University of Hong Kong |
Wang, Siyu | The University of Hong Kong |
Ma, Ye | The University of Hong Kong |
Kwok, Chun | The University of Hong Kong |
Wang, Qingyang | The University of Hong Kong |
Xi, Ning | The University of Hong Kong |
Keywords: Intelligent Control and Systems, Mobile Robotics
Abstract: This paper presents an approach for solving a challenging problem in the integration of low-level system sensing and control with high-level task scheduling, planning and perception in mobile robotic manipulator systems for autonomous disinfection. First, a unified event-based reference will be introduced. Then, the approach will be implemented in the task of disinfection by a mobile manipulator. The task of disinfection involves both continuous and discrete events. The interactions between discrete and continuous actions will also be considered. The disinfection task in a multiple-object venue is used to illustrate the proposed approach. The experimental results demonstrate the advantages of this method.
|
|
12:06-12:18, Paper SaS2A.4 | |
Dynamic Modeling Method of Six-Axis Force/Torque Sensors Based on IGWO-GRU (I) |
|
Xue, Yufei | Anhui University |
Gao, Lifu | Hefei Institutes of Physical Science, Chinese Academy of Science |
Cao, Huibin | Institute of Intelligent Machines, Chinese Academy of Sciences |
Sun, Yuxiang | Institute of Intelligent Machines, Chinese Academy of Sciences |
Keywords: Sensing, Haptic System, Deep Learning
Abstract: The strain-based six-axis force/torque sensor often faces challenges in accurately measuring complex and refined dynamic phenomena, leading to significant dynamic errors. To overcome this issue, it is essential to establish a high-precision dynamic model. A dynamic modeling method for six-axis force/torque sensors based on deep learning is presented in this paper. The proposed method utilizes the gated recurrent unit (GRU) neural network to construct an accurate dynamic model. The GRU neural network is capable of capturing long-term dependencies and handling sequential data effectively, making it suitable for modeling the interdimensional coupling between the dynamic force/torque and the sensor's dynamic response. To address the subjective judgment involved in selecting hyperparameters for deep learning, the improved gray wolf optimization (IGWO) algorithm is employed. In this study, the proposed IGWO-GRU method is applied to dynamically model six-axis force/torque sensors under negative step excitation and impact excitation. The evaluation of the method yields an average mean squared error (MSE) of 0.056417 and 0.05590 for the six channels in the respective cases. Through comparison with other algorithms, the results demonstrate the accurate modeling effect and the generalizability of the proposed method. Moreover, the findings provide new insights for the dynamic modeling of other types of sensors.
|
|
SaS2B |
Room E |
Control Methods and Mechatronic Designs towards Higher Level Autonomous
Medical Robots 2 |
Invited Sessions |
Chair: Su, Baiquan | Beijing University of Posts and Telecommunications |
Co-Chair: Wang, Junchen | Beihang University |
Organizer: Su, Baiquan | Beijing University of Posts and Telecommunications |
|
11:35-11:45, Paper SaS2B.1 | |
A 2-DOFs Planar Grinding End-Effector for Robotic Laminectomy |
|
Duan, Wei | Beihang University |
Ji, Xuquan | Beihang University |
Hu, Lei | Beihang University |
Liu, Wenyong | Beihang University |
Keywords: Medical Robotics, Path and Motion Planning, Emerging Technologies and Applications
Abstract: Robotic end effector accurately operated in small and confined surgical area in laminectomy can improve the efficiency of robotic surgery. This paper proposed a two-DOFs planar grinding end-effector for robotic laminectomy, which can move along any curve within the planar operation area. The end-effector adopts an arc RCM mechanism, which kinematic analysis gives a working space of an annular sector with a minimum radius of 137 mm, a maximum radius of 162.3 mm, a central angle of 1.65 rad. The structure rationality of the end-effector is verified through static analysis. The prototype was developed and controlled based on TwinCAT-PC architecture. Experiments of positioning accuracy give a repeated positioning accuracy of 0.1 mm. Experiments of ex vivo bone grinding gives an operation accuracy of 1.5 mm in a simulated clinical environment. The proposed prototype can be used for robotic laminectomy in combination with the navigation robotic arm.
|
|
11:45-11:55, Paper SaS2B.2 | |
A Disturbance Compensation Method Using Adaptive Neural Network for Robotic Manipulator |
|
Yang, Siqin | Beihang University |
Lu, Chunheng | Beihang University |
Luo, Xuejin | Beihang University |
Wang, Junchen | Beihang University |
Keywords: Intelligent Control, Intelligent Control and Systems, Medical Robotics
Abstract: In contact with the hardest human tissues, such as bones and teeth, performing force control precisely is the key for improving operation effect and ensuring surgical safety. This article introduces an intuitive six degree-of-freedoms (6-DoF) task space PD control to track a given trajectory based on a linear and decoupled model for six dimensional pose (rotational and positional) displacement. Then, an adaptive neural network (NN) controller is designed to deal with the nonlinearities of the system. A learning method based on the radial basis function NN (RBFNN) is involved in controller design to compensate for the manipulator's dynamic uncertainties. The stability of the controller is proved by using Lyapunov stability principles. Finally, the effectiveness of the proposed methods are validated through a group of setpoint control and trajectory tracking control simulations on a redundant robotic manipulator. The mean error of setpoint control with robotic uncertainties after compensation was 4.4704 times 10^{-9}, 1.4203 times 10^{-8}, -1.2235 times 10^{-7}, in axis X, Y, Z respectively. The maximum error of trajectory tracking in task space was less than 1 mm.
|
|
11:55-12:05, Paper SaS2B.3 | |
Instability Detection During Physical Human Robot Interaction Based on Wavelet Decomposition |
|
Lu, Chunheng | Beihang University |
Chen, Chenghao | Beijing Children's Hospital, Capital Medical Universtity |
Song, Zhihong | Beihang University |
Wang, Junchen | Beihang University |
Keywords: Human-Robot Interaction and Cooperation, Intelligent Control and Systems, Human-Machine Interface
Abstract: Physical human robot interaction is an important research topic in robotics. Admittance control has been widely used for collaborative tasks because of its compatibility with low-cost robots. However, stability issue arises especially when the interacting environment (including humans) becomes stiff. A common way to guarantee stability is changing control parameters, especially the damping parameters, but it will cause heavy react during interaction. Therefore an effective way to detect the instability and change parameters only when oscillation occurs is necessary. In this paper, we analyzed the frequency of normal operation and oscillation during interaction process, and proposed a real time instability detection method based on wavelet decomposition. Experiments on human interaction with an admittance-controlled robot were performed and the effectiveness of the proposed instability detection method was confirmed.
|
|
12:05-12:15, Paper SaS2B.4 | |
Root Canal Therapy Assisted Robot System |
|
Deng, Yingyan | Beihang University |
Piao, Yang | Peking University School and Hospital of Stomatology |
Lu, Chunheng | Beihang University |
Bao, Xudong | Peking University School and Hospital of Stomatology |
Dong, Yanmei | Peking University School and Hospital of Stomatology |
Wang, Junchen | Beihang University |
Keywords: Medical Robotics, Human-Robot Interaction and Cooperation, Grasping and Manipulation
Abstract: A robot aided system for root canal therapy was developed, which realized image registration, intra-operative visual navigation, human robot cooperative compliance control and robot drilling safety constraints. First, image registration includes two parts: using the Iterative Closest Point (ICP) to register the point cloud of Cone-Beam Computed Tomography (CBCT) image and oral scanning image, and using the method based on feature points to register the oral scanning model and visual space. Then, the mapping between the dental drill frame and the robot flange frame was obtained through robot hand eye calibration, and the image space was connected to the visual space using a binocular vision system, thus realizing visual navigation. Finally, the admittance control was used to guide the robot to locate the root canal orifice, and a linear virtual fixture was applied to constrain the drilling direction. The results of surgical navigation experiment and model drilling experiment show that the image registration error is less than 1mm, the robot drilling position error is less than 1 mm, and the angle error is about 5 degrees, which meets the clinical accuracy requirements.
|
|
SaS3A |
Room D |
AI-Enhanced Humanoid and Rehabilitation Robotics |
Invited Sessions |
Chair: Chen, Fei | The Chinese University of Hong Kong |
Co-Chair: Li, Qiang | Bielefeld University |
Organizer: Wang, Cong | Delft University of Technology (TU Delft) |
Organizer: Chen, Fei | Chinese University of Hong Kong |
Organizer: Chen, Jianyu | Tsinghua University |
Organizer: Li, Gaofeng | Zhejiang University |
Organizer: Li, Qiang | Bielefeld University |
Organizer: Xie, Ming | Nanyang Technological University |
Organizer: Wang, Xin | Chinese University of Hong Kong |
Organizer: Ren, Qinyuan | Chinese University of Hong Kong |
|
14:00-14:15, Paper SaS3A.1 | |
Dexterous and Human-Like Mobile Manipulation: From Autonomous to Collaborative (I) |
|
Chen, Fei | The Chinese University of Hong Kong |
Keywords: Human-Robot Interaction and Cooperation
Abstract: Modern industry demands the heavy usage of robotic mobile manipulators with high autonomy and intelligence to work ether independently or collaboratively with human workers. For this purpose, the robot should be able to accomplish various manipulation tasks without prior knowledge of the status of objects and human workers in a highly unstructured and dynamic environment. Therefore, it is important for the mobile manipulation robots to achieve human level capability in terms of perception of environment, planning and learning of manipulation skills dexterously, adaptively and intelligently. In this talk, we will cover various topics and issues along the direction to develop collaborative human-like mobile manipulation robots the team has been achieved in the past years and demonstrate various setup of autonomous mobile manipulator for various challenging scenarios.
|
|
14:15-14:25, Paper SaS3A.2 | |
Design and Development of the Dual-Arm Robot DARO (I) |
|
Liu, Yun | Zhejiang Lab |
Huang, Qiulan | ZHEJIANG LAB |
Zhang, Pu | Zhejianglab |
Yan, Mindong | Zhejiang Lab |
Liang, Dingkun | Research Institute of Interdisciplinary Innovation, Zhejiang Lab |
Kong, Lingyu | Zhejiang Lab |
Wang, Xin | Zhejiang Lab |
Xie, Anhuan | Zhejiang University |
Gu, Jason | Dalhousie University |
Zhu, Shiqiang | Zhejiang Lab |
Keywords: Humanoid Robots, Biologically Inspired Robotics, Robot Design
Abstract: This paper presents the design and development of a humanoid dual-arm robot called DARO, along with its improved version, DARO-N. DARO and DARO-N consist of two arms and humanoid hands, each with 6/7 degrees of freedom for the arms and 7 active degrees of freedom for the hands. Modular commercial actuators are employed as the driving joints of the arms, allowing for reduced development time and cost. The design principles of modularity, simplicity, lightweight construction, human likeness, information perception capability, and scalability are incorporated into the design. Furthermore, this work establishes a control framework for the humanoid dual-arm robot and conducts simulations and multiple physical experiments to validate the performance of the robots.
|
|
14:25-14:35, Paper SaS3A.3 | |
Bio-Inspired Control of a Rehabilitation Robot Actuated by Pneumatic Artificial Muscles (I) |
|
Cheng, Maotong | Zhejiang University |
Zhu, Wenxin | Zhejiang University |
Ren, Qinyuan | Zhejiang University |
Keywords: Rehabilitation and Assistive Robotics
Abstract: Wearable rehabilitation robots can bring great help to the elderly and sport injury patients. However, most of the current traditional rehabilitation robots adopt rigid actuators, which limits their compliance and human-robot interaction performance. In this paper, a novel upper limb rehabilitation robot actuated by pneumatic artificial muscles is developed and its dynamic model is constructed for control issue. In order to achieve more accurate and adaptive rehabilitation robot control, this paper proposes a bio-inspired control architecture and cerebellum-like controller network. To demonstrate the control effect of the proposed theoretical method, a series of simulations are conducted on the robot, and the results show that the proposed controller has good tracking performance and adaptation.
|
|
SaS3B |
Room E |
Control Methods and Mechatronic Designs towards Higher Level Autonomous
Medical Robots 3 |
Invited Sessions |
Chair: Su, Baiquan | Beijing University of Posts and Telecommunications |
Co-Chair: Wang, Junchen | Beihang University |
Organizer: Su, Baiquan | Beijing University of Posts and Telecommunications |
|
14:00-14:10, Paper SaS3B.1 | |
Cooperative Manipulator Control Based on IgH Ethercat Master |
|
Luo, Xuejin | Beihang University |
Yang, Siqin | Beihang University |
Xu, Hao | Beihang University, Beijing |
Lu, Chunheng | Beihang University |
Wang, Junchen | Beihang University |
Keywords: Intelligent Control and Systems, Human-Robot Interaction and Cooperation
Abstract: Cooperative manipulators are widely used in the scenarios where the human and the robot work closely to perform tasks such as assembling, medical care and surgical operation. More advanced control laws such as model-based force regulation with space constraint are increasingly needed to endow the robot with safe interaction capability in medical scenarios. These advanced control laws require complete control of the robot manipulator with minimal control latency. Unfortunately, commercial manipulators do not provide users with low-level access to the controller. Therefore, we developed a six-axis robotic manipulator for cooperative tasks with human. A control system based on EtherCAT with high real-time performance and satisfied motion control effect is proposed in this paper. The open source IgH EtherCAT is employed as the master and each drive of the axis motors serves as the slave node. A Linux + Xenomai dual-kernel system is configured to ensure the real-time performance. A motion control system based on IgH EtherCAT is built. On this basis, the free drag control in task space based on admittance control is achieved. The experimental results show that the communication time was less than num{4e4} nanosecond (ns), and the cycle jitter was below 20{mu}s. The maximum average joint velocity error during the task space free drag was 6.85 times 10^{-4} rad/s.
|
|
14:10-14:20, Paper SaS3B.2 | |
Preliminary Design of Robotic System for Closing Incision by Rotating Bandage Method |
|
Chen, Anqi | Beijing University of Posts and Telecommunications |
Ma, Xudong | Beijing University of Posts and Telecommunications |
Sun, Yixin | Beijing University of Posts and Telecommunications |
Liu, Wenyong | Beihang University |
Wang, Junchen | Beihang University |
Li, Changsheng | Beijing Institute of Technology |
Hu, Yida | Harvard Medical School |
Su, Baiquan | Beijing University of Posts and Telecommunications |
Keywords: Medical Robotics, Robot Design
Abstract: Wound treatment is routine work in hospital and bandaging of wound happens numerous even in a hospital scale. However, manual bandaging is accompanied with potential infection and it is tedious and manpower-wasting besides. This paper proposed a customized bandaging mechatronic system for avoiding bacteria from hand contact. The design of the propose bandaging system is developed and the working procedure of the by the system is given step by step. To the best knowledge of the authors, the customized mechatronic system is the first bandaging robotic system for bandaging purpose. After loading a customize bandage package, the proposed system lays a solid foundation for automatous bandaging method without human intervention during bandaging.
|
|
14:20-14:30, Paper SaS3B.3 | |
Teleoperation of Multi-Functional Neurosurgical Tool |
|
Ma, Xudong | Beijing University of Posts and Telecommunications |
Yao, Lechen | The Affiliated High School of Peking University |
Chen, Anqi | Beijing University of Posts and Telecommunications |
Wang, Junchen | Beihang University |
Li, Changsheng | Beijing Institute of Technology |
Liu, Wenyong | Beihang University |
Hu, Yida | Harvard Medical School |
Su, Baiquan | Beijing University of Posts and Telecommunications |
Keywords: Medical Robotics, Tele-Robotics/Networked, Cloud Robotics, Robot Design
Abstract: Fluid removal and electrotome coagulation are routine procedures in neurosurgery operations, and the doctor's remote control can reduce physical fatigue. In this paper, an integrated multi-functional surgical tool which can realize electrocoagulation and liquid aspiration is designed. At the same time, the multi-functional integrated surgical device is combined with the remote operation method to realize the remote operation of the instrument. This multifunctional instrument includes CCD-imaging, suction and electrical coagulation. The system can complete the remote operation function of the special design tool, and can isolate the doctor from the operation. The teleoperation of this instrument is integrated with the teleoperation of other functional instruments to realize the teleoperation of the complete surgical process.
|
|
14:30-14:40, Paper SaS3B.4 | |
Gesture Control Method for Oral Blood Mixed Liquid Removal Robot |
|
Ma, Xudong | Beijing University of Posts and Telecommunications |
Kuang, Ziao | School of Computer Science (National Pilot Software Engineering |
Li, Han | Beijing University of Posts and Telecommunications |
Chen, Anqi | Beijing University of Posts and Telecommunications |
Li, Changsheng | Beijing Institute of Technology |
Liu, Wenyong | Beihang University |
Wang, Junchen | Beihang University |
Hu, Yida | Harvard Medical School |
Gao, Li | Peking University School and Hospital of Stomatology |
Su, Baiquan | Beijing University of Posts and Telecommunications |
Keywords: Medical Robotics, Human-Machine Interface
Abstract: All of oral surgeries are accompanied by bleeding. Manual removal of blood is carried out by a nurse and dozens of repetitive blood suction operations are implemented by the nurse during a single oral surgery. In our previous work, a robot for removal of blood-water mixed liquid is designed. However, the robot requires an initial order to autonomous removal of blood mixed liquid. In this paper, a hand gesture control method for typical operations of the robot is developed. Gesture dataset are developed at first and a deep learning neural network is trained for detection of hand gesture. The gesture control procedures for four typical operations are provided in detail. In a group of experiments, it’s demonstrated that an oral surgeon controls the robot’s motion by presenting prescribed hand gesture. The proposed method could be an alternative manner for blood mixed liquid removal.
|
|
SaS4A |
Room D |
Revolutionizing Interaction: Embodied Intelligence and the New Era of
Human-Robot Collaboration |
Invited Sessions |
Chair: Qi, Wen | Politecnico Di Milano |
Co-Chair: Su, Hang | Politecnico Di Milano |
Organizer: Qi, Wen | Politecnico Di Milano |
Organizer: Ntalampiras, Stavros | University of Milan |
Organizer: Yue, Zhengjun | Technische Universiteit Delft |
Organizer: Ovur, Salih Ertug | Imperial College London |
Organizer: Su, Hang | Politecnico Di Milano |
Organizer: Chen, Yue | Georgia Institute of Technology |
Organizer: Aliverti, Andrea | Department of Electronics, Information and Bioengineering, Polit |
|
14:40-14:50, Paper SaS4A.1 | |
An Edge-Fog-Cloud-Based Hierarchical Adaptive Model for Human-Robot Interaction (I) |
|
Fan, Haoyu | South China University of Technology |
Li, Zhijie | South China University of Technology |
Su, Hang | Politecnico Di Milano |
Qi, Wen | Politecnico Di Milano |
Keywords: Human-Robot Interaction and Cooperation, Artificial Intelligence, Deep Learning
Abstract: With the rapid advancement of intelligent devices and the Internet of Things (IoT), a vast amount of detailed information can now be captured and analyzed from the human body. This progress has led to the development of more comprehensive and versatile Body Area Networks (BANs). Despite this promising landscape, the field lacks mature global applications, and numerous challenges await resolution, particularly in Human-Robot Interaction (HRI).To address the limitations of HRI, this paper focuses on recognizing complex human activities and achieving rapid, stable transmission of high-load data. An Edge-Fog-Cloud-based hierarchical adaptive system is proposed, utilizing multiple IMU sensors for human activity recognition (HAR). The system introduces an adaptive strategy to enhance the accuracy of identifying complex human behavior. By incorporating hierarchical communication and classification methods, the system improves the communication network's real-time performance and the algorithms' architectural structure, ensuring real-time capabilities for the HRI system.
|
|
14:50-15:00, Paper SaS4A.2 | |
A Review of Mobile Robots in Extreme Environments (I) |
|
Li, Yuhan | South China University of Technology |
Wang, Shengzhi | South China University of Technology |
Du, Juan | South China University of Technology |
Qi, Wen | Politecnico Di Milano |
Keywords: Mobile Robotics, Artificial Intelligence, Deep Learning
Abstract: This paper surveys mobile robotics in extreme environments, focusing on four application areas: the oil and gas industry, underwater exploration, space exploration, and search and rescue operations. Various methodologies and technologies are used to overcome these challenges, including terrain-aided navigation, sensor fusion, and robust SLAM algorithms. This paper comprehensively examines the challenges specific to each environment, such as low visibility, lack of GPS, communication limitations, and potential robot damage.
|
|
15:00-15:10, Paper SaS4A.3 | |
Integrating Sensor Fusion for Teleoperation Control of Anthropomorphic Dual-Arm Robots (I) |
|
Wang, Feilong | Changchun University of Science and Technology |
Chen, Furong | Changchun University of Science and Technology |
Yang, Xiaolong | Changchun University of Science and Technology |
Yong, Qi | ESIEE Paris |
Dong, Yanling | Shandong University |
Zhang, Xinming | Changchun University of Science and Technology |
Zheng, Long | Weihai Institute for Bionics, Jilin University |
Su, Hang | Politecnico Di Milano |
Keywords: Human-Robot Interaction and Cooperation, Humanoid Robots
Abstract: In order to achieve accurate teleoperation control of anthropomorphic two-arm robots, we adopt the sensor fusion method, which fuses data from multiple sensors (such as cameras and data gloves). By processing the information from these sensors, we are able to control the motion of the anthropomorphic dual-arm robot and the bionic hand in real-time. In this paper, a sensor fusion-based teleoperation control system for anthropomorphic dual-arm robots is proposed, which utilizes an extended Kalman filter to fuse data from multiple sensors. By analyzing and processing the position and posture information of the human arm obtained from the sensors and performing the corresponding kinematic mapping, we are able to control the movements of the anthropomorphic two-arm robot and the bionic hand in real-time. Eventually, users will be able to control humanoid two-arm robots through stable and reliable remote operation. In order to verify the effectiveness of the system for different users, we invited several operators to participate in the experiment. The experimental results show that the system can not only achieve a better human-computer interaction experience but also be effective for different users.
|
|
15:10-15:20, Paper SaS4A.4 | |
Sensor Fusion Based Humanoid Robotic Arm Control (I) |
|
Chen, Furong | Changchun University of Science and Technology |
Wang, Feilong | Changchun University of Science and Technology |
Yang, Xiaolong | Changchun University of Science and Technology |
Yong, Qi | ESIEE Paris |
Dong, Yanling | Shandong University |
Gao, Yi | Changchun University of Science and Technology |
Zheng, Long | Weihai Institute for Bionics, Jilin University |
Su, Hang | Politecnico Di Milano |
Keywords: Human-Robot Interaction and Cooperation, Humanoid Robots
Abstract: This paper aims to study the teleoperation method based on multi-sensor fusion to achieve precise control of the robotic arm. The paper proposes a remote operation method based on multi-sensor fusion, which combines Kinect sensor, goniometer and data glove to realize the precise measurement and control of the upper limb joint angle. The adaptive Kalman filter algorithm is introduced to automatically adjust the fusion weight according to the degree of occlusion of the shoulder, elbow and wrist positions, thereby improving the accuracy and stability of joint angle measurement. The effectiveness and feasibility of the proposed method are verified by experiments, which proves the high control accuracy and stability in the remote operation of humanoid robotic arm.
|
|
SaS4B |
Room E |
Intelligent Motion Control of Unmanned Underwater Vehicles |
Invited Sessions |
Chair: Ji, Daxiong | Zhejiang University |
Organizer: Ji, Daxiong | Zhejiang University |
|
14:45-14:55, Paper SaS4B.1 | |
Adaptive Path Following Control Based on Nonlinear Disturbance Observer for a Six Thrusters AUV (I) |
|
Ji, Daxiong | Zhejiang University |
Xu, Minghui | Zhejiang University |
Wang, Xinwei | Zhejiang University |
Ye, Zhangying | Zhejiang University |
Keywords: Path and Motion Planning
Abstract: In this paper, we investigate the path following problem of a six-thruster autonomous underwater vehicle (SAUV) moving on the horizontal plane. Neural network adaptive PID control method based on NDO (BNNPID) is proposed. External disturbance and model uncertainty are taken into account in controller design. A nonlinear disturbance observer (NDO) is designed for disturbances observation. Theoretical derivation and simulation experiments have demonstrated the stability of this observer. We combine NDO and PID controller and use BP neural networks to adjust controller parameters online. In order to improve the correlation between controller parameters and system state, adaptive control is introduced. The robustness and effectiveness of the proposed controller are tested and validated through simulation results.
|
|
14:55-15:05, Paper SaS4B.2 | |
Adaptive Formation Structure Control for USVs with Security Protection Mechanism (I) |
|
Yin, Shilin | Dalian Maritime University |
Zhang, Guoqing | Dalian Maritime University |
Huang, Chenfeng | Dalian Maritime University |
Keywords: Multi-Robot Systems, Path and Motion Planning
Abstract: Underactuated surface vehicles (USVs) formation control tasks are varied and complicated under the background of USVs formation practice. It is impractical for USVs formation to maintaining fixed structure in engineering, e.g. cooperative transportation, multiple-USV game, and formation collision avoidance. For this purpose, a formation control scheme with formation structure transform and security protection mechanism is investigated. In this scheme, an adaptive potential ship is developed in the guidance part. In order to achieve formation structure transform task, while avoiding inter-vehicle collision and external obstacles. For the control strategy, the radial basis function neural networks (RBF-NNs) are employed to approximate the uncertainty items in the model. Meanwhile, a NNs-based disturbance observer (DOB) is designed to reduce the control inputs, by estimating marine disturbance in practice engineering. Based on the minimal learning parameter (MLP) technique, the weights compression of DOB can effectively simplify the observer’s diagram structure. Benefiting from this design, the observer only require known parameters to estimate marine disturbance. Considerable effort has been made to obtain the semi-globally uniform ultimate bounded (SGUUB) stability by the theoretical analysis. Finally, the effectiveness of the proposed algorithm is verified by the numerical simulation.
|
|
15:05-15:15, Paper SaS4B.3 | |
Path Planning Algorithm for Long-Range Autonomous Underwater Vehicles Based on Environmental Features (I) |
|
Ji, Daxiong | Zhejiang University |
Cheng, Huifang | Zhejiang University |
Keywords: Path and Motion Planning, Mobile Robotics
Abstract: To solve the time optimal path planning problem of long-range autonomous underwater vehicle (AUV) in the spatially non-uniform ocean current environment, a path planning algorithm based on environmental features is studied. Firstly, a method for analyzing the distribution of ocean current velocity is proposed to pre analyze the ocean current environment. According to the analysis results, an environmental features based particle swarm optimization algorithm (EFPSO) is proposed for path planning. This algorithm consists of two layers. The first layer is a reference path search algorithm based on environmental features, which is used to identify and quickly locate the region where the global optimal path is located based on environmental features. The second layer is a path search algorithm for key region, which is used to further optimize the path in the region where the reference path is located. Simulation experiments show that EFPSO can avoid falling into local optima and stably obtain low-cost paths while increasing a small amount of computation time.
|
|
SaR1A |
Room D |
Small-Scale Robots |
Regular Sessions |
Chair: Wei, Junyi | University of Essex |
Co-Chair: Zhang, Tieshan | City University of Hong Kong |
|
15:25-15:35, Paper SaR1A.1 | |
The Control System for 5-DOF Magnetic Levitation Capsule Robot (I) |
|
Zeng, Zijin | Beihang University |
Wang, Fengwu | Beihang University |
Zhao, Jiawei | Beihang University, School of Mechanical Engineering and Automati |
Wang, Chutian | Beihang University |
Li, Chan | Beihang University |
Feng, Lin | Beihang University |
Keywords: Medical Robotics
Abstract: For the successful and efficient operation of capsule endoscope robots in the human stomach, high-accuracy and multi-degree of freedom (MDOF) motion control of capsule robots is crucial. This paper constructs a human-sized magnetic control system and proposes a 5-DOF motion control method that utilizes electromagnetic coils and a localization approach based on a simulated annealing-particle swarm optimization algorithm, aiming to enhance the accuracy of deflection control. The results indicate that the proposed methods achieve excellent levitation stability and improved translation and orientation control accuracy, highlighting the potential benefits of employing this method in the control system for capsule robots.
|
|
15:35-15:45, Paper SaR1A.2 | |
Equilibrium Theory Based Power Sharing Allocation Algorithm for Wireless Capsule Endoscopy Relays |
|
Li, Ye | BUAA Universiy |
Zhao, Jiawei | Beihang University, School of Mechanical Engineering and Automati |
Wei, Junyi | University of Essex |
Feng, Lin | Beihang University |
Keywords: Medical Robotics, Sensor Networks
Abstract: Wireless capsule endoscopy (WCE) is an effective tool for gastrointestinal diseases diagnosis. Compare to the traditional endoscopy, more and more patients and medical institutions prefer WCE due to its user-friendly and non-cross infection. However, the size limitation of WCE would lead to a problem of insufficient energy. This paper will propose an relay-based power allocation algorithm for WCE transmission system based on market equilibrium scheme. Considering power resources of a WCE as assigned resources and the WCE transmission networks revenue is defined by a logarithm utility function. Either relay and direct transmission links aim to maximize its own profit by power resources sharing. In order to solve the multi-objective optimization problem, the optimization problem can be divided into two single-objective problems: firstly, maximizing the profit of the direct link; secondly, maximizing the profit of the relay-based links. The multi-objective optimisation problem can be solved by calculating the equilibrium point of the power utility function. The convergence of the algorithm can be verified by numerical results.
|
|
15:45-15:55, Paper SaR1A.3 | |
High Precise Metallic Helical Microstructure Fabrication by Rotational Nanorobotic Manipulation System with Tilted Mandrel Compensation |
|
Zhang, Tieshan | City University of Hong Kong |
Ren, Hao | City University of Hongkong |
Li, Gen | City University of Hong Kong |
Guo, Dong | City University of Hong Kong |
Shang, Wanfeng | Shenzhen Institutes of Advanced Technology, Chinese Academy of S |
Shen, Yajing | The Hong Kong University of Science and Technology |
Keywords: Micro/Nano Robotics
Abstract: Metallic helical microstructures are significant components applied in many fields, such as mechatronics, photonics, microelectromechanical systems (MEMS) and biomedical engineering, but their fabrication with high precision remain challenge. Herein, we report a general controllable approach for high-precise micro metallic helical structure fabrication by a self-developed rotational nanorobotic manipulation system. To reduce the fabrication error induced from the tilted mandrel, firstly, we propose an end-effector trajectory planning method to pre-compensate the tilted angle during rotation. Further, a hybrid vision and position feedback strategy is also developed to automate the fabrication process for obtaining a helical structure with uniform pitch distance consequently. The experimental results verify that our method is able to fabricate various micro helical structures with different dimensions with high precision up to 6 um (less than 2 pixel equivalent). This research paves an effective solution for the high precise 3D helical structure fabrication, which would prompt their applications in engineering and inspire other microstructure manufacturing as well.
|
|
SaR1B |
Room E |
Soft Robots |
Regular Sessions |
Chair: Yang, Yongliang | Shenyang Institute of Automation, CAS |
Co-Chair: Zhang, Yi | University of Electronic Science and Technology of China |
|
15:20-15:35, Paper SaR1B.1 | |
RoboFold: A Platform for Protein-Folding Inspired Robot Swarms (I) |
|
Zhang, Feng | Shenyang Jianzhu University |
Gu, Yuping | Shenyang Jianzhu University |
Yuan, Shuai | Shenyang Jianzhu University |
Yang, Yongliang | Shenyang Institute of Automation, CAS |
Keywords: Soft Robotics, Biologically Inspired Robotics, Multi-Robot Systems
Abstract: Natural swarms have inspired various controlling algorithms for swarm robotics, while few of them were programmed in a similar way as the natural swarms. Defining a reliable programming method is still a daunting challenge in swarm robotics. In nature, the information of 3D protein structure is programmed in the sequence of amino acids of the protein. To mimic the programming method of protein folding, we present a platform for investigating protein-folding inspired programming method for robot swarms. This platform includes three parts: a vibration machine, protein-like robot sequences, and recording devices. The preliminary results presented here demonstrated the effectiveness of this platform and the feasibility of protein-folding inspired programming method for robot swarms.
|
|
15:35-15:45, Paper SaR1B.2 | |
A Bionic Mimosa Soft Robot Composed by PNIPAM-PEGDA Copolymer |
|
Wang, Xiaowen | Yantai University |
Yang, Wenguang | Yantai University |
Ge, Zhixing | Shenyang Institute of Automation |
Keywords: Soft Robotics, Smart Structures, Materials, Actuators
Abstract: Deformation is very important for biological survival, more and more researchers are committed to 3D deformation of 2D materials. In this research, we make 3D deformation by applying different internal stresses to 2D materials. At the beginning, single layer PNIPAM hydrogel sheets with different combinations of microchannels were fabricated by UV curing technology. Because of the uneven distribution of internal stresses, they deformed into different 3D structures. In order to maintain these 3D structures though without continuous stimulation, PNIPAM-PEGDA bilayer structure was prepared. It was found that the bilayer structure could respond to both temperature and solvent composition due to the difference of solubility between PNIPAM and PEGDA. Based on the bilayer structure, we made the bionic Mimosa, which could not only realize the open-closing reversibly, but also maintained the normally closed state without continuous stimulation. This research provides a reference for the development of directionally programmable soft materials and has great potential in the fields of bionics and biomedicine.
|
|
15:45-15:55, Paper SaR1B.3 | |
Intelligent 3D-Printed Magnetic Micro Soft Robotic Hand with Visual Feedback for Droplet Manipulation |
|
Zhou, Aiwu | Nanyang Technological University |
Zhang, Yi | University of Electronic Science and Technology of China |
Keywords: Micro/Nano Robotics, Robot Vision and Computer Vision, Intelligent Control
Abstract: Micro robotic hand, which manifests in many forms and is actuated via different mechanisms, is a popular tool for precise manipulation of small-scaled objects. However, conventional micro robots have limited controllability due to their simple structure and lack of sensing elements for status and environment feedback. In this work, a new type of 3D-printed robotic hand with computer vision-based feedback control is presented. The proposed micro soft robotic hands is actuated by a simple magnetic mechanism and shows the ability to manipulate soft matters such as liquid droplets. Identifying objects, especially clear droplets that do not have a strong contrast with the background, presents a great challenge for visual feedback. Therefore, computer vison-based AI technology is implemented to identify and track the micro robotic hand and target objects, which provides a visual feedback for the closed-loop control of desired actions in order to accomplish accurate and precise operations.
|
|
SaR2A |
Room D |
Human-Robot Interaction and Cooperation |
Regular Sessions |
Chair: Wang, Wenxue | Shenyang Institute of Automation, CAS |
Co-Chair: Gao, He | BeiHang University |
|
16:15-16:25, Paper SaR2A.1 | |
Recognition of Dynamic Grasping Motion through Feature Extraction for Teleoperation |
|
Dai, Yanping | Shenyang Institute of Automation, Chinese Academy of Sciences |
Chen, Wenyuan | Shenyang Institute of Automation, Chinese Academy of Sciences |
Xi, Ning | The University of Hong Kong |
Wang, Wenxue | Shenyang Institute of Automation, CAS |
Keywords: Tele-Robotics/Networked, Cloud Robotics, Human-Robot Interaction and Cooperation, Deep Learning
Abstract: Detecting human movement intentions and transmitting human operational capabilities to robots is of great significance for teleoperations. However, the difference in kinematic structure between human hands and robotic manipulators brings difficulties to teleoperation mapping. In this paper, we present an approach to hand grasp recognition and mapping based on feature extraction to drive heterogeneous remote manipulator. The hand grasp data set HUST recorded with a data glove was used for this study. The data were preprocessed and their features were extracted by PCA. A neural network-based classifier was obtained through offline training of human grasping activities. We compared different network models and selected the network with the best performance based on the training time and the task requirements of teleoperation. The experimental results showed that the recognition accuracy of dynamic grasping motion with feature extraction is higher than that without feature extraction and can be up to 97.47%. Finally, we validated the feasibility of this method to tele-operate an Barrett hand through online experiments. Our proposed method provides an effective approach of mapping natural human grasping operations to heterogeneous robotic teleoperation systems.
|
|
16:25-16:35, Paper SaR2A.2 | |
Development of a Simulation Environment for Robot Soccer Game with Deep Reinforcement Learning and Role Assignment |
|
Zhong, Hanzhong | Tsinghua University |
Zhu, Haier | Tsinghua University |
Li, Xiang | Tsinghua University |
Keywords: Deep Learning, Multi-Robot Systems
Abstract: The robot soccer game has been recognized as an excellent scenario to test the gaming algorithm of multi-agent systems. This paper develops a new simulation platform for the robot soccer game, and it has the advantage of open architecture, such that the formation control scheme, the path planning strategy for multiple robots, and many other algorithms can be implemented and tested. Specifically, both a Deep Reinforcement Learning (DRL) scheme and a role-assignment-based method have been successfully realized in this platform to drive multiple robots to play the soccer game, including 2V2, 3V3, 4V4, and so on. It is believed that the developed simulation environment can be used for data collection and transfer learning (TL), hence bridging the gap of Sim2Real technique in actual implementations.
|
|
16:35-16:45, Paper SaR2A.3 | |
The Impact of Commercial Chatbots on Customer Relationship Quality: The Mediating Role of Social Perceptions and Psychological Distance |
|
Tu, Yangjun | Hunan University |
Xiao, Lijun | Hunan University |
Liu, Wei | Hunan University |
Yang, Zhi | Hunan University |
Keywords: Human-Robot Interaction and Cooperation
Abstract: More businesses are choosing to use chatbots instead of traditional human customer support. However, this transition may have significant implications for customer relationship quality. This study aims to comparing the impact of commercial chatbots and human customer service on customer relationship quality and the mediating effects of social perceptions (competence and warmth) and psychological distance. Using a one factor between-subjects experimental design (service agent types: chatbot vs. human), 296 valid data samples were collected on the Credamo platform to analyze the main effects of service agent types on customer relationship quality and the mediating effects of social perceptions and psychological distance. The results of this study indicate that compared to human customer service representatives, chatbots have a more significant positive impact on customer relationship quality, including satisfaction, trust, and commitment. Further analysis reveals that perceived competence and psychological distance mediate the effects of service agent types on relationship quality, while perceived warmth does not exhibit mediating effects. This study expands and enriches the research on service agent types by introducing higher-order and mediating psychological process variables such as social perceptions and psychological distance, revealing the specific mechanisms through which service agent types affect customer relationship quality.
|
|
16:45-16:55, Paper SaR2A.4 | |
The Impact of Chatbot Avatars on Relationship Quality in Commercial Scenarios |
|
Tu, Yangjun | Hunan University |
Xiao, Lijun | Hunan University |
Yang, Zhi | Hunan University |
Keywords: Human-Robot Interaction and Cooperation
Abstract: Chatbots have been widely applied in commercial scenarios as supplements or replacements for traditional human customer service. When interacting with customers, chatbots often present themselves through avatars, playing a significant role in the interaction between businesses and customers. However, there is currently no conclusive evidence on which avatar type is most effective in improving relationship outcomes, especially in the case of conversation breakdown. Therefore, this study aims to explore the impact of different avatar types of chatbots in commercial settings on relationship quality. Using a one factor between-subjects experimental design (avatars: human-like vs.animal vs.company logo), 505 valid data samples were collected on the Mturk platform. The results of this study indicate that the avatar of chatbots has a significant impact on satisfaction and commitment, while its effect on trust is not significant. According to post-hoc comparisons, animal avatars had a significantly greater impact on satisfaction and commitment compared to human-like avatars and company logo avatars. This indicates that the use of animal avatars for chatbots can achieve higher levels of satisfaction and commitment than human-like avatars and company logo avatars. This study makes significant contributions to the fields of human-computer interaction theory, design concepts, and customer relationship theory.Additionally, it provides valuable insights for businesses in deploying chatbots
|
|
16:55-17:05, Paper SaR2A.5 | |
A Safety Motion Planning Algorithm for Mobile Manipulator Based on Improved LSTM Neural Network and Capability Map |
|
Wan, Jiahao | Beihang University |
Tao, Yong | Beijing University of Aeronautics and Astronautics |
Liu, Xiaonan | Beijing New Materials and New Energy Technology Development Prom |
Song, Yian | Beihang University; School of Mechanical Engineering & Automatio |
Gao, He | BeiHang University |
Keywords: Human-Robot Interaction and Cooperation, Path and Motion Planning, Industrial Robotics and Factory Automation
Abstract: Currently, indoor narrow environments, such as assembly in large aircraft cabins, suffer from the inefficiency and compromised quality of manual operations. The existing mobile manipulators for human-machine collaboration still have certain deficiencies, including high redundancy, insufficient constraints, and weak active safety protection capabilities. In light of these challenges, we present a safety motion planning algorithm for mobile manipulator based on improved LSTM neural network and Capability Map (CM). We have modified the robot's capability index (CI) based on the safety potential field function. This provides the spatial representation of robot operation ability for motion planning. In addition, we have added a security gate mechanism to LSTM, enabling more accurate prediction of personnel movements in high-risk areas. The experimental results demonstrate that this algorithm holds great potential in human-machine collaboration. By doing so, the algorithm enables more accurate prediction of the movement trajectory of personnel and facilitates the avoidance of motion conflicts between robots and personnel.
|
|
SaR2B |
Room E |
Artificial Intelligence |
Regular Sessions |
Chair: Xu, Jing | Tsinghua University |
|
16:15-16:25, Paper SaR2B.1 | |
Deep Reinforcement Learning Integrated RRT Algorithm for Path Planning |
|
Liu, Huashan | Donghua University |
Gu, Yuqing | Donghua University |
Li, Xiangjian | Donghua University |
Xiao, Xinjie | Donghua University |
Keywords: Path and Motion Planning, Artificial Intelligence, Industrial Robotics and Factory Automation
Abstract: Rapidly-exploring random tree (RRT) algorithm, featured with strong exploration capability, is widely used in path planning tasks. However, it is difficult for RRT to find the optimal path due to its inherent characteristics of random sampling. In this paper, we propose an improved RRT algorithm integrated with deep reinforcement learning (IRRT-DRL), which can effectively search for feasible paths and exploit previous experience for optimization. It includes a unique reward function and a dynamic waypoint selection mechanism that automatically adjusts the interval between adjacent waypoints to help the agent bypass obstacles. Experimental results have verified the feasibility and superiority of the proposed approach.
|
|
16:25-16:35, Paper SaR2B.2 | |
Expert-Guided Triple-Memory Deep Reinforcement Learning for Robotic Motion Planning |
|
Liu, Huashan | Donghua University |
Zhang, Guojing | Donghua University |
Li, Xiangjian | Donghua University |
Xiao, Xinjie | Donghua University |
Keywords: Artificial Intelligence, Path and Motion Planning, Industrial Robotics and Factory Automation
Abstract: Off-policy deep reinforcement learning (DRL) mostly use memory to improve the processes of data sampling. However, it is difficult for DRL to obtain positive feedback in the early stage of the training. Although an expert memory can solve the problem of sparse rewards in the early stage of training, it can not ensure to obtain the optima, and even leads to overfitting, which affects the performance of the agent in the later stage. In this paper, an expert-guided triple-memory (ETM) framework is proposed to be applied in off-policy DRL, so as to enhance the convergence performance and enhance the stability of the algorithm. Experimental results show that, the proposed approach is feasible and superior in the obstacle-avoidance motion planning tasks for robots.
|
|
16:35-16:45, Paper SaR2B.3 | |
Fast Identification of Stacked Steel Pipes Based on Improved NCC Algorithm |
|
Sun, Chuanzhu | Qingdao University |
Mingjie, Yao | Qingdao University |
Fu, Chaoxing | Qingdao University |
Keywords: Artificial Intelligence, Industrial Robotics and Factory Automation, Robot Vision and Computer Vision
Abstract: In order to solve the problem that the rotation-invariant NCC algorithm has a large number of operations in each pyramid layer in the matching of rotating targets, this paper proposed to use a combination of Hough transform and pyramid layering strategy to accelerate the algorithm. In the coarse matching stage at the top of the pyramid, the Hough transform was introduced to detect the straight line in the search image to calculate the angle of the steel pipe, thus setting the angle constraint and eliminating the selection of the rotation step and the subsequent estimation of the rotation angle in the top of the pyramid, speeding up the matching speed of the algorithm and improving the real-time performance of stacked steel pipe detection. The experimental results showed that this improved template matching algorithm can improve the recognition speed of stacked steel pipes with a high accuracy rate. The study has certain theoretical significance and application value for improving the speed of visual recognition and localisation of stacked steel tubes.
|
|
16:45-16:55, Paper SaR2B.4 | |
Direct Monocular Visual Odometry Based on Lidar Vision Fusion |
|
Fang, Baofu | HeFei University of Technology |
Pan, Qing | Hefei University of Technology |
Wang, Hao | Hefei University of Technology |
Keywords: Robot Vision and Computer Vision, SLAM and Navigation
Abstract: Lidar-assisted visual odometry is a widely used method for pose estimation. However, existing lidar-visual odometry methods suffer from depth association errors, and single-point feature-based methods have insufficient accuracy and are prone to tracking failures, leading to inaccurate pose estimation. In this paper, we proposed a direct monocular visual odometry method based on lidar visual fusion. Firstly, high-gradient pixels of the lidar point cloud projection are extracted, and an initial pose estimation is performed by minimizing photometric errors, avoiding inaccurate feature-depth associations. Then, the point-line features in keyframes are combined and associated with the current frame for pose refinement, and the line features are also matched using the minimization of photometric errors. The evaluations on the KITTI Odometry and Nuscene datasets, and compared to lidar odometry and similar lidar-assisted visual odometry methods, show that our method achieves better pose estimation accuracy and robustness in the majority of scenarios.
|
|
16:55-17:05, Paper SaR2B.5 | |
Bayesian Optimization of Double-SAC Robotic Arm Path Planning Using Recent Prioritized Experience Replay |
|
Yuan, Shuai | Shenyang Jianzhu University |
Wen, Ke | Shenyang Jianzhu University |
Zhang, Feng | Shenyang Jianzhu University |
Xiong, Jianqiang | Shenyang Jianzhu University |
|
|
17:05-17:15, Paper SaR2B.6 | |
Research of Improved TD3 Robotic Arm Path Planning Using Evolutionary Algorithm |
|
Zhang, Feng | Shenyang Jianzhu University |
Xiong, Jianqiang | Shenyang Jianzhu University |
Yuan, Shuai | Shenyang Jianzhu University |
Wen, Ke | Shenyang Jianzhu University |
Keywords: Artificial Intelligence, Path and Motion Planning, Biologically Inspired Robotics
Abstract: Nowadays, application of automated intelligent robot arm devices to improve industrial production efficiency has become a popular research field in the world. The previous off-line path planning method of robotic arm has the inadequacies of low efficiency and slow speed. Although the deep reinforcement learning has accomplished many achievements in the path planning of control manipulator, there are still some problems such as long training time and low planning accuracy. To solve the abovementioned issues, we propose an improved Twin Delayed Deep Deterministic policy gradient (TD3) algorithm (Improved Cross-Entropy Method-TD3: ICEM-TD3) for the path planning of the robotic arm. First, this paper combines evolutionary strategies with TD3 to generate action networks. Then the exploration of TD3 algorithm in the action space is replaced by the exploration in the parameter space. In addition, this paper designs a new reward function to weaken the redundancy of planning and accelerate the convergence speed of training. Finally, the Gazebo simulator is adopted to verify the proposed algorithm, and the results illustrate that the proposed algorithm can greatly improve the accuracy of the path planning of the manipulator using deep reinforcement learning.
|
|
SaR3A |
Room D |
Industrial/Field Robotics and Factory Automation |
Regular Sessions |
Chair: Li, Xiang | Tsinghua University |
|
17:05-17:20, Paper SaR3A.1 | |
Robotic Dexterous Manipulation of Deformable Objects: Theory and Experiments (I) |
|
Li, Xiang | Tsinghua University |
Keywords: Multi-Robot Systems
Abstract: The task involving deformable objects (DOs), such as cables, tissues, and flour, can be found in many applications, such as manufacturing, medical care, and home service. Controlling the robot to autonomously and intelligently manipulate DOs can automate such tasks and hence provide better service for human society. However, compared with the rigid object, it is challenging for the robot to manipulate DOs, mainly because of its deformable nature. As a result, the deformation model is usually unknown and varies significantly; the feature may be occluded and unobservable; the planned path needs to be shaped online according to the unforeseen deformation; the object’s infinite degrees of freedom are difficult to control. This talk discusses the robotic dexterous manipulation method for DOs, and it covers all the aspects of modeling, perception, planning, and control. Real-world experiments and applications of the proposed method are presented.
|
|
17:20-17:30, Paper SaR3A.2 | |
Research on Trajectory Planning and Control of Hydraulic Stacking Robot |
|
Li, Deran | Hefei University of Technology |
Cao, Jian | Hefei University of Technology |
Li, Feiteng | Zhejiang University |
Zhu, Xiaocong | Zhejiang University |
Keywords: Industrial Robotics and Factory Automation, Path and Motion Planning, Robot Design
Abstract: To ensure the smoothness of hydraulic-robot’s motion and the control accuracy of the end effector trajectory for stacking, a fuzzy PID controller has been employed with the trajectory planning in joint space. Wherein, a trajectory planning algorithm combining quintic polynomial with trapezoidal velocity method is proposed to reduce the motion time. This algorithm ensures smooth angular velocity and acceleration, preventing impacts and vibrations during the robot’s operation. The transfer function of the hydraulic robot’s electro-hydraulic proportional control subsystem is established. And a fuzzy PID controller is designed to track the end effector’s motion trajectory. Simulation experiments demonstrate that the robot has fast motion response and small tracking errors of posture in Cartesian space. The feasibility and effectiveness of the optimization solution have been proved through comparing and analyzing the robot’s motion tracking results.
|
|
17:30-17:40, Paper SaR3A.3 | |
Tip Motion Control and Vibration Suppression of Rotary Flexible Manipulator with Piezoelectric Actuators |
|
Song, Shuaili | Zhejiang University |
Gao, Yingying | Zhejiang University |
Zhu, Xiaocong | Zhejiang University |
Cao, Jian | Hefei University of Technology |
Keywords: Industrial Robotics and Factory Automation
Abstract: The flexible manipulator is widely used in robots, aerospace, medical, manufacturing and other fields due to its light weight, fast response, high adaptability, low energy consumption and high load to weight ratio. However, the flexible manipulator is prone to vibration, and it is difficult to achieve high-speed and high-precision motion control. In this paper, the dynamic model of the rotary flexible manipulator with piezoelectric actuators is established based on Lagrange equation. The optimal configuration criterion and the optimal solution via genetic algorithm are designed for the piezoelectric actuators on the rotary flexible manipulator. The flexible manipulator is decomposed into fast and slow subsystems by singular perturbation principle. A two-time scale composite control strategy with an adaptive robust controller in the slow subsystem and a LQR controller in the fast subsystem is proposed. Experiments show that the flexible manipulator can achieve both good trajectory tracking and vibration suppression under the joint drive of motor and piezoelectric actuators.
|
|
17:40-17:50, Paper SaR3A.4 | |
Type Synthesis of 5-DOF 3T2R Redundantly Actuated Parallel Robots for Super-Large Components with Complicated Surfaces |
|
Jiang, Bingshan | Zhejiang Lab |
Kong, Lingyu | Zhejiang Lab |
Xie, Anhuan | Zhejiang University |
Wang, Xin | Zhejiang Lab |
Huang, Guanyu | Zhejiang Lab |
Zhang, Lan | Zhejiang Lab |
Zhao, Pengyu | Zhejiang Lab |
Tian, Xinyu | Beijing Jiaotong University |
Keywords: Robot Design
Abstract: Aerospace represents the development of a national science and technology. It is an important foundation for exploring space and an important guarantee for the construction of aerospace power. There are many large workpieces in the aerospace field. The box insulation layer of large workpieces is an important processing problem. A piece of special equipment with a 1:1 ratio is no longer in line with the development trend of processing the end-surface insulation layer of the large box. This article proposes a new thick processing scheme, which consists of the XYZ shaft long guide rail and the processing unit. The processing unit is a redundantly actuated parallel robot, which is composed of 3T2R (T represents translation, R represents rotation) redundantly actuated parallel mechanism (RAPM). This paper focuses on the type synthesis of five degrees of freedom (5-DOF) redundantly actuated parallel mechanisms (RAPMs). Based on Lie group theory and configuration evolution, three classes of the RAPMs are synthesized by designing the moving platform I and the limbs. This work lays the foundation for solving the processing method of the end surface of the large box.
|
|
17:50-18:00, Paper SaR3A.5 | |
Modeling and Motion Analysis for Flexible Legs of a Bio-Falling Cat Robot |
|
Cao, Jian | Hefei University of Technology |
Xue, Liang | Hefei University of Technology |
Zhu, Xiaocong | Zhejiang University |
Wang, Shanbo | Hefei University of Technology |
Han, Shun | Hefei University of Technology |
Keywords: Biologically Inspired Robotics, Mobile Robotics
Abstract: A bio-falling cat robot is designed to have possible motion abilities of flipping, walking, bounce, landing and buffering in the unstructured environment. The kinematics using standard Denavit-Hartenberg method and the dynamics using Lagrange method are analyzed for the flexible legs driven by pneumatic muscles. The mathematical model of the pneumatic muscle is established and combined with the dynamics of the flexible legs to obtain the relationship between air pressure and joint angles. Then, the three-dimensional model of the robot is imported into Adams and the co-simulation technology of MATLAB and Adams is developed to analyze behavior of the virtual prototype. Moreover, the gait planning and foot trajectory planning with compound cycloid is proposed. The simulation results of the trot gait for the bio-falling cat robot verify its motion capability.
|
|
SaR3B |
Room E |
Intelligent Control |
Regular Sessions |
Chair: Li, Qiang | Bielefeld University |
Co-Chair: Xu, Tianbo | Shenyang Jianzhu University |
|
17:20-17:30, Paper SaR3B.1 | |
Distributed Underwater Jet Propulsion Based Maneuverable Rudderless AUV: Design Fabrication and Testing |
|
Wang, Zhangtao | Beihang University |
Wu, Wenle | Beihang University |
Wan, Yi | Beihang University |
Wang, Xingjian | Beihang University |
Wang, Shaoping | Beihang University |
Keywords: Mobile Robotics, Field Robotics, Robot Design
Abstract: Underwater vehicles are an important technical means for the development and utilization of marine resources. With the continuous deepening of marine development, the application scenarios of Autonomous Underwater Vehicles (AUV) have been further expanded to offshore areas with dense reefs, narrow and turbid pipeline caves, and environments that require precise mechanical operation and attitude control. Traditional propeller and rudder control systems face severe challenges such as insufficient maneuverability and susceptibility to underwater object entanglement and failure. To meet the needs of AUV for propulsion and attitude control in complex application scenarios, a distributed jet propulsion method is proposed in this paper. In this study, a prototype of distributed jet propulsion based maneuverable rudderless AUV was designed and built, which realized the multi-degree-of-freedom pose adjustment of AUV and verified the superior mobility and reliability of distributed jet propulsion.
|
|
17:30-17:40, Paper SaR3B.2 | |
Sorting Space Configuration Comparison for the Multi-Agent Robotic Sorting System |
|
Starostenko, Aleksey | Moscow Institute of Physics and Technology |
Kozin, Filipp | Moscow Institute of Physics and Technology |
Gorbachev, Roman | Moscow Institute of Physics and Technology |
Khvostikova, Olga | Moscow Institute of Physics and Technology |
Gerasimov, Sergei | Moscow Institute of Physics and Technology |
Zakharova, Ekaterina | Moscow Institute of Physics and Technology |
Zaripov, Mikhail | Moscow Institute of Physics and Technology |
Davydenko, Egor | Moscow Institute of Physics and Technology |
Gonnochenko, Aleksei | Sberbank Russia |
Keywords: Multi-Robot Systems, Intelligent Transportation Systems, Intelligent Control and Systems
Abstract: Multi-agent robotic sorting systems configuration comparison is proposed in this paper. Sorting system provide item sorting task for managed fleet of robotic agents and it's path planning on sorting space graph in real time. State-of-art multi-agent path finding algorithm generate deadlock-free scheduling to perform sorting tasks. Throughput comparison research for the different configurations of sorting spare is conducted. In this paper we study throughput of three sorting space configurations with sorting center model. Robotics agent model, sorting space model and sorting system models provides environment for fleet management system. Each agent receives an action sequence from scheduling system, which takes into account the movement, the tasks of all robotic agents, the states of sorting system environment for specific sorting center configuration. The throughput of the sorting center is estimated. Configurations represents different approaches to sorting center space organization.
|
|
17:40-17:50, Paper SaR3B.3 | |
Deep Vision Guided Mobile Robot Dual Axis PID Dynamic Real Time Adjustment Following |
|
Li, Yang | Guizhou University |
Yang, Guanci | Guizhou University |
He, Ling | Guizhou University |
Zhao, Le | Guizhou University |
Zhao, Tianyi | Guizhou University |
Sun, Haoran | Guizhou University |
Keywords: Home and Personal Robot Systems, Artificial Intelligence, Intelligent Control
Abstract: The real-time tracking of moving objects by social robots is crucial for ensuring object-oriented services. This paper presents a real-time following algorithm for mobile robots based on deep vision object detection. Firstly, using a binocular depth vision camera, real-time target information from a first-person perspective is collected, and a lightweight deep learning model is employed to detect the relative position of the human body posture with respect to the social robot. Subsequently, to address the issue of unsmooth trajectory during the tracking process, a dual-axis PID algorithm is developed for cooperative control of the robot's speed and heading angle. Given the inherent randomness in human body movement, target loss can occur during tracking. To enhance the algorithm's robustness, a dynamic tracking mechanism is introduced. Finally, we implement a vision-driven two-wheel differential social robot following algorithm on the ROS system and validate it through three real-world experimental scenarios. The mean values of the target acquisition rate, tracked target position, and angle smoothness are found to be 0.81, 0.46, and 1.02, respectively.
|
|
17:50-18:00, Paper SaR3B.4 | |
Passivity Analysis for Singular Semi-Markovian Jump Systems Based on T-S Fuzzy Model |
|
Xu, Tianbo | Shenyang Jianzhu University |
Jin, Meixi | Shenyang Jianzhu University, No. 25, Middle Hunnan Road, Hunnan |
Qi, Wenhai | Qufu Normal University |
Wang, Jin | Shenyang Jianzhu University |
Keywords: Intelligent Control and Systems, Intelligent Control
Abstract: This article mainly addresses the passivity analysis problem for the fuzzy singular semi-Markovian jump systems(SS-MJSs) in a finite-time domain. The study of finite-time passivity based on the framework of singular semi-Markovian is a new field. Compared with conventional systems, singular systems cannot guarantee a unique solution, so they will face more challenges in system analysis. In this article, a systematic analysis method is proposed for SS-MJSs, which reduces the conservatism of the research. Firstly, finite-time stability conditions for fuzzy SS-MJSs are proposed. Then, the fuzzy passive control and disturbance observer based control (DOBC) are combined to address the multiple disturbance problem under the singular semi-Markovian model. Finally, the passive control scheme is designed to satisfy finite-time performance for fuzzy SS-MJSs. Based on this theoretical analysis, the effectiveness of the results is further demonstrated through numerical simulation.
|
|
SaPP |
Hall |
Poster Session |
Poster Sessions |
Chair: Xiu, Wei | Chinese Institute of Electronics |
|
14:00-18:00, Paper SaPP.1 | |
Research on Feature Point Localization Technology Based on Binocular Vision |
|
Liu, Lei | University of Jinan |
Jun, Wei | University of Jinan |
Mu, Qingsheng | University of Jinan |
|
|
14:00-18:00, Paper SaPP.2 | |
Towards Autonomous Robotic Ultrasound Scanning Using the Reinforcement Learning-Based Volumetric Data Navigation Method |
|
Shen, Cuifeng | Institute of Automation, Chinese Academy of Sciences |
Deng, Zhaokun | Institute of Automation, Chinese Academy of Sciences |
Wang, Jinzhi | College of Engineering, Peking University |
Wang, Shuangyi | Chinese Academy of Sciences |
Chen, Chen | Institute of Automation, Chinese Academy of Sciences |
Keywords: Deep Learning, Robot Vision and Computer Vision, Medical Robotics
Abstract: Ultrasound scanning is an indispensable diagnostic tool in modern medicine, but obtaining an accurate standard plane relies heavily on the sonographer's skill and expertise. This paper presents a novel approach toward autonomous robotic ultrasound scanning using a reinforcement learning-based navigation method with volumetric ultrasound data. The proposed method enables an ultrasound robot to autonomously acquire high-quality 2D standard planes from patients, thereby reducing the need for human intervention. The core of the approach is a deep reinforcement learning algorithm that detects 2D standard planes within the 3D ultrasound volume and guides the ultrasound probe to extract high-quality standard planes. To assess the effectiveness of the 2D standard plane detection method, an evaluation was conducted using a real-world ultrasound dataset, and a comparison was made between its performance and that of other state-of-the-art autonomous approaches. The simulation experiment was also conducted to validate the autonomous robotic ultrasound acquisition workflow. Results demonstrate that the reinforcement learning-based volumetric data navigation method outperforms existing techniques in terms of standard plane accuracy and is very effective for volumetric data-based robotic navigation.
|
|
14:00-18:00, Paper SaPP.3 | |
Towards Automated Ultrasound Scanning Using Vision-Based Navigation: From Physician Skill Learning to Robotic Reproduction |
|
Hao, Mingrui | Institute of Automation, Chinese Academy of Sciences |
Zhang, Pengcheng | Institute of Automation, Chinese Academy of Sciences |
Chen, Chen | Institute of Automation, Chinese Academy of Sciences |
Wang, Shuangyi | Chinese Academy of Sciences |
Keywords: Human-Robot Interaction and Cooperation, Robot Vision and Computer Vision, Medical Robotics
Abstract: Ultrasound scanning is widely employed for disease diagnosis and procedure guidance. However, conventional ultrasound depends on the sonographer’s manual operation, resulting in poor repeatability. Although robot-assisted ultrasound has been extensively studied, it has been a challenge in the field to further automate the examination and make the robotic system more practical for primary use. In this study, we propose a visual navigation method based on the skill learning of the sonographer. The key information about the probe's movement, including position and angle, was recorded using vision-based methods on 14 cases based on a professional sonographer's practice. Then, an automatic scanning method based on visual guidance, human-machine collaboration, and path planning was designed using a novel parallel mechanism-based ultrasound robot. The strategy of traversal scanning is a key consideration in facilitating efficient coverage of desired ranges and involves two primary steps: waypoint selection and waypoint planning. To compare the efficiency of different schemes proposed for these two respective stages, a simulation study and an actual running time test of the robot were conducted to assess the weighted coverage, energy, and time consumption. The results demonstrate that the proposed traversal strategy can effectively cover the statistical area formed by the physician's scanning experience, and the new workflow is promising to be tested in-vivo in our future work.
|
|
14:00-18:00, Paper SaPP.4 | |
Research on Absolute Positioning Error Compensation of Serial 6-DOF Industrial Robot (I) |
|
Cui, Rongyue | University of Jinan |
Jun, Wei | University of Jinan |
Liang, Liang | SIASUN Robot & Automation CO., Ltd |
Keywords: Industrial Robotics and Factory Automation, Intelligent Control and Systems
Abstract: To improve the absolute positioning accuracy of the robot end, a kinematic model of the tandem robot considering the D-H parameter error is established; a laser tracker is used to track and detect the motion positioning of the robot end, and the least squares method is used to solve the robot D-H parameter error; the robot motion model is corrected and compensated according to the geometric error, and it is experimentally verified that the method in this paper can identify and compensate the D-H parameter errors existing in the robot motion model. ABB IRB2400 robot as a research object, the maximum error of absolute positioning of the robot before the error compensation is 1.6732mm, and the average error is 1.110422mm. The absolute positioning accuracy of the robot is significantly improved.
|
|
14:00-18:00, Paper SaPP.5 | |
Research on Capacitance Compensation for Temperature Changes in Laser Cutting Systems (I) |
|
Hu, Jingru | University of Jinan |
Jun, Wei | University of Jinan |
Luan, Xianye | Shandong SIASIN Industrial Software Research Institute Co., Ltd |
Gong, Fei | Shandong SIASIN Industrial Software Research Institute Co., Ltd |
Keywords: Intelligent Control and Systems, Sensing, Haptic System, Sensor Networks
Abstract: To address the problem that changes in ambient temperature during the laser cutting process will change the capacitance value measured between the nozzle of the cutting head and the plate, thus affecting the cutting effect. In this paper, the capacitance value is collected by the STM32F405RGT6 microprocessor and PCAP02, and the relationship curve between the capacitance value and the temperature value is collected to get the slope value of the curve in each temperature interval, and the compensation capacitance value is obtained by using the compensation formula. By verifying it on the CODESYS platform, the compensation is turned on when the ambient temperature changes and the compensated capacitance value is restored to the calibrated capacitance value within 1ms, the Z-axis change curve fluctuates normally, and the dynamic following accuracy is restored to 0.1mm. The design enables the laser cutting system to be unaffected by changes in ambient temperature that may affect cutting results.
|
|
14:00-18:00, Paper SaPP.6 | |
Aircraft Skin Seam Location and Step Difference Measurement Based on 3D Point Cloud |
|
Yu, Haiwen | Nanjing University of Aeronautics and Astronautics |
Chen, Bai | Nanjing University of Aeronautics and Astronautics |
Duan, Jinjun | Nanjing University of Aeronautics and Astronaut |
Ding, Liping | Nanjing University of Aeronautics and Astronautics |
Keywords: Robot Vision and Computer Vision, Industrial Robotics and Factory Automation
Abstract: In order to meet the requirements of aircraft skin assembly quality, it is necessary to eliminate step differences around the seam of the skin. The skin seam location and step difference measurement before grinding are the key steps to determine the grinding accuracy. This paper proposes a point cloud processing algorithm of seam location and step difference measurement for grinding trajectory generation of the curved components. First, extract the boundary features of the preprocessed workpiece point cloud, and divide the point cloud into different regions; Second, recognize the boundary feature points, and construct the boundary line by fitting the boundary points; Finally, calculate the step difference based on the boundary line of both sides and generate the solution for grinding track generation based on the seam boundary position and the step difference. We use the line structured light based vision measurement platform to verify the proposed algorithm, and the results show that the measurement system and algorithm can achieve the seam boundary location of curved components accurately. The research work in this paper is extendable to applications in the machining of curved workpieces.
|
|
14:00-18:00, Paper SaPP.7 | |
Design and Implementation of a New Spherical Robot with an Elastic Pawl |
|
Fang, Shihao | Beijing University of Civil Engineering and Architecture |
Qin, Jianjun | Beijing University of Civil Engineering and Architecture |
Cao, Yu | Beijing University of Civil Engineering and Architecture |
Shao, Pai | Beijing University of Civil Engineering and Architecture |
Keywords: Mobile Robotics, Robot Design, Rescue Robotics
Abstract: In order to improve the obstacle-surmounting performance and adaptability of the spherical robot in unstructured terrain, a new spherical robot with an elastic pawl is designed. While retaining the original structural characteristics of the spherical robot, the elastic pawl deformation mechanism and axial expansion mechanism are introduced to enable the robot to switch various motion modes. The parameters of the deformation mechanism are designed through the establishment of a mathematical model, and the kinematic analysis of the obstacle-surmounting process is carried out. ADAMS dynamic simulation software is used to simulate the robot in a variety of typical terrain environments. On this basis, the physical prototype is manufactured and assembled, and the traveling test under various terrains is carried out, verifying the robot structure's feasibility and practicability.
|
|
14:00-18:00, Paper SaPP.8 | |
Development and Kinematic Accuracy Analysis of Supernumerary Robotic Limbs |
|
Liao, Ziyu | NUAA |
Chen, Bai | Nanjing University of Aeronautics and Astronautics |
Zheng, Qian | Nanjing University of Aeronautics and Astronautics |
Xu, Jiajun | Nanjing University of Aeronautics and Astronautics |
Tu, Zhendong | Nanjing University of Aeronautics and Astronautics |
Wu, Hongtao | Nanjing University of Aeronautics and Astronautics |
Keywords: Rehabilitation and Assistive Robotics, Robot Design
Abstract: This paper presents a supernumerary robotic limbs (SRLs), which have an additional four degree-of freedoms (DOF) robotic limb mounted on the waist of the wearer. In order to analyze the kinematic accuracy of the SRLs, this paper built the kinematic model and adjoint error model based on the Product of Exponential (POE) theory. For facilitating the parameter identification of the adjoint error model, this paper established a linear error model with only position measurement of end-effector. The experiment result verify the validity of the error model and it is enough precise for most practical manipulations. Meanwhile, comparing the kinematic accuracy of fixed base and floating base of SRLs, the results show that motion precision will be significant loss when the SRLs worn on the human body. This indicates that in the process of collaboration, SRLs may not be suitable for assigning tasks with high demands for motion accuracy.
|
|
14:00-18:00, Paper SaPP.9 | |
Research on Tail Gas Optimization Method Based on Adaptive Dynamic Programming |
|
Song, Xue | Changchun University of Science and Technology |
Yang, Yang | Changchun University of Science and Technology |
Song, Yaru | Changchun University of Science and Technology |
Li, Mingqiu | Changchun University of Science and Technology |
Keywords: Artificial Intelligence, Intelligent Control, Emerging Technologies and Applications
Abstract: With the increase in the number of motor vehicles, motor vehicle emissions have become one of the most important factors affecting air quality. For the problem of excessive motor vehicle emissions at signal intersections, this paper proposes a tailpipe optimization method based on adaptive dynamic programming. The proposed method uses back-propagation (BP) neural networks for the model network, execution network and evaluation network. According to the relationship between motor vehicle delay time and tailpipe emission, the tailpipe emission model is established, and then the model network is trained separately, and the optimal weights and thresholds of BP neural network are obtained by using the fast convergence speed and powerful search ability of genetic ant colony fusion algorithm. Finally, the obtained weights and thresholds are brought into the execution network and evaluation network for optimization to achieve the goal of minimizing exhaust emissions at signalized intersections by changing the duration of signal lights. The feasibility and effectiveness of the proposed method are verified by simulation experiments. The experimental results show that the proposed method can significantly reduce vehicle exhaust emissions, resulting in higher fuel efficiency and lower emission levels for vehicles at signalized intersections.
|
|
14:00-18:00, Paper SaPP.10 | |
Trajectory Planning of Robot End-Effector Position and Orientation Based on Cubic Spline |
|
Liu, Yang | Northeastern University |
Du, Zhenjun | SIASUN Robot & Automation |
Liu, Mingmin | SIASUN Robot & Automation |
Tao, Tao Yongliang | Northeastern University |
Li, Chengpeng | SIASUN Robot & Automation |
Keywords: Industrial Robotics and Factory Automation, Path and Motion Planning, Intelligent Control
Abstract: Trajectory planning for the end effector of industrial robots has been a hot topic in the field of robotics. A continuous and smooth trajectory of the end effector's pose is beneficial for avoiding unnecessary impacts and vibrations, thereby improving the performance and lifespan of industrial robots. In recent years, researchers have predominantly focused on the trajectory planning of multi-directional variations in orientation using quaternions, while giving less attention to the coordinated planning of position and orientation. To address this issue, this paper proposes an efficient trajectory planning method based on improved spline curve interpolation. The orientation of key pose points, characterized by multi-directional variations, is represented using quaternions. By employing a quaternion dimension reduction algorithm, these quaternions are transformed into vector forms in three-dimensional space, allowing for unified representation of position and orientation within the same dimension. Taking the chord length between key pose points as a parameter, the position and orientation components are subjected to cubic spline curve interpolation with chord length as the independent variable. The proposed method is applicable for coordinated planning of position and orientation trajectories, ensuring high-order continuity of the pose trajectory. The effectiveness and rationality of the proposed method are validated through a set of key pose points featuring multi-directional
|
|
14:00-18:00, Paper SaPP.11 | |
A Human-Machine Interactive Simulation Environment for Intelligent Analysis of Human Factors Reliability |
|
Wang, Chunlei | Shanghai University |
Wang, Xiao | Shanghai University |
Keywords: Human-Robot Interaction and Cooperation, Artificial Intelligence, Human-Machine Interface
Abstract: Human factors reliability analysis holds a crucial significance within the domain of human-machine interactive systems. Conventional analysis methods rely on real-world interaction data and are subject to analysis by domain experts, yet they possess limitations concerning operability and subjectivity. This paper introduces a Human Factors Reliability Research Task Simulation Platform, aiming to investigate human intelligence and behavior within simulated interactive environments. The platform not only serves as an interactive simulation environment but also integrates an intelligent analysis module for human factors reliability analysis. Specifically, three types of tasks, including simulations of multiple sub-tasks, have been developed to suit typical aerospace missions. By merging human-machine interaction simulation with the actual operational process, the platform effectively monitors various aspects of participant data. Through the interactive process between individuals and the environment, we have successfully and accurately predicted the error probabilities of participants in actual tasks, achieving outcomes comparable to domain experts.
|
|
14:00-18:00, Paper SaPP.12 | |
Dynamic Modeling of the Hybrid Manipulator |
|
Chen, Weifeng | Tianjin University |
Liu, Haitao | Tianjin University |
Li, Guangxi | Tianjin University |
Bai, Yujie | Tianjin University |
Keywords: Industrial Robotics and Factory Automation, Robot Design
Abstract: This paper presents an approach for dynamic modelling of the hybrid manipulator based on the principle of virtual work and screw theory. The proposed approach is developed under the framework of the generalized Jacobian and Hessian matrix, leading to the calculation of the intensity of the wrench imposed on the actuated joint in a straightforward manner. A case study is developed to verify the proposed approach.
|
|
14:00-18:00, Paper SaPP.13 | |
An Approach for Upper Limb Movement Intention Recognition Using EEG and sEMG Fusion Based on the MCPSA-CIIM |
|
Zhang, Weiming | Xi’an Jiaotong University |
Zhang, Xiaodong | Xi’an Jiaotong University |
Xu, Cheng | Science and Technology on Complex System Control and Intelligent |
Zhou, Guchuan | Science and Technology on Complex System Control and Intelligent |
Zhang, Teng | Xi’an Jiaotong University |
Wang, Yu | Xi’an Jiaotong University |
Keywords: Brain-Machine Interface, Deep Learning, Medical Robotics
Abstract: The upper limb motion intention recognition method based on EEG and sEMG fusion has achieved significant results in fields such as prosthetic control. However, most existing fusion methods use manual means to select features, which cannot capture temporal and spatial features at different scales, and ignore the correlation features between the two types of signals. To address these issues, this article proposes a fusion recognition method for upper limb motion intention EEG and sEMG based on Multi-scale Convolution, Polarized Self- Attention, and Cross Intelligence Integration Module. Extract multidimensional temporal and spatial features of EEG and sEMG through multi-scale convolution, and introduce polarized self-attention mechanism to filter the extracted multi-scale features. Simultaneously using cross enhancement strategy to extract correlation features between EEG and sEMG. Finally, the features are input into the classification network for recognition. This method was validated on the Jeong database, and the results showed that compared with CNN-LSTM and EEGNet, the recognition accuracy of this method increased by 2.63% and 3.15%, respectively.
|
|
14:00-18:00, Paper SaPP.14 | |
Research on Dermatological Classification Algorithm Based on the Fusion Model |
|
Zhang, Chenjie | Changchun University of Science and Technology |
Wang, Shuai | Changchun University of Science and Technology |
Zhang, Yu | Changchun University of Science and Technology |
Hu, Hanping | Changchun University of Science and Technology |
Keywords: Artificial Intelligence, Machine Learning, Deep Learning
Abstract: A new dermatology classification algorithm is proposed to use CNN instead of Patch Embedding part of Swin Transformer for local feature extraction. The method gives full play to the advantages of CNN in extracting local features, while exploiting the excellent ability of Transformer in extracting global features, realizing the complementary advantages between the two. In addition, the SimAM self-attentive mechanism is introduced in the CNN part, and after training on the ISIC2019 dataset, the fused network achieves an accuracy of 90.6%, which improves the performance by 7.4% and 2.1% relative to the methods applying ResNet and Swin Transformer independently. These results demonstrate the significant advantages of the fused algorithm in the field of dermatology classification.
|
|
14:00-18:00, Paper SaPP.15 | |
Garbage Image Classification Algorithm Based on Swin Transformer |
|
Hu, Hanping | Changchun University of Science and Technology |
Wang, Shuai | Changchun University of Science and Technology |
Zhang, Chenjie | Changchun University of Science and Technology |
Pan, Yue | Changchun University of Science and Technology |
Keywords: Artificial Intelligence, Machine Learning, Deep Learning
Abstract: By comparing the accuracy of VGG16, ResNet101 and Swin Transformer on the garbage classification dataset, it is proved that Swin Transformer has an advantage in garbage classification tasks. Experimental results show that, with the self-attention mechanism and multi-scale feature modeling capabilities, Swin Transformer exhibits higher accuracy than traditional convolutional neural network algorithms. The accuracy of the Swi Transformer algorithm reaches 97%, which is 2.9% higher than VGG16 and 1.1% higher than ResNet101.
|
|
14:00-18:00, Paper SaPP.16 | |
Magnetically Controlled Nanorobotic Swarm for the Glioma Therapy |
|
Li, Chan | Beihang University |
Hu, Zihan | Beijing University of Aeronautics and Astronautics |
Sun, Hongyan | Beihang University |
Wang, Chutian | Beihang University |
Zeng, Zijin | Beihang University |
Feng, Lin | Beihang University |
Keywords: Micro/Nano Robotics, Medical Robotics
Abstract: Gliomas have a specific pathogenesis, high mortality rate and very low cure rate. In the current treatment of glioma, the presence of the blood-brain barrier severely hinders drug delivery. To address this problem, we propose a drug delivery strategy via temozolomide-loaded micro-nano robots, which is expected to break the blood-brain barrier and precisely enrich in the glioma region. We have validated in vitro the effect of driving control of the drug-loaded magnetically controlled micro-nano robot. A specially designed magnetic drive control system using a rotating magnetic field allows for cluster control of the drug-loaded magnetically controlled micro-nano robots, finding a relationship between the speed of movement and the frequency of magnetic field rotation, and a preliminary attempt to break the blood-brain barrier in healthy mice. This study provides a good solution strategy for drug-laden magnetically controlled micro-nano robots to break the blood-brain barrier for the treatment of glioma.
|
|
14:00-18:00, Paper SaPP.17 | |
Multi-Class Industrial Texture-Less Object Detection Method in Cluttered and Occluded Scenes |
|
Li, Sicong | Shenyang Institute of Automation, Chinese Academy of Sciences |
Zhu, Feng | Shenyang Institute of Automation, Chinese Academy of Sciences |
Wu, Qingxiao | Shenyang Institute of Automation Chinese Academy of Sciences |
Keywords: Industrial Robotics and Factory Automation, Robot Vision and Computer Vision, Deep Learning
Abstract: in industrial scenes, template matching algorithms have the advantages of robustness and fastness, but most of them tend to focus on single-class object detection. When facing multi-class object detection tasks, the performance degrades significantly. Deep neural networks are naturally suitable for this task, but rotation regression is still a problem that has yet to be solved perfectly, making it challenging to meet the requirements of industrial robot sorting and grasping scenes. We proposed a hybrid two-stage method called key-pair matching (KPM) algorithm, which can perform robust, fast, and accurate detection of multi-class industrial workpieces. In the first stage, we choose one off-the-shelf CNN detector to predict the horizontal bounding boxes to solve the classification and coarse localization problems in complex scenarios, in the second stage, features based on key-shapes primitives are used to calculate translations and rotations. Experiments show that our proposed method outperform the other comparison methods regarding detection performance and localization accuracy under reasonable runtime, making it suitable for industrial objects detection in cluttered and occlusion scenes
|
|
14:00-18:00, Paper SaPP.18 | |
A New CCA-Based Method for Improving SSVEP-Based BCI System Classification |
|
Cui, Yingxue | Zhengzhou University of Light Industry |
|
|
14:00-18:00, Paper SaPP.19 | |
Deep Learning Based Defect Detection Algorithm for Solar Panels |
|
Li, Jiaqi | University |
Li, Hongxu | Yanshan University |
Wu, Yifan | University of Alberta |
Zhou, Hailiang | Hebei Baoding Jiasheng Photovoltaic Technology Co., Ltd |
Manfredi, Luigi | University of Dundee |
Li, Peng | Yanshan University |
Zhang, Hong | University of Alberta |
Keywords: Deep Learning, Artificial Intelligence, Robot Vision and Computer Vision
Abstract: Defect detection of solar panels plays an essential role in guaranteeing product quality within automated production lines. However, traditional manual inspection of solar panel defects suffers from low efficiency. This paper proposes an enhanced YOLOv5 algorithm (EL-YOLOv5) fused with the CBAM hybrid attention module to ensure product quality. The algorithm focuses on detecting five common types of defects that frequently appear on photovoltaic production lines, namely hidden cracks, scratches, broken grids, black spots, and short circuits. This study utilizes publicly available solar panel datasets, as well as datasets collected from actual photovoltaic production lines. These datasets are annotated accordingly and used to train the proposed algorithm. The experimental results demonstrate that the proposed algorithm achieves good performance on both the public and actual solar panel defect datasets. Particularly in actual datasets, where defect features are often less apparent and defects are smaller in size, the proposed algorithm can still detect even minor black spots.
|
|
14:00-18:00, Paper SaPP.20 | |
Event-Potential Method for Mobile Robot Motion Planning and Control |
|
Sheng, Yongji | The University of Hong Kong |
Ye, Jiajie | The University of Hong Kong |
Wang, Siyu | The University of Hong Kong |
Liu, Xinyu | The University of Hong Kong |
Xi, Ning | The University of Hong Kong |
Keywords: Mobile Robotics, Path and Motion Planning, Intelligent Control and Systems
Abstract: Motion planning and control of mobile robots is a challenging but crucial task for autonomous navigation. Since the actual environment is usually highly dynamic and has extremely high uncertainty and unpredictability, motion planning and control methods need to handle unexpected events and uncertainties in the environment, such as dynamic obstacles. Existing motion planning and control methods are usually decoupled processes. When unexpected events occur, the planner replans a new feasible plan based on new sensory information, and then the controller executes the new plan. Existing methods usually perform the perception, replanning, and control loop in real time. Replanning is an inefficient and time-consuming process since the robot needs to process new perception information and restart the planner. Also, the plan is not fully executed before the next loop which wastes most of the plan. We tackle this problem by introducing an event-potential method: a novel event-based potential field motion planning and control method that introduces event reference instead of time reference, which can locally modify the preplanned velocity along the trajectory using the resultant force generated by the attractive force from a trajectory and repulsive force from obstacles for obstacle avoidance. After completing obstacle avoidance, the event reference allows the robot to continue to track the original trajectory without replanning.
|
|
14:00-18:00, Paper SaPP.21 | |
Kinematics Solution of Cable-Driven Parallel Robots Considering Cable Sagging (I) |
|
Gao, Feng | University of Science and Technology of China |
Gao, Lifu | Hefei Institutes of Physical Science, Chinese Academy of Science |
Wang, Daqing | Hefei Institutes of Physical Science, Chinese Academy of Science |
Jiang, Man | Hefei Institutes of Physical Science, Chinese Academy of Science |
Keywords: Intelligent Control and Systems, Intelligent Control, Robot Design
Abstract: This paper discusses the kinematics solution method of a large-span redundantly restrained cable-driven parallel robot (CDPR).The kinematics solution of large-span redundantly restrained CDPR involves challenges related to kinematic and static coupling, when considering the influence of cable mass. This paper focuses on deriving the kinematic equations of the CDPR based on the catenary model, and the cable tension solution for the redundantly restrained CDPR is obtained with the minimum p-norms of the cable tensions as the optimization objective. Moreover, a numerical calculation method is proposed for the forward and inverse kinematics of the CDPR, which is based on the trust-region dogleg (TRD) algorithm.Additionally, to enhance the efficiency of the numerical calculation method, we present an inverse kinematics solution method that combines the back-propagation (BP) neural network with the particle swarm optimization (PSO) algorithm and the trust-region dogleg algorithm (BPPSO-TRD). The simulation results highlight the importance of considering the cable mass when solving the kinematics for large-span CDPR. Compared to the TRD algorithm, the BPPSO-TRD algorithm can achieve 47.59% reduction in solution iteration and 22.22% reduction in solution time per case. This validates the effectiveness and reasonableness of the algorithm, while establishing a theoretical foundation for the position control of large-span redundantly restrained CDPR.
|
|
14:00-18:00, Paper SaPP.22 | |
Modeling and Simulation of Submarine Launched UAV Formation and Formation Reconfiguration with Initial Discrete (I) |
|
Jia, Huayu | University of Chinese Academy of Sciences |
Zheng, HuiLong | University of Chinese Academy of Sciences |
Zhou, Hong | University of Chinese Academy of Sciences |
Zhang, Qian | University of Chinese Academy of Sciences |
Li, XinSong | Shenyang Aerospace University |
Keywords: Intelligent Control, Intelligent Control and Systems, Multi-Robot Systems
Abstract: In order to meet the needs of homogeneous submarine-launched UAV clusters in complex environments, UAV formation and formation reconfiguration algorithms with discrete initial states are investigated. A fast cluster formation method is proposed, and the effectiveness of this method is verified through comparative experiments, which can effectively reduce the entanglement of UAV trajectories in the process of clustering, reduce the probability of collision, and effectively shorten the time of forming cluster formation. Based on second-order consensus and DWA localized path planning algorithms, we improve the obstacle avoidance constraint function, design the speed and distance constraint cost function, propose the space-time coupled formation and formation reconstruction method, and verify the feasibility and effectiveness of this method, which provide some technical support for the subsequent transformation to engineering applications.
|
|
14:00-18:00, Paper SaPP.23 | |
Edge Detection and Alignment Control Method for Mobile Robots Based on Multi-Lidar Fusion |
|
Gao, He | BeiHang University |
Tao, Yong | Beijing University of Aeronautics and Astronautics |
Liu, Xiaonan | Beijing New Materials and New Energy Technology Development Prom |
Zhang, Yufan | Beihang University |
Liu, Haitao | BUAA |
Wan, Jiahao | Beihang University |
Keywords: Industrial Robotics and Factory Automation, Mobile Robotics, Intelligent Control
Abstract: Currently, there is a growing demand for intelligent maintenance of large-scale infrastructure, prompting a shift from manual to semi-automatic solutions in the control schemes of inspection robots. Mobile inspection robots play a crucial role in tasks such as flip-over inspection, fall prevention warning, defect identification, all of which necessitate edge detection and alignment processes. Laser alignment, with its high precision and low computational complexity, offers significant advantages over manual observation and visual alignment methods. This paper addresses practical engineering requirements by proposing a lidar-based structural edge detection algorithm and an automatic alignment controller. To validate the effectiveness of the proposed method, edge detection and alignment control experiments were conducted in a simulated environment. The results show that the edge detection algorithm reduces the direction detection error to 2.7°. The developed alignment controller exhibits a convergence time approximately 40% shorter than the standard PID controller. The proposed scheme effectively meets the accuracy and real-time requirements for edge detection and automatic alignment functions in mobile inspection robots and holds significant application potential for the infrastructure project's intelligent maintenance.
|
|
14:00-18:00, Paper SaPP.24 | |
Toward Safe Human-Robot Interaction: More Efficient Safety Monitoring Technology for Robotic Arms (I) |
|
Gong, Zihang | YanTaiUniversity |
Liu, Zhaowei | School of Computer and Control Engineering, Yantai University |
Su, Hang | Politecnico Di Milano |
|
|
14:00-18:00, Paper SaPP.25 | |
A Novel Dynamic Motion Planning Based on Error Tolerance Batch Informed Tree |
|
Cao, Zhimin | Suzhou City University |
Keywords: Path and Motion Planning
Abstract: In a static environment for robotic manipulators, many present motion planning algorithms can generate collision-free paths effectively. But in a human-robot work scenario or a unstructured environment, these algorithms may fail or difficultly avoid obstacles in real-time planning due to missing updating environment model. To realize dynamic obstacle avoidance, a novel dynamic motion planning algorithm, called error tolerance batch informed tree (et-BIT*) is proposed. Experiments showed that the et-BIT* planner can avoid a suddenly appeared obstacle and bypass it by choosing a candidate motion path.
|
|
14:00-18:00, Paper SaPP.26 | |
Gait Planning and Motion Control Based on Vrep Simulation for Quadruped Robot (I) |
|
Zhou, Linqi | Nanchang Hangkong University |
Chen, Zhihua | Nanchang Hangkong University, Beijing Institute of Technology |
Liu, Jun | Nanchang Hangkong University |
Liu, Zhi | Beijing Institute of Technology |
Chen, Yumeng | Nanchang Hangkong University |
Zhang, Liting | Nanchang Hangkong University |
Keywords: Intelligent Control, Mobile Robotics, ROS, Software System for Robotics Application
Abstract: Gait planning of quadruped robots plays an important role in achieving less walking, including dynamic and static gait. In this article, a static and dynamic gait control method based on center of gravity stability margin is proposed. Firstly, the robot model and kinematics modeling are introduced. Secondly, the robot's foot static and dynamic gait were planned and the foot trajectory was designed. Finally, two types of gait of the robot were simulated using Vrep simulation software, and the differences in stability and speed between the coordinated gait with speed and stability in the static and dynamic gait of a 12 degree of freedom robot were analyzed, verifying the effectiveness of the gait control method proposed in this paper.
|
|
14:00-18:00, Paper SaPP.27 | |
Motion Planning and Tracking Control Method Based on CoppeliaSim for UGV in Irregular Terrain (I) |
|
Huang, Jiale | School of Information Engineering, Nanchang Hangkong University |
Chen, Zhihua | Nanchang Hangkong University, Beijing Institute of Technology |
Zhong, Yongbo | Nanchang Hangkong University |
Zeng, Wenrui | Nanchang Hangkong University |
Ji, Chuanmin | NCHU |
Keywords: Intelligent Control, Sensing, Haptic System, Mobile Robotics
Abstract: Path planning and tracking algorithms are one of the cores of autonomous navigation technology for unmanned ground vehicle (UGV). In this paper, a local obstacle avoidance method based on event-triggered control is proposed to ensure that the UGV avoids obstacles in the tracking process. First, the path planning and tracking control problem of the UGV is modeled and formulated. Second, the improvement and logical correlation of the relevant algorithms are described in detail. Finally, a joint simulation of MATLAB and V-rep is performed. The simulation results show that event-triggered control can largely improve the reliability of the UGV path planning and tracking process, and verify the effectiveness of the algorithm.
|
|
14:00-18:00, Paper SaPP.28 | |
Embroidery Robotics: An Innovative Approach to AI-Based Online Learning and Image Generation (I) |
|
Chen, Ling | Hunan Normal University |
Zhang, Tianxin | Hunan Normal University |
Chen, Jing | Hunan Normal University |
Keywords: Home and Personal Robot Systems, Grasping and Manipulation, Intelligent Control and Systems
Abstract: In this paper, we present an innovative approach to employ artificial intelligence for teaching a dual-arm robot embroidery tasks through online learning. Utilizing the principle of Teaching by Demonstration, we introduce an online learning algorithm that efficiently captures and reproduces the embroidery trajectories of a human expert on a 2D plane. The algorithm's performance was evaluated quantitatively and qualitatively, showing a promising level of accuracy in reproducing the original embroidery patterns. We further validated the feasibility and potential of our approach through a user feedback survey involving 30 professional embroiderers. Despite certain areas identified for improvement, notably the finer details in embroidery, the overall positive response confirmed the viability of our approach. This study signifies a crucial step towards automating intricate manual tasks and broadens the horizons for applications of AI and robotics in various industries.
|
|
14:00-18:00, Paper SaPP.29 | |
A Fuzzy Inference System on Probabilistic Roadmap for Multi-Robot Path Planning |
|
Replogle, Brandon | California Polytechnic State University, San Luis Obispo |
Yu, Xiao-Hua | California Polytechnic State University, San Luis Obispo |
Keywords: Path and Motion Planning, Machine Learning
Abstract: In recent years, multi-robot systems have been widely used in many applications such as warehouse inventory tracking, homecare, natural resource monitoring, emergency search and rescue operations, etc. Probabilistic roadmap (PRM) is a typical path planning algorithm that determines an optimal trajectory once the robot start and goal configurations are specified. However, when the number of robots in the system increases, it converges very slowly and may even fail to find the solution. Fuzzy inference system resembles human reasoning and decision-making process and is very robust and efficient when dealing with complicated situations. In this research, a fuzzy inference system is proposed and combined with the probabilistic roadmap algorithm for multi-robot path planning. Computer simulation results for three different environments show that, compared with the standard PRM algorithm, this approach is very effective to reduce computational cost which is especially important for real-time applications. In the most complicated scenario studied in this paper, the proposed approach can reduce the run time by 23.92% with a trade-off of increased average path length of 6.83%.
|
|
14:00-18:00, Paper SaPP.30 | |
Study on the Grinding Speed Model of Eccentric Shaft Components of the RV Reducer for Robotic Arm |
|
Jiang, Zhemin | Beijing Beikai Electric Co |
Fang, Haicheng | Beijing Beikai Electric Co |
Li, Chao | Beijing Beikai Electric Co |
Liu, Shilei | Beijing Beikai Electric Co |
Keywords: Industrial Robotics and Factory Automation, Smart Structures, Materials, Actuators, Intelligent Control and Systems
Abstract: This paper establishes a model for the grinding motion relationship of eccentric shaft profiles in the RV reducer used for plane rigid body motion of the robotic arm. It clarifies the velocity and acceleration constraint relationship between the X and C axes and provides a quantitative expression for the feed speed of the grinding point. The changing trends of the feed speed and acceleration of the grinding wheel under constant angular speed and constant linear speed conditions are analyzed and compared. It is concluded that workpieces with large eccentricity are not suitable for constant linear speed grinding, providing a basis for further analysis of the causes of profile errors and error compensation in the dynamic grinding process.
|
|
14:00-18:00, Paper SaPP.31 | |
An Event-Driven Asynchronous Feature Tracking Method (I) |
|
Xu, Haidong | Soochow University |
Jin, Shizhao | Soochow University |
Yu, Shumei | Soochow University |
Sun, Rongchuan | Soochow University |
Sun, Lining | Soochow University |
Keywords: SLAM and Navigation, Robot Vision and Computer Vision, Mobile Robotics
Abstract: Compared to traditional frame-based cameras, event cameras offer a few advantages due to their bio-inspired characteristic, including low latency and high dynamic range, as each pixel can independently respond to changes in brightness when generating event streams. However, the unconventional output of event cameras brings challenges to basic computer vision problems. Recent research in event-based feature tracking has focused on using information from event streams to achieve asynchronous tracking of feature positions. We proposed an improved feature tracking method with complementary information from both the event and frame-based cameras to update feature positions asynchronously. A nearest neighbor algorithm is used for data association to filter new feature points, and the optimization quality is evaluated by the cost function values. This method utilizes the detection of FAST corners on the RGB camera to initialize the missing objects or targets, which are points exhibiting significant local pixel brightness changes. The algorithm has been implemented in C++ and estimated using publicly available datasets. The experimental results indicate that our method boosts the feature age by 9% ~ 24% while maintaining tracking accuracy compared to the Event-based Kanade-Lucas-Tomasi tracker (EKLT) method.
|
|
14:00-18:00, Paper SaPP.32 | |
Design of an Impact-Resistant Elbow Mechanism for Wheel-Legged Robot Crawling |
|
Ma, Xs | Beijing Institute of Technology |
Gao, Junyao | Beijing Institute of Technology |
Qiu, Xuejian | Beijing Institute of Technology |
Zhao, Lingxuan | Beijing Institute of Technology |
Chen, Xuechao | Beijing Insititute of Technology |
Yu, Zhangguo | Beijing Institute of Technology |
Huang, Qiang | Beijing Institute of Technology |
Keywords: Humanoid Robots, Smart Structures, Materials, Actuators
Abstract: Wheel-legged robots utilize their arms for efficient crawling in narrow and passages. However, during the transition from a standing to a crawling motion, wheeled-legged robots, which are underactuated systems, are susceptible to falling and potential damage. Additionally, the elbow joint requires a high torque during robot crawling. Therefore, we designed an elbow mechanism driven by a linear actuator for BIT-W robot. The torque requirement of the elbow joint is analyzed by constructing a crawling model of the BIT-W robot. Subsequently, a novel optimization method is proposed to enhance impact resistance by optimizing the mechanical parameters, based on the analysis. Finally, The experiments conducted validate the elbow mechanism's capability to provide stable support for crawling and motion mode switching. Moreover, the calculated impact mitigation factor (IMF) results substantiate the mechanism's effective enhancement of the arm's ability to withstand impacts.
|
|
14:00-18:00, Paper SaPP.33 | |
Movement Primitives with Explicit Constraints for Imitation Learning of Pick-And-Place Action |
|
Dong, Yue | Beijing Institute of Technology |
Yu, Zhangguo | Beijing Institute of Technology |
Chen, Xuechao | Beijing Insititute of Technology |
Wang, Chenzheng | Beijing Instiitute of Technology |
Huang, Qiang | Beijing Institute of Technology |
Keywords: Education Robotics, Human-Robot Interaction and Cooperation, Intelligent Control and Systems
Abstract: Movement primitives are the basic units of motions for complex behaviors. They are widely used in imitation learning because they provide leaning and generalizing operator for demonstrated behaviors. As a trajectory-level encoding method, movement primitives have encountered problems when learning and generalizing under certain constraints. This paper proposes Movement Primitives with Explicit Constraints (MPs-EC) for the imitation learning of pick-and-place actions. This methodology learns constraints from demonstration and then reproduces them in generalized trajectories. Meanwhile, an optimization method converts the learned constraints into the scalar weight of a target object to realize constraint reproduction. In this way, the stable constraint reproduction in the generalized trajectory ensures successful tasking. A simulation of fixed orientation pick-and-place action and an experiment of dual-arm pick-and-throw action verify the effectiveness of the proposed movement primitives.
|
|
14:00-18:00, Paper SaPP.34 | |
Depth Control of Cable Patrol Autonomous Underwater Vehicle Based on Reinforcement Learning (I) |
|
Zhang, Yuenan | Hangzhou Dianzi University |
Zhang, Xueting | Hangzhou Dianzi University |
Zeng, Fanzong | Hangzhou Dianzi University |
Sun, Shangwei | Hangzhou Dianzi University |
Liu, Jingbiao | Hangzhou Dianzi University |
Zhou, Zhe Fang | Hangzhou Dianzi University |
Guo, Chenhang | Hangzhou Dianzi University |
Wang, Dongrui | Hangzhou Dianzi University |
Keywords: Intelligent Control, Machine Learning
Abstract: In order to improve the depth control method of underwater vehicle and reduce the interference and influence of external factors such as currents, this paper analyzes the motion control and dynamics modeling for an independently designed cable patrol autonomous underwater vehicle(CP-AUV), and proposes a control algorithm for CP-AUV that integrates the Actor-Critic algorithm and PID control. The algorithm uses traditional PID control as the controller of the CP-AUV and uses the Actor-Critic algorithm to adjust the PID control parameters to make it better adapted to different disturbances and different adjustment depths. The effectiveness of the method is verified by simulation, and compared with the traditional PID control, the reinforcement learning-based PID control retains the simple and practical advantages of PID, but can also adjust the parameters according to the motion environment, which enhances the robustness of the algorithm and the accuracy of parameter tracking while improving the control accuracy.
|
|
14:00-18:00, Paper SaPP.35 | |
Developing a Comprehensive Model for the Prevention of Tension Neck Syndrome: A Focus on Musculoskeletal Disorder Prevention Strategies |
|
Sohani, Behnaz | University of Lincoln |
Adigun, Ifeoluwa Joshua | University of Lincoln |
Aliyu, Aliyu | University of Lincoln |
Rahmani, Amir | Castlet Ltd |
Goher, Khaled | University of Nottingham |
Keywords: Soft Robotics, Medical Robotics, Rehabilitation and Assistive Robotics
Abstract: Tension Neck Syndrome Musculoskeletal Disorder (TNS MSD) causes discomfort in the muscles around the neck and shoulder, especially of the elderly. It is one of the leading causes of early retirement. This paper provides initial results on the efficacy of a novel ergonomic-oriented neck support designed to mitigate and alleviate TNS MSD and focuses on the design of an adaptive neck supporter by integrating a soft actuator massager to help deliver a soothing massage. The design was carried out using simulations, prototyping, and measurements. The massager and adaptive neck supporter prototype were validated by Finite Element Analysis prior to fabrication to assess the feasibility of the design concept. After the massager prototype was fabricated, it was tested to validate the initial concept. Future work will be focused on fabricating the full-scale adaptive neck supporter prototype as well as upgrading and optimising the design concept.
|
| |