Keywords: Biomimetic and Bioinspired Robots, Aerial and Flying Robots, Performance Evaluation and Optimization
Abstract: Natural fliers glide and minimize wing articulation to conserve energy for endured and long range flights. Elucidating the underlying physiology of such capability could potentially address numerous challenging problems in flight engineering. However, primitive nature of the bioinspired research impedes such achievements, hence to bypass these limitations, this study introduces a bioinspired non-cooperative multiple objective optimization methodology based on a novel fusion of PARSEC, Nash strategy, and genetic algorithms to achieve insect-level aerodynamic efficiencies. The proposed technique is validated on a conventional airfoil as well as the wing cross-section of a desert locust (Schistocerca gregaria) at low Reynolds number, and we have recorded a 77% improvement in its gliding ratio.

Keywords: Computer Vision and Visual Servoing, Object Recognition
Abstract: Optical data is one of the crucial information resources for robotic platforms to sense and interact with the environment being employed. Obtained image quality is the main factor of having a successful application of sophisticated methods (e.g., object detection and recognition). In this paper, a method is proposed to improve the image quality by enhancing the lighting and denoising. The proposed method is based on a generative adversarial network (GAN) structure. It makes use of the attention model both to guide the enhancement process and to apply denoising simultaneously thanks to the step of adding noise on the input of discriminator networks. Detailed experimental and comparative results using real datasets were presented in order to underline the performance of the proposed method.

Keywords: Contact: Modeling, Sensing and Control, Manipulation Planning and Control, Grasping
Abstract: In this study, we present a novel method for ‘peg-in-hole’ assembly using a multi-joint three-fingered gripper. In previous studies, we have conducted peg-in-hole assembly based on spiral search method, not knowing the exact hole position. However, search time varies greatly depending on the error between the peg and the hole, it was difficult to estimate tact time of the assembly task. In this paper, in order to reduce the search time deviation of the hole, we propose a method that estimating the direction of the hole first before conducting the search motion. Direction of the hole is estimated by ‘kinesthetic sensing’ using the fingers grasping the peg. The feasibility of the proposed method has been verified through peg-in-hole assembly experiments using the three-fingered robot gripper.

Keywords: Biomimetic and Bioinspired Robots, Robotic Hands, Soft Robotics
Abstract: We present a novel approach for fabricating cable-driven robotic systems. Particularly, we show that a biomimetic finger featuring accurate bone geometry, ligament structures, and visco-elastic tendons can be synthesized as a single part using a mutli-material 3D printer. This fabrication method eliminates the need to engineer an interface between the rigid skeletal structure and elastic tendon system. The artificial muscles required to drive the printed tendons of the finger can also be printed in place. MuJoCo, a physics simulation engine which can be used to generate control strategies, is used to develop a model of the non-linear platform. A physical test bed is used to compare the simulation results to a printed prototype. This lays the groundwork for a new robotics design approach, where the fabrication and assembly is automated.

Keywords: Industrial Robots, Manipulation Planning and Control
Abstract: Recent developments in industrial robotics use real-time trajectory modification to improve throughput and safety in automatic processes. Online trajectory scaling is often used to this purpose. In this paper, we propose a feedback trajectory scaling approach that is able to recover from the delay introduced by the speed modulation and improves the path-following performance thank to an additional inner control loop. Simulation and experimental results on an industrial 6-degree-of-freedom robot show the effectiveness of the proposed approach compared to standard algorithms.

Keywords: Simultaneous Localization and Mapping (SLAM), Multi-Robot Systems
Abstract: Simultaneous localization and mapping (SLAM) system has an important role in providing an accurate and comprehensive solution for situational awareness in unknown environments. In order to maximize the situational awareness, the wider field of view is required. It is possible to achieve a wide field of view with an omnidirectional lense or multiple perspective cameras. However, calibration of such systems is sensitive and difficult. For this reason, we present a practical solution to a multi-camera SLAM system. The goal of this study is to obtain robust localization and mapping for multicamera setup without requiring pre-calibration of the camera system calibration. With this goal, we associate measurements from cameras with their relative poses and propose an iterative optimization method to refine the map, keyframe poses and relative poses between cameras simultaneously. We evaluated our method on a dataset which consists of three cameras with small overlapping regions, and on the KITTI odometry dataset which is set in stereo configuration. The experiments demonstrated that the proposed method provides not only a practical but also robust SLAM solution for multi-camera systems.

Keywords:
Abstract: This talk gives an overview of techniques we develop to enable robust, fast and flexible manipulation. We make extensive use of machine learning to enable robots to learn and improve their manipulation abilities. Machine learning is also key to account for uncertainty in perception and model prediction. The talk will present examples in which robots can learn to manipulate objects that deforms as a result of being manipulated, such as cutting fruits and vegetables. It will also present our most recent work on modeling human ability to acquire very fine manipulation skills for watchmaking.

Keywords:
Abstract: In this talk, I will present our recent studies on transferring human demonstrated tasks to robotic manipulators using automatic task and motion planning. The talk includes two parts. In the first part, I am going to review the difficulties faced by conventional and modern robotic manipulation systems. After that, I will present an opensource platform developed in our lab to solve these problems. I will primarily focus on the development workflow of the open-source platform and show the details of its task and sequence planning component. In the second part, I will present several past projects developed using the platform. Most of the contents are video-based and self-explanatory.

Keywords:
Abstract: Human-in-the-loop robot control is an effective means for robot programming that will help pave the road for a society populated with ubiquitous robots. As in human-human collaboration, in the human-robot shared control, both the robot and the human can adapt their control policies based on each other’s behavior. If managed properly, this co-evolving shared control framework may lead to higher task performance and effective robot behavior synthesis through synergy. However, currently, there is no general rule to ensure that a human-robot shared system evolves towards the sought synergy. Therefore, besides robot learning, the human sensorimotor learning induced by the human-robot shared control systems needs to be studied for a ‘robot-enabled society’. In this talk, a recent study in the aforementioned lines will be presented.

Keywords: Industrial Robots, Robot Surveillance and Security, Aerial and Flying Robots
Abstract: In this paper, we suggest an unmanned aerial vehicle (UAV) system on industrial chimneys using UAV's high mobility. The UAV follows along the waypoints given by operators and recognizes the front obstacles by using the stereo camera and re-plans the path. By applying a chimney temperature state (fire detection) classifier using Support Vector Machine (SVM), we solved the issue of a lack of data to detect. The SVM classifier is trained by small-scale experimental datasets. To apply different scale data for the SVM classifier, we extracted vector with Histogram of Oriented Gradients(HOG) and checked the appropriate scale by estimating similarity (Wasserstein distance) of varying scale reference images. The autonomous flying tests were conducted at the industrial environment and the feasibility of the chimney inspection algorithm has been done with lab-scale experiments.

Keywords: Medical Robotics and Computer-Integrated Surgery, Modular Robots, Actuation and Actuators
Abstract: The continuum robot systems for transoral surgery to alleviate the difficulties surgeons encounter in clinical cases have been developed worldwide. Especially in a tongue root cancer case, a surgeon has to reverse the patient’s tongue for the lesion to be exposed and approach long surgical tools to the lesion to perform resection in the patient’s mouth, which is very difficult and time-consuming due to the limitation of the existing instruments. Therefore, in this paper, the authors proposed design of a two-section continuum robot for transoral tongue cancer surgery. A reference path for transoral robotic surgery in tongue root cancer case is defined, a design of a twosection continuum robot is proposed, which can closely follow the reference path with 3 mm wire control, and the design is validated by computer simulation. Moreover, because of the bending and stiffness characteristics of the continuum robot, the designed robot is thought to be able to access the lesion in a patient’s mouth with rigidity capable of performing resection. The validation of the design though additional design feasibility, in-vitro, and in-vivo experiment should be followed.

Keywords: Medical Robotics and Computer-Integrated Surgery, Motion Planning and Obstacle Avoidance, Manipulation Planning and Control
Abstract: This paper proposed an optimization-based algorithm for underactuated surgical robots to perform auto-suturing, in which the needle needs 6 DOFs to fully define its orientation and position. A four DOF robot arm was used to illustrate the optimization approach. Both the algorithm and benchtop tests are presented.

Keywords: Medical Robotics and Computer-Integrated Surgery, Performance Evaluation and Optimization, Robotic Hands
Abstract: In recent years, people have been forced to stay in sitting postures for a long period of time in many situations. These prolonged sitting position causes blood flow stagnation in lower limbs, and lead to thrombus. It is said that pulmonary thromboembolism is caused by this thrombus circulation. To overcome these problems, a method of promoting blood flow using the muscle pumping action by repeatedly contracting and relaxing by sending electrical stimulation to the lower limb muscle has been proposed. In this research, we hypothesize that stimulating when the blood flow pumped out from the heart reaches the stimulation site can boost the blood circulation. Therefore, we propose a method to synchronize the stimulation timing with the timing blood flow reaches maximum value which is predicted from the measured heartbeat-timing.

Keywords: Modular Robots, Motion Planning and Obstacle Avoidance
Abstract: This paper proposes the approach of the locomotion of the Variable Topology Truss (VTT) system on sloped surfaces. To fully use the flexibility of the system, the locomotion algorithm on uneven surfaces is needed. As the initial research of locomotion on uneven surfaces, we discuss VTT’s locomotion on a single sloped surface. The slip condition of VTT on a sloped surface was analyzed by a static approach. If the required friction forces of two contact points are larger than the maximum static frictions, the system will slip on the surface. The improved locomotion algorithm of VTT, which includes moving on a sloped surface, will be developed in the future.

Keywords: Multisensor Data Fusion, Physical and Cognitive Human-Robot Interaction
Abstract: Similar to human-human social interaction, the process of inferring a user’s personality traits during human-robot interaction plays an important role. Robots need to be endowed with such capability in order to attract user engagement more. In this study, we present our on-going research on obtaining variable-length multimodal features and their fusion to enable social robots to infer human personality traits during face-to-face human-robot interaction. Multimodal nonverbal features, including head motion, face direction, body motion, voice pitch, voice energy, and Mel-frequency Cepstral Coefficient (MFCC), were extracted from videos and audios recorded during the interaction. The different combinations of multimodal features were verified, and their classification performance was compared.

Keywords: Manipulation Planning and Control, Computer Vision and Visual Servoing, Learning From Humans
Abstract: Robotic manipulation has been applied to a particular setup and a limited number of known objects. In order to cope with these limitations, robots need to be capable of manipulating novel objects. In this work, we proposed a computationally efficient 2-stage framework for planar pushing, allowing a robot to push novel objects to a specified pose with a minimum number of steps. We developed three modules: Coarse Action Predictor (CAP), Forward Dynamic Estimator (FDE), and Physical Property Estimator (PPE). CAP predicts a mixture of Gaussian distribution of actions. FPE learns the causality between action and successive object state. PPE based on Recurrent Neural Network predicts the physical center of mass (PCOM) from the visual center of mass (VCOM) and robot-object interaction. Our preliminary experiments show the promising results to meet the required capability of pushing novel objects.

Keywords: Aerial and Flying Robots, Performance Evaluation and Optimization
Abstract: To design an intuitive, easy to use drone controller, we developed a gesture-based drone controller using a Myo armband which requires only one hand to control a drone. A preliminary usability test showed that our controller is not yet comparable to an existing joystick or mobile app in performance, but is promising in providing enhanced situation awareness, consistency, and integrated functions. Future design directions are discussed with the results.

Keywords: Robotic Systems Architectures and Programming, Rehabilitation and Healthcare Robotics
Abstract: Children with autism spectrum disorder (ASD) have defects in eye contact and interaction with others. We propose a behavioral intervention-based interactive robot system for preschool children with ASD. The system can perform social interaction with the subject using the robot's appearance and LCD, and aims to quantify the psychological stress index of the subject through the user's head pose and biosignal analysis information.

Keywords: Physical and Cognitive Human-Robot Interaction
Abstract: Children with autism spectrum disorder (ASD) exhibit deficiencies in social communication and interaction, such as anxiety about performing joint attention or eye contact with others. In this paper, we propose a method to quantitatively analyze the psychological response from the anxiety-induced video stimulation by using the biometric sensors and analytic approaches. In order to analyze the psychological response, mean heart-rate and autonomic parasympathetic nervous system of heart rate variability analysis method is applied to the the photoplethysmography signal, and the histogram method is applied to the electrodermal activity sensor signals. A total of 10 college students were tested. In some cases, it was possible to classify states according to intensity of visual stimulus.

Keywords: Physical and Cognitive Human-Robot Interaction, Learning From Humans
Abstract: In these days, interests about robotics and deep learning are getting higher. Robots are developed for helping people on every part of human life, and the deep learning is essential for improving robot performance. Most robots use vision data and depth data to perform better in various fields, also in gesture recognition. However, using the vision sensor alone, it is hard to discriminate the similar gestures or situations. Misrecognizing the gestures can cause a big danger, especially in emergency situation. Thus, we tried to detect the emergency situation in real-time by using audio data. To judge the emergency in real-time, we propose a novel 3D convolutional neural network using Mel spectrogram for features. By this study, we found that detector with the proposed network can detect the specific situation such as an emergency, and finally, can classify the gestures with the sound data.

Keywords: Human-Robot Augmentation, Biomimetic and Bioinspired Robots
Abstract: Towards the control of a human augmentation exoskeleton for running assistance, it is essential to understand the human intention and muscle synergy during walking and running. For this purpose, an experimental study was conducted to investigate the characteristics of surface electromyography (sEMG) during the walking and running conditions. The GRF data and the muscle activities data by 16 muscles sEMG were collected simultaneously under 1.25 m/s -2.25 m/s walking and 2.5 m/s – 4.0 m/s running. Specifically, the two back muscle activities by sEMG were collected to find the back muscle function during walking and running. The MAV and EMG peak features of sEMG were calculated to investigate the change of muscle activities along with the speed changes. Based on our measurements and analysis, the walking and muscle activity pattern of the human could be changed with the different speeds during the walking condition. The muscle activity pattern for running shows a consistent pattern at different speeds.

Keywords: Robotic Systems Architectures and Programming, AI Reasoning Methods for Robotics, Learning From Humans
Abstract: Intelligence and dexterity of robot have been extended for decades, and robots like co-bot began to permeate to workspace of humans. Therefore, sharing of intelligence is required for both humans and robots. XAI(eXplainable Artificial Intelligence) is a good example of intelligence sharing from robots or AIs to humans. The proposed system in this paper is vice versa. We wish that intelligence of humans is shared to robots and AIs. Concretely, We discussed and analyzed the process of robot understanding furniture assembly instruction manual which is designed for human. Additionally, we proposed a system for AI to understand this. The assembly instructions are given to AI as 2D information, and 3D CAD files are also given as 3D information. We show our system that provides correspondence between 3D and 2D information using Faster R-CNN.

Keywords: Learning From Humans, Legged Robots, Dynamics and Control
Abstract: This paper proposes an inverse optimal control (IOC) framework which incorporates demonstrations with mixed qualities. The proposed method utilizes the benefits of sub-optimal demonstrations which can provide information about textit{what not to do} and supplies training data near states unvisited by optimal demonstrations. The main idea of the proposed method is to find the value function which satisfies the optimality condition over optimal demonstrations and violates it over sub-optimal demonstrations. We conduct experiments on three environments and empirically show that the proposed method outperforms the original IOC algorithm, which uses only optimal demonstrations.

Keywords: Medical Robotics and Computer-Integrated Surgery, Dynamics and Control, Micro/Nano Robots
Abstract: The medical micro-robotic applications are becoming more eminent by ever-flourishing possibilities presented by the robotics and biomedical communities in the past decade. Artificial and bio-hybrid designs started flooding the literature accompanied by fewer examples of closed-loop control of such systems until recently. The current focus is mainly on MRI systems and electromagnetic coil configurations as it has been long since they proliferated in mainstream medicine. In this paper, a closed-loop control approach to a rather rarely investigated robotic arrangement is proposed. A detailed mathematical model for the dynamics of an open kinematic chain and a single bacterium cell coupled with the help of a magnetic field is established, and a bilateral adaptive control is auspiciously fitted on top to maneuver the cell in a simulated environment.

Keywords: Robotic Hands, Grasping, Soft Robotics
Abstract: This paper presents a new type of a robotic palm based on a granular jamming mechanism to improve grasping performance. The granular jamming principle is adopted to alter the shape and stiffness of the robotic palm by controlling a transition between a solid-state and a fluid-state of a granular material used. The robotic palm incorporates a specifically designed granular chamber that is optimized for dealing with large volume change. The control system is also developed for the proposed granular jamming mechanism to be electrically operated without any pneumatic pump. In addition, the stiffness of the palm can be precisely regulated by the feedback control of the negative pressure applied to the granular chamber. We evaluate the shape-adaptability of the robotic palm for various objects. As a result, the robotic palm could accommodate the various shapes of the testing objects by conformably altering its shape during contact. Moreover, the stiffness-controllability is also investigated for the three different sizes of granular materials. The stiffness increases up to 30 fold under fully jammed state for the small size of the grain. Finally, we evaluate the grasping performance of the robotic palm with a commercially available robot hand. 1.7 times higher grasping force was attained with the conformably deformed and stiffened surface, compared to the flat skin of the rigid palm. Therefore, the stiffness-controlled robotic palm can improve grasping performance with the enhanced shape-adaptability and stiffness-controllability.

Keywords: Mechanism and Design, Underwater Robotics, Biomimetic and Bioinspired Robots
Abstract: In this paper, we proposed an underwater walking mechanism for the underwater amphibious robot that uses one degree of freedom (DOF) actuators. For this walking mechanism, we developed a unique spring-hinge type paddle that enables the amphibious robot to walk on the seabed. We proposed a simplified 2-D model of the robot. Then, we analyzed rough-terrain capability of this mechanism by using following terms: the paddle-length, the hinge-length, the distance to the obstacle, and the maximum sweep angle. We developed an experimental robot for a feasibility test of the effectiveness of proposed walking mechanism, and we performed ground and water tank experiments with this robot. As a result, we confirmed that the robot walked stably with the proposed mechanism.

Keywords: Physical and Cognitive Human-Robot Interaction, Social and Socially Assistive Robotics, Behavior-Based Systems
Abstract: With the recent expansion of the geriatric population, caregivers and therapists have become increasingly important for improving the quality of life of elderly persons; therefore, having properly trained professionals in these fields is critical. To ensure accurate and efficient elderly care training, this study focuses on the development of an automated care training assessment method based on fuzzy-logic for calculating care training quantitative score (CTQS) for improving care skill of caregivers or therapists. Five elderly care experts and three trainees performed the exercise in flexion and extension using the elbow part of care training assistant robot (CaTARo). We obtained quantitative data from the CaTARo, such as elbow external angle, angular velocity, torque, pressure, for all participants and evaluated performance by combining four key parameters using fuzzy-logic method. As a result, we found that the result of trainees had showed a significant difference between pre-test and post-test, and the CTQS of trainees in post-test could be improved by training with CaTARo than that of pre-test.

Keywords: Physical and Cognitive Human-Robot Interaction, Social and Socially Assistive Robotics
Abstract: In order for social robots to be usably accepted by users, they need to exhibit social interaction skills in a certain degree when they encounter human even in unexpected environments, like human can do. Evaluation of Social Interaction (ESI) is one of well-defined methods to evaluate human social intelligence with 7 classes and 27 items of social interaction skills. Among them, the physically supporting social interaction(PSSI) class with 5 items are dominant to evaluate social robot intelligence, according to the occupation therapist evaluation. In this paper, focusing on the 3 items in the PSSI class, we propose an evaluation method of social robot intelligence in terms of social interactive motion, apply it to a social robot called Wendy and compare it with human social intelligence.

Keywords: Robotic Systems Architectures and Programming, Social and Socially Assistive Robotics, Wheeled Mobile Robots
Abstract: Visual perception is a fundamental capability necessary for intelligent mobile robots to interact properly and safely with the humans in the real-world. Recently, the world has seen revolutionary advances in deep learning has led to some incredible breakthroughs in vision technology. However, research integrating diverse visual perception methods into robotic systems is still in its infancy and lacks validation in real-world scenarios. In this paper, we present a visual perception framework for an intelligent mobile robot. Based on the robot operating system middleware, our framework integrates a broad set of advanced algorithms capable of recognising people, objects and human poses, as well as describing observed scenes. In several challenge scenarios of international robotics competitions using two mobile service robots, the performance and acceptability of the proposed framework are evaluated.

Keywords: Telerobotics, Human-Robot Augmentation, Computer Vision and Visual Servoing
Abstract: In this paper the novel teleoperation method "Hands-Free" is presented. Hands-Free is a vision-based augmented reality system that allows users to teleoperate a robot end-effector with their hands in real time. The system leverages OpenPose neural network to detect the human operator hand in a given workspace, achieving an average inference time of 0.15 s. The user index position is extracted from the image and converted in real world coordinates to move the robot end-effector in a different workspace.

The user hand skeleton is visualized in real-time moving in the actual robot workspace, allowing the user to teleoperate the robot intuitively, regardless of the differences between the user workspace and the robot workspace.

Since a set of calibration procedures is involved to convert the index position to the robot end-effector position, we designed three experiments to determine the different errors introduced by conversion. A detailed explanation of the mathematical principles adoptes in the work is provided in the paper.

Finally, the proposed system has been developed using ROS and is publicly available at the following GitHub repository: https://github.com/Krissy93/hands-free-project.

Keywords: Robotic Systems Architectures and Programming
Abstract: Robotic systems are complex and commonly require experts to program the motions and interactions between all the different components. Operators with programming skills are usually needed to make the robot perform a new task or even to apply small changes in its current behavior. For this reason many tools have been developed to ease the programming of robotic systems. Online programming methods rely on the use of the robot in order to move it to the desired configurations. On the other hand, simulation-based methods enable the offline teaching of the needed program without involving the actual hardware setup. Virtual Reality (VR) allows the user to program a robot safely and effortlessly, without the need to move the real manipulator. However, online programming methods are needed for on-site adjustments, but a common interface between these two methods is usually not available. In this work we propose a VR-based framework for programming robotic tasks. The system architecture deployed allows the integration of the defined programs into existing tools for online teaching and execution on the real hardware. The proposed virtual environment enables the intuitive definition of the entire task workflow, without the need to involve the real setup. The bilateral communication between this component and the robotic hardware allows the user to introduce changes in the virtual environment, as well into the real system. In this way, they can both be updated with the latest changes and used in a interchangeable way, exploiting the advantages of both methods in a flexible manner.

Keywords: Rehabilitation and Healthcare Robotics, Soft Robotics
Abstract: Highly intensive and repetitive training is crucial to enhance a hand function for stroke survivors. Continuous rehabilitation after hospital discharge has to be implemented for better hand function improvements. Many robotic devices have been developed for home rehabilitation and daily assistance purposes, which the device has to be portable, comfortable, and easy-to-use. However, active devices tend to have a sizeable actuating unit to extend fingers with flexor hypertonia, and passive devices cannot assist flexion, which results in low grip strength. In this study, an elastic hand orthosis was developed to enable stroke survivors to pinch and stabilize ordinary objects. Since a high force was needed to extend fingers actively, a 3D printed structure was placed at the palm to extend the finger passively. The twisted string actuation(TSA) was adopted to replicate pinching with high strength by bending the structure. The rigidity of the structure was optimized based on the measured finger joint stiffness. The functional ability of the orthosis was tested with one stroke survivor. The stroke survivor can passively extend the finger and pinch various objects with a sufficiently large force. The proposed orthosis is expected to help the patients to conduct activities of daily living(ADL).

Keywords: Soft Robotics, Actuation and Actuators, Biomimetic and Bioinspired Robots
Abstract: This work introduces the concept, design and manufacturing of a pre-twisting based inflatable torsional actuator for robotic application. This actuator is composed of six individually inflatable chamber and each three chambers are connected with each other. As a result, the developed actuator has maximum twisting angle as 45.2 degree on clockwise direction and 43.1 degree on counterclockwise direction.

Keywords: Soft Robotics, Actuation and Actuators, Biomimetic and Bioinspired Robots
Abstract: This work introduces the concept, design and manufacturing of an artificial muscle driven by positive and negative pressure linear actuator. This linear actuator is capable of producing large forces while being lightweight and efficient. A load cell together with a linear sensor were installed onto the end of the actuator to verify the force and the linear displacement. Based on the actuator dimensions, it is able to lift loads of over 110 times its own weight. This sort of artificial muscle could be useful in lightweight robotic arms.

Keywords: Soft Robotics, Actuation and Actuators, Mechanism and Design
Abstract: In the field of soft robotics, the various soft actuators have been suggested such as fluidic elastomer actuator (FEA), electroactive polymer (EAP) actuator, and shape memory alloy (SMA). However, each soft actuator has its own limitations such as the existence of external equipment, controllability, and limited output force. To solve these issues, designing a new soft actuator by combining different soft actuators has been attempted previously. In this paper, the novel soft actuator which is designed by combining PVC-gel and hydraulically amplified self-healing electrostatic actuator (HASEL) is presented. The proposed actuator shows a high output force (3.35 N) at the low operating voltage (4kV), which has been the goal of the dielectric EAP actuator for many years.

Keywords: Soft Robotics, Actuation and Actuators
Abstract: Layer Jamming Brake has a strong tensile resistance as the vacuum is increased, but at low vacuum pressures the tensile resistance is very weak. To overcome that problem, this paper introduces a layer jamming brake that combined with an Inflatable Pouch can exert more force by combining the positive pressure from the Inflatable Pouch with the negative pressure from the Layer Jamming Brake. The actuator can lift a load of 10 kg with 30 kPa of positive pressure and 60 kPa of negative pressure.

Keywords: Soft Robotics, Biomimetic and Bioinspired Robots
Abstract: This paper describes the design and fabrication of a low-cost crawling robot using origami folding techniques. Robotics today have become easily accessible to many educators, students and makers. Largely, these robots are made from rigid structures fabricated with the use of casting, welding, machining, or injection-molding. The proposed robot is a low-cost, untethered origami robot that can be fabricated from scratch using personal fabrication machines, such as laser cutters and 3D printers. The robot is comprised of materials and hardware that are readily available at craft stores. The body of this origami robot is mainly composed of Vellum, a durable and flexible paper material that through an etched and altered Miura folding pattern becomes volumetric. This robot is named Mollusca, inspired by the wave-like locomotion of Mollusks.

Keywords: Soft Robotics, Mechanism and Design
Abstract: A wearable robot is a robot that is worn by a user and can always interact with the robot regardless of the user's location. On the other hand, most wearable robots have a limited mounting position and do not move from the mounted position in most cases. However, in order to provide appropriate support depending on the user's situation, It is necessary that the mounting position of the robot can be changed according to the situation. Therefore, in this study, we propose a robot mechanism that moves the user's arm up and down in order to interact with the user in various ways. By rotating the whole body, the movement become possible. In the movement experiment, it was proved that the proposed mechanism could move on the average thickness of human arm.

Keywords: Soft Robotics, Search and Rescue Robotics
Abstract: Soft robots that locomote through growth at the tip have the potential to explore rough, unstructured environments and constrained space due to their complaint nature. However, the continuous release of new material from the robot tip has makes attaching sensors, cameras, and other equipment to the robot tip a difficult challenge. Many studies to date have proposed a light and durable cap that encases the tip of the robot and serves as a mount for camera and other sensor equipment. In this paper, we propose a method to keep the sensor or camera on the tip without using a rigid cap that limits the inherent compliance of the soft robot. The proposed method makes use of the inside structure of the growing robot to access the tip, while independently controlling the equipment We present the initial testing results for a work in progress contribution.

Keywords: Aerial and Flying Robots, Dynamics and Control, Mechanism and Design
Abstract: The RVM (Robot-based Vibration Suppression Modules) is proposed for the manipulation and transport of a large flexible object. Since the RVM is easily attachable/detachable to the object, this RVM allows distributing over the manipulated object so that it is scalable to the object size. The composition of the system is partly motivated by the MAGMaS (Multiple Aerial-Ground Manipulator System) [1]-[3], however, since the quadrotor usage is mechanically too complicated and its design is not optimized for manipulation, thus we overcome these limitations using distributed RVMs. For this, we first provide an optimized RVM design with the minimum number of rotors, so that the feasible thrust force is maximized while it minimizes undesirable wrench and its own weight. Then, we elucidate a controllability condition and an optimal placement with multiple distributed RVMs. Experiments are performed to demonstrate the effectiveness of the proposed theory.

Keywords: Dynamics and Control, Manipulation Planning and Control, Mechanism and Design
Abstract: We have developed the Deformable Shadow Casting Robot (DeSCaRo) that has 39 degrees of freedom. The purpose of this robot is to create shadow art. Shadow art is one of the fields of art attracting attention from people in recent years. Thus, DeSCaRo can generate shadow art and give impact to people. In this study, DeSCaRo is used as an information providing device in the Intelligent Space. It is described whether the presentation of information about shadow pictures is effective.

Keywords: Mechanism and Design, Aerial and Flying Robots, Contact: Modeling, Sensing and Control
Abstract: Unmanned Aerial Vehicles (UAVs) have been taking an important place in our daily lives. They have been used for several different usage purposes, e.g., mapping, surveillance, inspection, and many others. One of the widely-used areas of UAVs is visual inspection since they can gather information from areas beyond human-reach or from environments hostile to humans. Their usage for the inspection purpose is mostly limited to non-contact type sensors such as optical cameras and/or laser sensors. Lately, wall-climbing UAVs have been proposed to gather data using contact-type sensors. However, the major drawback is that they can solely be used for planar surfaces. In this paper, preliminary design studies of a compact robotic mechanism to allow UAVs to land on any shape of surfaces is presented. This mechanism is intended to be light-weight, modular, expandable, self-balanced, attachable to most of the existing UAVs.

Keywords: Mechanism and Design, Modular Robots
Abstract: Spiral Zipper is a prismatic actuator which drives a flexible band into a helix to form a length-changeable tube. The band can be extended or retracted by applying a torque to the band tube; in this case, torque is transmitted using a friction drive wheel. However, driving with friction causes ear on the driving components, which decreases the friction force and results in increased slip. In this paper, an improved mechanical design using a multi-armed friction drive wheel, the diameter of which can be adjustable, for adjusting the friction force is presented. The friction adjustment ability of the multiarmed friction drive wheel is validated by experiment, and it is shown to counteract the negative effects of wear.

Keywords: Mechanism and Design, Wheeled Mobile Robots
Abstract: Robot technology is already proving high efficiency in the industrial field. These robot technologies are also applicable to smart farms. In this paper, we are going to present about development of AGV for smart farm among robot technologies. The AGV is one of the most efficient robot technologies that can be applied to agricultural field. The AGV can be used in smart farm to assist with the transport of crops mainly and it needs functions of hot water pipe driving, autonomous driving, worker following, call by operator and control systems. The driving on the hot water pipe is done by using that as a rail, which requires special wheels. The autonomous driving requires magnetic line recognition and RFID. The ultrasonic sensor and IR sensor are used for worker following. The LoRA (Long Range) is used for AGV call and sending data to control system. The control system is made by Linux and it has LoRA, LTE to communicate with AGV and workers. The proposed AGV allows for the attachment of equipment such as lifts and conveyors to be used for a variety of tasks.

Keywords: Mechanism and Design, Robotic Hands
Abstract: In this paper, we propose a wearable robot arm in consideration of weight and usability. Based on the features of the existing wearable robot arms, we focused on the issues of weight and usability. When a human is working physically with his hands, behavior of the hands can be divided into two types. One is that shoulder and elbow joints move before doing the task with the hands. The other is that wrist joints move during working. We found that these features can be applied into the wearable robot arms, and we proposed ”Hybrid Actuation System (HAS)” in combination of two types of joints. In this paper, we have improved the passive joint based on the maximum holding torque.

Keywords: Robotic Hands, Mechanism and Design, Grasping
Abstract: This work presents one linkage-driven robotic hand which has three fingers. Each finger is constructed by stacking one five–bar mechanism over one double parallelogram. This finger mechanism can be used to achieve the parallel and adaptive power grasping. Through adding a rotational actuator to the base of each finger, many types of grasping modes can be achieved by using this robotic hand. Development and experimental evaluation of this three-finger hand are reported in this work.

Keywords: Telerobotics, Industrial Robots, Haptics
Abstract: Currently, automated welding systems for mass production are used at industrial filed, but automatic welding cannot be used to manufacture small productions or repair machines and structures in a hazard environment. In an environment where automatic welding is not available, the operator must complete the task by controlling the robot remotely. However, it is not an easy task to do with teleoperation while only looking at some camera screens on the robot endeffector. Therefore, we designed a process for specific tasks to reduce the task’s difficulty so that it is easier for tele-operator to do the work. Based on this process, data retargeting and constraints of the master device can assist the operator in completing a specific task.

Keywords: Computer Vision and Visual Servoing
Abstract: In recent years, the use of 3DCG and 3D avatar has increased with the development of image processing technology in various digital contents. Motions in 3D avatars has become very complicated, and it takes a very long time to create them since they are made manually in most cases. In this study, estimating emotions from sentences and generating 3D avatar motions automatically from the estimated emotion information are challenged. As the first stage of the research, estimation of emotion from sentences is investigated in this paper.

Keywords: Medical Robotics and Computer-Integrated Surgery, Computer Vision and Visual Servoing
Abstract: Vitreoretinal surgery represents a medical procedure whose tasks exist on the edge of human capability, given the relative scale of possible human hand movements, and afflicted areas on the retina. Performing vitreoretinal surgery using teleoperated robots, as opposed to manually, can potentially greatly reduce the difficulty level and required skill, and minimize the risk for the patient. In this paper, a potential safety concern with existing microsurgery robots will be shown. A computer vision-based system will be proposed which can address this safety concern, as well as provide automatic microscope positioning. The CV system will segment the iris and trocars in real time using a convolutional neural network (CNN). The trocar positions are used to place each slave robot’s remote centre of motion (RCM), minimizing scleral strain. The iris position is used to automatically align the microscope with the iris centre.

Keywords: Object Recognition, Computer Vision and Visual Servoing
Abstract: Object detection has received much attention in recent years. However, relatively little research has been conducted on orientation estimations. In addition, scant research has focused on the orientation estimations of indoor objects that robots must recognize when performing their tasks. In this paper, we present a new method that predicts 2D bounding boxes and orientations simultaneously in real time for indoor scenes. We evaluate our method on the SUN RGB-D dataset, and the results show that our approach is very capable of making predictions for directional objects. With ablation experiments, we demonstrated that symmetric objects degrade training performance outcomes. Moreover, for an object with an identical shape in four directions, when we limit the orientation scope, the performance improves.

Keywords: Rehabilitation and Healthcare Robotics, Computer Vision and Visual Servoing, Object Recognition
Abstract: To control a rehabilitation robot system, the pose of the user is required in real-time. This study suggests a low-cost real-time optical motion capture system to provide motion data for a cable-driven gait rehabilitation robot. Band-shaped markers are designed in this study, and they have specific patterns that can be recognized with webcams. Band markers can be put around body segments such as shank, thigh, and arms, and used for identifying location and orientation of them. When two or more webcams recognize the same patterns of the band marker, 3D locations of specific contour points can be obtained by triangulation and based on them, the pose of the segments can be estimated. The rehabilitation robot in this study requires the absolute location of knees and orientation of thighs. Therefore, the band markers were put around thighs and used to control the robot system.

Keywords: Social and Socially Assistive Robotics, Object Recognition, Aerial and Flying Robots
Abstract: In recent years, hornets appear in urban areas and it is regarded as a problem since the hornets attack human. It is also dangerous to destroy hornet nests since hornets have poison and they attack the people who approach to the nest. In this research, finding hornet’s nests by tracking hornets by combining computer vision and aerial drone.

Keywords: Object Recognition
Abstract: With the spread of smartphones, the number of accidents related to walking while looking on a smartphone is increasing. In this research, we propose a system to notify the danger of collision with a lightweight mobile device in order to prevent a collision accident with a pedestrian. While the user is using his smartphone, the proposed application runs in background and monitors surroundings via the smartphone camera. The warning notification is delivered by displaying warning images on the screen or by the notification sound. Experiments were performed to verify the effects of the proposed system.

Keywords: Simultaneous Localization and Mapping (SLAM), Wheeled Mobile Robots
Abstract: Ultra-wideband(UWB)-based simultaneous localization and mapping(SLAM) researches has been widely studied to replace situations where vision-based sensors or global positioning system(GPS) signals are not available. In this paper, we propose an extended Kalman filter(EKF) SLAM framework for estimating position and orientation of a robot and UWB anchors. The framework consists of three main processes: initialization of position of a UWB anchor, pre-integration of an inertial measurement unit(IMU) sensor measurement, and the EKF framework. A state vector of the system augmented with bias of a gyroscope, scale factor of a gyroscope, and bias of an accelerometer is estimated from the UWB-IMU tightly-coupled EKF framework. With the performance test for positioning of a robot when UWB anchors are fixed or moved, it is verified that position root-mean-square-error(RMSE) is less than 20 cm.

Keywords: Social and Socially Assistive Robotics
Abstract: In this work-in-progress paper, we first propose that intelligence functions can enable social robots to communicate naturally in Korean and then work to design Key Intelligence Functions (KIFs), which are high-level functions that can be used across various contexts through the use of a basic and generic language comprehension intelligence. Chatbots are inherently limited, in that they are only able to generate a simple answer to domain-specific questions. As such, our Key Intelligence Functions work to further improve upon the functionality of chatbots. In order to test various representative examples of KIFs, we proposed and implemented a number of recognizers: an affirmative-negative recognizer, a choice recognizer for multiple-choice questions, a number extractor from a sentence, and a robot behavior command recognizer. We then recruited 52 undergraduate students to write interaction scenarios using the KIFs to evaluate their effectiveness. The scores obtained in this user study demonstrated the potential of our approach.

Keywords: AI Reasoning Methods for Robotics, Soft Robotics, Force and Tactile Sensing
Abstract: In this paper, we compared neural network models available for piezoelectric slip sensors that detect slip over large area. Slip detection sensors are useful when manipulating objects through robotic hands. When manipulating an object, you can apply proper gripping force if you know the object is slipping or holding well on the robot hand. For this purpose, we developed a large area slip detection sensor that can cover the robot hands like skin. The developed sensor measures vibration in entire area using one tactile cell. Various artificial neural networks are used to detect slips occurring on the surface by using simple vibration signals. Slip data is difficult to produce large amounts of data for learning. To solve this problem, the artificial neural network model, which is well trained with a small amount of data, was searched. The same data was learned and compared using basic ANN, CNN, RNN, and LSTM. As a result, we confirmed that CNN is the neural network model with the highest accuracy in our sensor.

Keywords: Dynamics and Control
Abstract: A study on fractional-order (FO) sliding mode Controller with sliding perturbation observer (SPO) for trajectory tracking is researched. FO-SMCSPO (FO sliding mode control with SPO) and FO-TSMCSPO (FO terminal sliding mode control with SPO) are proposed. FO sliding mode changes the integral-order derivative in sliding surface to fractional-order one. It results in an excellent tracking performance in starting with fast convergence speed, which shows a good performance in the part of transient response. However, a slow reaching exists in the terminal of reaching equilibrium. In order to compensate this drawback, terminal sliding mode (TSM) is applied instead of the linear sliding mode (LSM). The combination of fractional-order sliding surface with terminal sliding techniques is designed to have a fast convergence entirely. Sliding perturbation observer (SPO) is adopted to reduce the chattering problem. Comparative simulations demonstrated the desired performance.

Keywords: Force and Tactile Sensing, Modeling, Identification, Calibration
Abstract: One of the issues in eye surgery is absent of tool-to-tissue interaction force. Such problem may cause additional damage to human eyeball. Therefore, the main challenge for eye surgery instruments is to achieve force sensing capability. This paper presents the conceptual design of a single axis force sensing instrument for retinal microsurgery. The development of this force sensing instrument is based on slot module. The slot structure is created on the tubular shaft to make the discrepancy between the centerline and the neural axis of the tube, thus a strain can be induced when the instrument tends to bend. Euler-Bernoulli beam theory is used to model the force sensing module. To demonstrate the force sensing capability, the corresponding 3D model was built and analyzed with finite element analysis (FEA) method. The results indicate that Euler-Bernoulli beam model fits well with this slot-based force sensing module and also verify the force sensing capability of this eye surgery instrument.

Keywords: Foundations of Sensing and Estimation, Robotics in Hazardous Applications
Abstract: Searching and estimating source information such as the location and release rate, called as source term, has many applications across environmental, medical and security domains. For autonomous source search and estimation in a turbulent environment, this paper presents an information-theoretic search approach which extends the existing Infotaxis method by introducing the Gaussian mixture model (GMM). The use of the GMM for potential source locations, obtained from the particle filter, in determining candidates of the next sampling position better facilitates the balance between exploitation and exploration properties of Infotaxis, resulting in the better search and estimation performance. Numerical simulation results in various environmental setups show the superior performance of the proposed approach compared with original Infotaxis.

Keywords: Intelligent Robotic Vehicles, Range, Sonar, GPS and Inertial Sensing
Abstract: Road Markers provide important information to be followed by the autonomous vehicle for navigation. Therefore, the road markers detection is an important foundation in the development of intelligent vehicles. The camera and LiDAR are commonly used sensors for road marker extraction, but the LiDAR sensor is more robust over camera in the light changing situation. The LiDAR provides higher intensity value in the road marker area as compared to the other road surface. However, due to physical characteristics, the intensity value is lost over a long distance. So, Road Markers detection system can be made efficient by restoring long-distance intensity values. Some studies proposed a method to calibrate intensity loss, but it is hard to completely make a model of calibration. In the normal condition of road situation, the histogram of certain range point cloud have constant shape which can help to improve the road marker detection accuracy. Therefore, in this paper we propose the robust pre-processing method to extract long-range road marker surface points without intensity calibration.

Keywords: Manipulation Planning and Control, Multi-Robot Systems, Motion Planning and Obstacle Avoidance
Abstract: This study presents object transportation using two 4-degrees of freedom (4-DOF) manipulators. In object transportation, constraints exist wherein the end-effectors of each arm must maintain their relative position. Given that a 4-DOF manipulator exhibits a limited workspace that satisfies the constraints, it is necessary to identify the area within which the manipulators can move an object. Object transportation space that satisfies the constraints can be derived via suppressing the rotational motion of two end-effectors when two manipulators are driven in the same posture. In the case of 4-DOF manipulators used in the study, the object transportation space is divided into a plane and a curved surface as well as a cross line where the two spaces intersect. In order to move in each space, conditions must be satisfied for each space. Therefore, if the conditions are satisfied, object transportation can be performed across two spaces.

Keywords: Multisensor Data Fusion, World Modelling, Behavior-Based Systems
Abstract: In recent years, the various research on IoT(Internet of Things) has been actively conducted. Intelligent Space is a comprehensive spatial system which includes IoT. One of the functions in Intelligent Space is memory recollection which is composed of detecting and tracking of people and recording the activities of the people in the space. In this paper, based on the recorded activities of people, abnormality of people's behavior is defined and its detecting algorithm is proposed.

Keywords: Dynamics and Control, Telerobotics, Industrial Robots
Abstract: We present experimental validation to verify passivity-based control framework of large-scale dual-arm telerobotic system on flexible support. Our control framework utilized passive decomposition approach to achieve two different control objective; 1) user-commanded trajectory tracking of robot arm, and 2) vibration suppression of flexible support. We performed experiment for validation with 28-DOF large-scale telerobotic system for height-operation tasks in nuclear power plants.