Keywords: Control Application in Mechatronics, Identification and Estimation in Mechatronics
Abstract: Contouring motion control is a critical issue for multi-axis machining in modern industry. This paper presents a three-dimensional global task coordinate frame (3D-GTCF) based contouring control method. Compared to the existed GTCF, which can only be used for 2-D contouring tasks, the proposed task coordinate frame is designed for the 3-D curve. By integrating a numerical method to calculate the contouring error and the corresponding transformation matrix, the proposed approach is able to tackle complex free-form contouring following tasks without their explicit analytical expressions. It’s noted that the construction of the proposed 3-D task coordinates is only determined by the geometric shape of the desired contour, and is not relevant to the desired motion on the contour. Then a direct/indirect adaptive robust controller is adopted to realize contour control in the task space. Simulation results under different contouring tasks validate the effectiveness of the proposed 3D-GTCF based contour control approach.

Keywords: Learning and Neural Control in Mechatronics, Robot Dynamics and Control, Neural and Fuzzy Control in Mechatronics
Abstract: In this work, a design of novel 4-DOF parallel robot is first introduced. Compared to the existing parallel robots, the advantages of this robot are on large translational workspace, large driving force, which extremely suitable for pick and place operating scenario. However, due to the complicated closed-chain structure of the parallel robot, the vital dynamic control solution for this scenario is restricted by the fact that the dynamic model can not meet the real-time calculation requirements. In order to introduce the prior knowledge of robot dynamics into the control system, an online learning control of neural networks based on off-line pre-training by the local dynamic model is proposed which has high computational efficiency and strong robustness. The effectiveness of the proposed control algorithm is verified by two simulation scenarios.

Keywords: Control Application in Mechatronics, Identification and Estimation in Mechatronics
Abstract: An adaptive trajectory tracking controller is proposed for linear systems with input delay, unknown plant parameters and constant disturbance. The controller employs predictor feedback to compensate the effect of input delay and the projection-type adaptation laws are designed to tackle parametric uncertainties. By the stability analysis with a Lyapunov function in logarithmic form, the controller guarantees the convergence of the tracking error when the design parameters such as adaptation gains meet certain conditions. An application to the speed control of the rotary motor is conducted. The simulation and experiment results demonstrate the effectiveness of the proposed controller.

Keywords: Control Application in Mechatronics
Abstract: To achieve ideal tracking motion performance for industrial motion stages, cascaded iterative learning control (CILC) strategy is deeply explored and investigated in this paper. CILC is essentially based on executing traditional ILC for several rounds. Specifically, CILC executes a new round of ILC by registering and clearing the feedforward compensation signal. The strategy and accuracy of CILC have been presented and analyzed. The theoretical analysis shows that the repetitive error can be eliminated completely regardless of the imperfection of filters. Comparative experimental investigations are executed on a linear motor-driven stage. Experimental results show that CILC method can achieve significant accuracy improvement compared with standard ILC method, although ILC performs rather well in industry due to its practically excellent control accuracy. CILC has good application potential for industrial mechatronic motion systems.

Keywords: Sensors and Sensing Systems, Sensor Integration, Data Fusion, Control Application in Mechatronics
Abstract: With no need for position sensors, sensorless control makes permanent magnet synchronous motor (PMSM) more compact, more economical, and easier to manufacture. Due to the advantages of no model limitation and low noise level, flux estimator is a promising method of sensorless control, while the drift caused by pure integrator limits its performance so hitherto it is not used in precision motor control. In this paper, a new flux estimator with drift elimination capability is designed to achieve high-accuracy flux estimation. The proposed drift eliminator using the orthogonality of α and β-axis can estimate the drift in variable situations with high dynamic performance. Theoretical interpretation is given while comparative simulations and experiments consistently show that the proposed estimator can eliminate the flux drift phenomenon efficiently. Sensorless control experiments are also conducted on a permanent magnet linear synchronous motor (PMLSM), where the position estimation error (PEE) can be lower than 50 µm. The results certify that the proposed estimator can be used in high-precision sensorless control, both for constant speed and variable speed trajectories.

Keywords: Mobile Robots, Robot Dynamics and Control
Abstract: In this paper, a novel omnidirectional mobile robot built with decoupled powered caster wheels is studied.The kinematic and dynamic models of this robot are established and linearized, from one-quarter body to the whole robot. The decentralized relay test procedure is proposed to excite the robot under bounded motions, so that the dynamic parameters are identified on a variety of moving paths. The effectiveness of the modeling and parameter identification methods are verified by simulation.

Keywords: Control Application in Mechatronics, Actuators
Abstract: PVC gels represent a novel type of electroactive polymer actuators with a number of appealing characteristics, including low cost, high compliance, large strain, high stress, fast reaction, and thermal stability. Despite their vast potential in a variety of applications, modeling and control of nonlinear dynamics of PVC actuators has received little attention. In this paper, we first present a data-driven approach to modeling nonlinear dynamics of PVC gel actuators. A Hammerstein model, consisting of a nonlinear module cascaded with linear dynamics, is proposed to capture the pronounced dependence of the voltage input -- displacement output frequency response on the input amplitude and bias. A control scheme is then designed based on the model, where an inverse compensator for the nonlinear element is combined with a PID feedback controller. A disturbance observer is further introduced for the estimate and rejection of influence of imperfect inverse compensation and model uncertainties. Experimental results are presented to support the efficacy of the proposed modeling and control approach. In particular, for a number of reference trajectories, the proposed control scheme results in over 80% reduction of tracking error in comparison with a well-tuned PID controller.

Keywords: Actuators, Biomechatronics, Modeling and Design of Mechatonic Systems
Abstract: A general system modeling and simulation technique is described for discrete muscle-like actuators. These actuators are composed of smaller, modular contractile units connected together, each of which may be controlled individually. Units are activated (recruited) in a manner similar to muscle cell recruitment, whereby the contractions of each individual smaller unit sum to produce a contraction of the entire actuator. Each unit has a lumped-parameter dynamic model, and a mathematical technique is described for automatically constructing a state-space realization of the entire actuator from the individual lumped-parameter models of the units, based on the manner in which they are connected. The method is demonstrated by simulating four different configurations of a particular type of muscle-like actuator driven by the contractions of individual miniature solenoids, demonstrating the flexibility of the approach, and how the system dynamics vary with the activation of the individual units.

Keywords: Actuators, Rehabilitation Robots
Abstract: With the rapid development of soft exoskeletons, traditional cylindrical pneumatic artificial muscles (PAM) can no longer meet people's needs of wearable exoskeleton for hand rehabilitation. We propose a novel flat pneumatic artificial muscle with the advantage of lightweight, small size, and fitting on the skin without squeezing, made of silicone rubber and embedded fibers. The fibers embedded in the silicone substrate can convert a radial expansion into axial contraction. By finite element modeling (FEM), we obtain the deformation and stress distribution of the PAM preliminary designed under different air pressures. Finally, the flat PAM with better structure and performance is fabricated. Some experiments were carried out to measure the contraction rate and output force. The PAM has a maximum output force of 20.4N and a maximum contraction rate of 9.8% and with a weight of only 19.8g. Finally, a simple wrist exoskeleton is made to verify its availability.

Keywords: Actuators, Compuational Models and Methods, Fuel Cells and Alternative Power Sources
Abstract: This paper presents a quantitative evaluation of a linear and rotary type of thermomagnetic motors for their suitability for heat energy harvesting. Mathematical models of the thermomagnetic motors are developed and the output characteristics of the thermomagnetic motors simulated using these models. The result shows that the power transfer in the linear motor is more effective than in the rotary device under similar operating conditions. The linear device does require a higher temperature source for activation but consumes less heat once the device is in motion. It is observed that output from each device can be further enhanced by appropriate selection of the operating point to match the design of the device. For instance, operating the rotary device at lower speeds and finding the right balance between heating and cooling rates does yield more output from the device.

Keywords: Actuators, Control Application in Mechatronics, Identification and Estimation in Mechatronics
Abstract: In this paper, we present the force ripple reduction method for iron-cored permanent magnet linear synchronous motors (PMLSMs), using position-dependent force constant. The iron-cored PMLSMs generate magnified force ripple due to magnetic saliency, which makes system vibrations and deteriorates the closed-loop performance. To suppress such motor force ripple, we propose the control method considering not only geometry-driven ripple (GDR) but also current-driven ripple (CDR) which is investigated in detail how it can be generated in PMLSMs. We also present the identification method for both current-driven force (CDF) and geometry-driven ripple in simple experimental setup without additional equipment, and it is validated experimentally with NRMSE (peak-to-peak normalized root mean square error) of 4.74 % and 1.01 % for CDF and GDR, respectively. Using the identified motor parameters, the force ripple is reduced by 91.5 % and 86.9 % in RMS and peak-to-peak value, without sacrificing the motor thrust and with small additional power consumption.

Keywords: Actuators, Control Application in Mechatronics, Actuators in Mechatronic Systems
Abstract: Elastic actuators show advantages in energy efficiency and safety in human-robot interaction. However, such actuators might be subject to faults due to their complexity, particularly in elastic and kinematic elements. We present a stiffness-fault-tolerant control strategy for a redundant actuator with multiple elastic actuation units. The control scheme is capable of detecting and compensating for stiffness changes in individual elastic elements. We develop the dynamics of the actuator and a model-based impedance control scheme. The obtained closed-loop dynamic interaction behavior of the system and its stability is analyzed. Oscillatory trajectory simulations are used to evaluate the control strategy. Results show that the controller is capable of tracking a reference trajectory with desired interaction impedance behavior under fault conditions and interaction disturbances, while exploiting redundancy to perform load compensation.

Keywords: Actuators, Control Application in Mechatronics, Identification and Estimation in Mechatronics
Abstract: Twisted string actuators (TSAs) have exhibited high performance in numerous mechatronic applications. Most existing studies on control of TSAs assume that the external force applied to the TSA is measurable or predictable. Furthermore, existing studies also assume that all the TSA parameters such as the motor properties and string stiffnesses are accurately known. However, the system parameters could be difficult to measure, could change over time due to general wear and tear, and creep. The external forces applied to the TSA could be difficult to predict or measure. In this study, parameter estimation and control strategies were developed for TSAs, assuming little or no knowledge about the system parameters. Firstly, a parameter estimation strategy which utilized the least squares algorithm and gradient algorithm is presented. Secondly, an adaptive control strategy based on model reference control with feedback linearization is proposed. The developed estimation and control strategies were tested to be effective through simulation. The performance of the proposed control strategy was compared with a proportional controller (PC) and a proportional controller with a feedforward term (PCFF). The average TSA length tracking errors for the proposed controller, PC, and PCFF were 2.6 × 10−4 m, 1.4 × 10−3 m, and 7.8 × 10−4 m, respectively.

Keywords: Actuators in Mechatronic Systems, Control Application in Mechatronics, Automotive Systems
Abstract: In this paper, taking hydraulic system of mining hydraulic excavator boom and stick as the plant, an adaptive robust impedance controller based on an extended state observer and backstepping method is designed. The energy-saving operation and position tracking performance are considered in system modeling and control law design. The system effects on electro-hydraulic proportional valves with open centers and flow regeneration valves are considered. When hydraulic cylinders are extended, the proposed control strategy adopts a pump-controlled pressure closed-loop to shorten the dead time. When hydraulic cylinders are retracted, less oil is output to meet the idle demand. The valve-controlled position subsystem with impedance mode ensures smoother digging processes. Comparative simulations show that the proposed control strategy has better position tracking and smaller velocity fluctuations. Compared with the system without flow regeneration valves, the proposed system can reduce energy consumption by 27.32%.

Keywords: Intelligent Sensors, Actuators in Mechatronic Systems, Design/control of MEMS-nano devices
Abstract: Physical intelligence (PI) is an emerging research field using new multi-functional smart materials in mechatronic designs. On the microscopic scale, PI principles give rise to unconventional transducers, which are especially useful for micro/nano-robot design with size and resource constrains. Since it is not easy to directly observe nanoscale multi-physics phenomenon, understanding their principles can be challenging. In this work, we bring PI principles into the metaverse to bridge this gap by developing two mixed reality scale models. The first example is a virtual reality (VR) 2D material twistronics visualizer to demonstrate the novel intelligent 2D materials with tunable properties as a rising field in condensed matter physics. Users can interactively control the cross-coupling multi-physics phenomena and observe the visualized material responses. The second example is centered around an Atomic Force Microscope (AFM) to illustrate its imaging and probe principles. For interaction, users can control the twist angle using atomic lattice models and feel the AFM cantilever force using custom haptic devices. We believe these tools can help precision mechatronic engineers understand and make better use of physical intelligence building blocks to design micro-electromechanical systems.

Keywords: Sensor Integration, Data Fusion, Mobile Robots, Machine Learning
Abstract: Visual-inertial odometry (VIO) systems traditionally rely on filtering or optimization-based techniques for egomotion estimation. While these methods are accurate under nominal conditions, they are prone to failure during severe illumination changes, rapid camera motions, or on low-texture image sequences. Learning-based systems have the potential to outperform classical implementations in challenging environments, but, currently, do not perform as well as classical methods in nominal settings. Herein, we introduce a framework for training a hybrid VIO system that leverages the advantages of learning and standard filtering-based state estimation. Our approach is built upon a differentiable Kalman filter, with an IMU-driven process model and a robust, neural network-derived relative pose measurement model. The use of the Kalman filter framework enables the principled treatment of uncertainty at training time and at test time. We show that our self-supervised loss formulation outperforms a similar, supervised method, while also enabling online retraining. We evaluate our system on a visually degraded version of the EuRoC dataset and find that our estimator operates without a significant reduction in accuracy in cases where classical estimators consistently diverge. Finally, by properly utilizing the metric information contained in the IMU measurements, our system is able to recover metric scene scale, while other self-supervised monocular VIO approaches cannot.

Keywords: Human -Machine Interfaces, Sensors and Sensing Systems, Automotive Systems
Abstract: Automated Driving Systems (ADS) are becoming ubiquitous to reduce the workload of drivers and improve road safety. However, present-day ADS lacks accurate and effective driver monitoring systems. Driver monitoring systems use physiological measurements, such as pupil dilation, eye-gaze, and eye-blinks, in order to monitor the cognitive load experienced by the drivers. With advances in eye-tracking technology, pupil dilation is emerging as a reliable measure of cognitive load in ADS. However, pupil dilation as a measure of cognitive load suffers from many factors, such as confounding effects, noise, and personal attributes, to name a few. Hence, in order to improve cognitive load estimation in ADS, other non-invasive measures must be studied and incorporated. In this paper, various eye-gaze metrics are studied and evaluated as a measure of cognitive load based on data collected from 16 drivers in a simulated driving scenario using a driving simulator.

Keywords: Opto-Mechatronic Sensors, Modeling and Design of Mechatonic Systems, Underwater robotics
Abstract: LED-based free-space optical communication is becoming a promising alternative to the traditional acoustics mode of underwater communication because of its low cost, high data rate, and low power consumption. However, a fundamental challenge associated with the free space optical communication systems is to establish and maintain Line-Of-Sight (LOS) between the two communicating agents. It becomes even more challenging when the agents are in motion, which necessitates an active alignment system that enables each agent to point itself towards the direction of LOS continuously. In this paper, we present a bi-directional active alignment control approach for LED-based communication system of two robots, where the robots navigate in a three-dimensional (3D) space. We propose an extended Kalman filter (EKF) based alignment approach, where the estimates of azimuthal and elevation components of heading bias with the LOS are used to correct the alignment. We introduce and implement a new circular scanning technique on a two-degree-of-freedom (DOF) rotational system, mounted on a robot, that enables consecutive independent measurements from a single photo-diode, which are necessary to satisfy the observability constraints of the EKF. Furthermore, we explore a synchronized alternating scheme to address the bidirectional nature of the problem. The scanning amplitude is further adjusted based on the EKF estimation covariance, to balance the trade-off between estimation accuracy and actuation effort. We compare the proposed approach with an extremum-seeking (ES) approach on both simulation and the experiments, where a setup of two robots with relative 3D motion is considered. The presented results support the efficacy of the proposed method in the presence of slow relative motion between the robots, and demonstrate the superiority of the proposed approach over the ES method over a wide range of distances between the robots.

Keywords: Intelligent Sensors, Opto-Mechatronic Sensors, Sensors and Sensing Systems
Abstract: For robots to perform advanced manipulation of objects, touch is a critical source of information, and a high-quality tactile sensor is essential. Image-based optical tactile sensors, and its inheritances, which have soft touch interfaces, can provide high-resolution tactile images of the contact geometry, contact pressure, and slip conditions. However, due to the lack of robustness provided by the current tactile sensors, the ability to grasp hard or sharp objects is minimal. In this work, we propose an image-based optical tactile sensor and overcome the above limitation of poor robustness by introducing a latex layer on the touch interface. We use a combination of silicone elastomer covered with a latex material and an acrylic sheet to support the silicone elastomer. A camera placed at the bottom of the sensor housing captures the deformation of the elastomer surface illuminated by an inner light. To evaluate the performance, we carried out a series of experiments. First, we evaluated the mechanical characteristics of the silicone elastomer with three types of coating, namely latex membrane, metallic coating, and no coating. The proposed latex membrane clearly outperformed the other two in terms of robustness. Second, we carried out the force-displacement experiments quantitatively to further study the sensitivity and robustness. Last, we validated the sensor performance in terms of its spatial resolution by applying the VGG-19 neural network for classifying touch patterns captured by the sensor. Overall, the proposed sensor achieved the desired robustness, sensitivity, and spatial resolution performance.

Keywords: Sensors and Sensing Systems, Human -Machine Interfaces, Neural Networks
Abstract: Motivated by the increased needs of home-based rehabilitation for stroke patients, more and more interest has been drawn towards developing body-fixed sensors for monitoring affected joint motion. Although relatively accurate bone geometries can be obtained by scanning technologies, most human joints are approximated by simple circles and spheres in order to reduce the highly nonlinear kinematics to a tractable form for motion studies; many human joint-motion sensing challenges remain open. In this paper, a magnetic tensor sensor (MTS) and its measurement model implemented on an artificial neural network (ANN) trained with back-propagation for tracing a human joint trajectory are presented. The effects of input configurations and datatypes on measurement accuracy of a MTS/ANN have been numerically investigated with published data and experimentally evaluated on a prototype pantographic exoskeleton worn on a human shank/foot. As demonstrated experimentally, the MTS/ANN system calibrates the sensor intrinsic parameters, accounts for the environmental magnetic effects on the measurements, and can be trained with both offline and user-specific data. While illustrated in the context of ankle joint motion sensing in the sagittal plane, the MTS/ANN can be potentially extended to 6-DOF joint motion measurements.

Keywords: Hybrid intelligent systems, Intelligent Sensors, Human -Machine Interfaces
Abstract: In industrial applications, mechanical and physiological thresholds may limit the capability of human manipulating machine via control devices, such as joysticks and steering wheels. These thresholds can result in loss of information in the control signals that are kept below the threshold of detection of the device or the human operator. One approach to mitigate these effects is stochastic resonance, by injecting additive noise into a signal to raise its energy content over the threshold of detection. Though this noise partially corrupts the signal, it can increase the detectability of the signal by the control device. This paper provides, for the first time, research towards using stochastic resonance to improve human performance in control tasks. In particular, it shows that using adaptive colored noise can improve the detectability of the steering control signals recorded from human participants. The approach converts a signal processing task to an optimization problem, where particle swarm optimization is employed to obtain the optimal color (or spectral exponent) of the injected additive noise, generated through an intelligent technique with fractional order filters. The results have shown that the proposed method improves the detectability of sub-threshold steering control signals. This method can be widely applicable to other industrial domains, such as energy harvesting and enhancing sensory perception.

Keywords: Sensors and Sensing Systems, Motion Vibration and Noise Control, Mechatronics in Manufacturing Processes
Abstract: This paper presents a six degree of freedom (DoF) active sample-tracking robotic 3D measurement system for inline applications. The integrated measurement platform (MP) of the state-of-the-art system is augmented by a compact and tailored in-plane tracking sensor system. Based on two position-sensitive devices and laser markers, relative motion between the 3D measuring tool on the MP and a sample surface is measured with sub-micrometer resolution. Applying a tailored high performance PID control architecture, a six DoF sample-tracking control bandwidth of about 450Hz is achieved. In a translational inplane DoF, vibrations with a dominant component at 66Hz are reduced from 9.44μm rms by a factor of 10 to 931nm rms. Experimental results successfully proof the system concept of robotic 3D measurements with sub-micrometer precision directly in a challenging vibration-prone environment.

Keywords: Legged Robots, Mobile Robots, Design Optimization in Mechatronics
Abstract: Among small-scale mobile robots, multi-modal locomotion can help compensate for limited actuator capabilities. However, supporting multiple locomotion modes or gaits in small terrestrial robots typically requires complex designs with low locomotion efficiency. In this work, legged and rolling gaits are achieved by a 10~cm robot having just two degrees of freedom (DoF). This is acheived by leg shaping that facilitates whole body rolling and event-driven control that maintains motion using simple inertial sensor measurements. Speeds of approximately 0.4 and 2.2 body lengths per second are achieved in legged and rolling modes, respectively, with low cost of transport. The proposed design approach and control techniques may aid in design of further miniaturized robots reliant on transducers with small range-of-motion.

Keywords: Walking Machines, Legged Robots
Abstract: For robots to successfully locomote in different environments, it's better to equip them with multiple models of locomotion. Most existing robotic systems realize multimodal locomotion by integrating multiple mechanisms into one single robot. These robots are generally cumbersome or challenging to control and actuate. Recently, shape morphing structures are utilized to enable reconfigurable and multi-functional robots. In this paper, we present a novel origami-inspired module that can change its shape and motion. The modules consist of joints that can be individually controlled to be soft or rigid, thus enabling the reconfiguration of the modules under actuation. To understand the reconfiguration capability, we numerically analyzed the programmable shapes and motions for a module. Using the modules, we developed a reconfigurable robot with four legs, each made from four serially connected modules. The robot can walk, crawl, and inch using the same mechanical structure. We anticipate that the proposed method can be leveraged to enable robots with physical intelligence to adapt their morphologies (e.g., body shapes, leg orientations) and behaviors (i.e., locomotion modes) in response to external environments.

Keywords: Humanoid Robots, Legged Robots, Walking Machines
Abstract: State estimation for legged locomotion over a dynamic rigid surface (DRS), which is a rigid surface moving in the world frame (e.g., ships, aircraft, and trains), remains an under-explored problem. This paper introduces an invariant extended Kalman filter that estimates the robot's pose and velocity during DRS locomotion by using common sensors of legged robots (e.g., inertial measurement units (IMU), joint encoders, and RDB-D camera). A key feature of the filter lies in that it explicitly addresses the nonstationary surface-foot contact point and the hybrid robot behaviors. Another key feature is that, in the absence of IMU biases, the filter satisfies the attractive group affine and invariant observation conditions, and is thus provably convergent for the deterministic continuous phases. The observability analysis is performed to reveal the effects of DRS movement on the state observability, and the convergence property of the hybrid, deterministic filter system is examined for the observable state variables. Experiments of a Digit humanoid robot walking on a pitching treadmill validate the effectiveness of the proposed filter under large estimation errors and moderate DRS movement. The video of the experiments can be found at: https://youtu.be/ScQIBFUSKzo.

Keywords: Legged Robots, Robot Dynamics and Control
Abstract: The paper reports on a model-based methodology for a leg-wheel transformable robot to autonomously determine the use of the wheeled or legged mode based on environmental RGBD information. A concentric and reduced-order dual-leg-wheel model was developed to explore the dynamic interactions between the leg-wheel and terrain. The simulation results revealed that terrain height variation acts as the key factor in determining the model’s terrain negotiability; therefore, it was utilized as the decision index for leg-wheel motion selection. To facilitate real-time motion selection on the empirical robot, a two-step selection strategy utilizing RGBD information was proposed. The first step utilized RGB images to classify seven types of terrain. If the classified terrain was potentially rough, the second step utilized depth images to compute the terrain height variation, which was then compared to the index derived by the dual-leg-wheel model to select the final operation mode. The strategy was implemented on a leg-wheel transformable robot and experimentally validated on various types of terrain. The results confirm that the robot’s behavior closely resembled the model’s prediction, and using the proposed strategy to select leg-wheel motion in the robot yielded the most energy efficient locomotion.

Keywords: Legged Robots, Robot Dynamics and Control, Modeling and Design of Mechatonic Systems
Abstract: Locomotion through resistive media is an organic occurrence during traversal of the natural world. Due to the complexities required to analyze the effect of these media on the dynamics of locomotion, controllers of legged robots generally neglect or treat them as disturbances. In this paper, we address the challenge of producing optimal locomotion control in resistive media. We do so by applying trajectory optimization techniques within a direct collocation framework onto a reduced-order resistive model of legged locomotion: the Fluid Field SLIP model. The results of this optimization led to five different optimal gaits being found for hopping in air, fluidized media, and at the interface between these fluids. When applying the optimal control policies to a single leg robotic hopper in mixed fluid it was found that the new controller was able to improve its efficiency by 54% over the previous controller. It achieved this by employing a novel "kickback" and retraction maneuver found by the optimizer. Which was found to improve efficiency even in un-optimized controllers when hopping in deep fluid.

Keywords: Legged Robots
Abstract: Modulation of joint stiffness is a natural ability of humans, which enables us to interact smoothly and stably with environments. Inspired by this observation, variable impedance control has also been studied for robots in various applications. However, the design of joint impedance profiles that fulfill feasible requirements and optimality is non-trivial. In this paper, we present a novel scheme to modulate the joint stiffness with guaranteed stability for torque-controlled quadruped robots. The joint stiffness and desired contact forces are optimized coordinately in a quadratic programming (QP) formulation, where the constraints of non-slipping contacts and torque limits are also imposed. Moreover, the stability during stiffness modulation is guaranteed by a tank-based passivity constraint. The effectiveness of the proposed stiffness modulation is validated on our quadruped robot CENTAURO, demonstrating that it is capable of generating more compliant contacts while ensuring necessary tracking performances.

Keywords: Compuational Models and Methods, Image Processing, Sensor Integration, Data Fusion
Abstract: Although the three-dimensional (3D) reconstruction result obtained by structure from motion and multi-view stereo from an image can visually confirm whether 3D reconstruction is achieved, it is not possible to scale the obtained 3D reconstruction result. Therefore, we verify whether the 3D reconstruction can be scaled based on the size of the two-dimensional (2D) code by simultaneously performing a 3D reconstruction of its target and its code whose size is known. Experimental results show that the scaled 3D reconstruction is highly accurate according to the size of the reconstructed 2D code.

Keywords: Robot Dynamics and Control, Control Application in Mechatronics
Abstract: In this paper a load-side acceleration control scheme for geared motors affected by unknown backlash and friction is presented. The proposed scheme consists on the combination of a backlash-deactivated disturbance observer (BDDOb), which conditionally estimates and compensates load-side disturbances, and a variable-order force-position-integrated disturbance observer (VOFPIDO) which quickly compensates nonlinear friction. Experimental results prove the effectiveness of the proposed method

Keywords: Aerial Robots, Unmanned Aerial Vehicles, Mobile Robots
Abstract: We have developed a flying robot which consists of a multi-rotor UAV with three-arm manipulator system. In this study, we propose a method of flight through narrow space and collision avoidance using robotic arms on the flying robot and performed collision avoidance simulations. In the proposed method, the multiple arms are stretched out to prevent the propellers from touching with obstacles when the UAV flies in narrow space. We confirmed the effectiveness of the proposed method through experiments and simulation.

Keywords: Rehabilitation Robots, Biomechatronics, Virtual Reality and Display
Abstract: An inexperienced patient who uses crutches tends to use them with an inappropriate method if an appropriate instruction is not provided, and this may cause a secondary accident like a falling. In order to prevent such an accident, the authors have been developing a crutch walk training system with ICT (Information and Communication Technology). In this paper, functions of the developed crutch walk training system are introduced.

Keywords: Robot Dynamics and Control, Control Application in Mechatronics
Abstract: In this paper, a single inertialization compensator (SIC) on the load-side is proposed for the robot actuator with the base/link dual resonance vibration (B/LDRV) model. The effectiveness of the proposed method is verified by simulations and experiments.

Keywords: Robot Dynamics and Control, Control Application in Mechatronics
Abstract: This paper proposes a robot motion control system based on the Finite-Diiference Time-Domain (FDTD) state observer and the FDTD dynamic torque compensation method to compensate for inertia fluctuations and interference forces caused by the robot motion. The effectiveness of the proposed method is verified by experiments with a robot manipulator

Keywords: Control Application in Mechatronics, Machine Learning
Abstract: This paper proposes a force sensing approach with the adaptive signal decomposition based on singular spectrum analysis. The disturbance observer and the singular spectrum analysis are designed for force sensing function. The adaptive signal decomposition is constructed based on a deep neural network to achieve fine force estimation in variable noise conditions. The effectiveness of the proposed method is verified by experimental results.

Keywords: Modeling and Design of Mechatonic Systems, Flexible Manipulators and Structures, Novel Industry Applications of Mechatroinics
Abstract: The robot hand was developed for use in environments such as food processing plants and medical facilities, where conventional robots are considered difficult to introduce. The finger which consists of a single sheet of parts, can be replaced, allowing for disposable use.

Keywords: Human -Machine Interfaces
Abstract: In recent years, automation has been progressing in factories in various industries, but production lines that use only robots cannot flexibly respond to changes in production items. Therefore, human-robot collaboration (HRC), in which robots and humans work in the same space, is widely introduced to flexibly respond to changes in the production line.However, current collaborative robots are not equipped with a function to care about and communicate the health conditions of the collaborator. Without this function, it is difficult to prevent human error and accidents from occurring, because the collaborator cannot be aware of his or her own health condition. To solve this problem, we focused on the social interaction of cooperating robots, and created a system that evaluates health conditions such as stress based on a person's heart rate and provides feedback of the evaluation as facial expressions. We also conducted experiments of collaborative work using the system and verified the system. The results confirm that the subjective fatigue values obtained by reflecting on the work and the corresponding health status feedback are possible.

Keywords: Human -Machine Interfaces, Vehicle Control
Abstract: Electric wheelchairs are used as a means of active mobility for the elderly and physically handicapped people as a welfare equipment. People who suffer from muscular dystrophy, or other diseases and disorders that disable a large part of their body have difficulty operating electric wheelchairs because they cannot grasp the joystick due to muscle weakness. Muscle strength may be reduced with muscular dystrophy, but the loss of range of motion of the fingertips may be gradual. This research aims to design an operation interface that allows multi-directional input without requiring force for people without grasping strength for personal mobilities. To focus on the body burden, we evaluated the EMG data obtained from the validation experiment using amplitude probability distribution function (APDF) analysiss, which is used in the field of occupational physiology, etc. We quantified the physical burden of input by APDF analysis and compared the results, suggesting that the physical burden of our proposed interface is lower than that of a joystick used for electric wheelchairs.

Keywords: Virtual Reality and Human Interface, Image Processing
Abstract: In recent years, there is a shortage of caregivers in Japan due to the increase in the elderly population. At the same time, the caregivers themselves are aging, and elderly caregivers are caring for the elderly people. As a result, the burden per caregiver has increased, which has become a social problem. In order to deal with this problem, care support robots and information and communication technology have been introduced, but the work that can be done by each support is limited. In addition, caregivers are exposed to various events and situations in the care environment, which increases the cognitive burden even for caregivers with advanced care skills. To solve these problems, we develop a support system that reduces workload and improves work efficiency by recognizing caregivers' work situations in an intelligent space and providing flexible work support according to the work situation. To this end, the Ministry of Health, Labor and Welfare (MHLW) currently states that work procedures are important for safe and efficient caregiving operations. Therefore, this study proposes an information presentation support system to enable caregivers to perform a series of tasks in the proper sequence. Through experiments, it is shown that the proposed information presentation support system improves work efficiency.

Keywords: Control Application in Mechatronics, Human -Machine Interfaces, Medical Robotics/Mechatronics
Abstract: —In recent years, real-world haptics has been studied in particular, and its role as a human support system has been emphasized in various fields. One of the studies that have been conducted on the preservation and reproduction of human motion is the motion copying system. Although this system can store accurate position and force using actuators in motion preservation, it is difficult to reproduce flexible human motion. Therefore, this paper discusses the possibility to reproduce the motion of a human being itself by using external sensors such as cameras and force sensors to store position and force information. In this paper, the feasibility is confirmed from simulation for the motion reproduction based on the measured human data by the external sensors.

Keywords: Mechatronics-Enabled Teaching and/or Training, Control Application in Mechatronics, Sensor Integration, Data Fusion
Abstract: In this research, a useful device in walking training for the visually impaired is developed. An IMU sensor is attached to a white cane and analyzed its movement. And then, to reduce the error of the sensor, the Kalman filter is applied to its output values. The validity is confirmed by the preliminary experiments.

Keywords: Mobile Robots, Legged Robots, Compuational Models and Methods
Abstract: The authors have developed a hexapod tracked mobile robot. This robot has six legs attached to the body. This robot is able to traverse a wide gapped-step by supporting the track driving using four front and rear legs while holding the target object by two middle legs in the transportation task. In this research, aiming at the autonomous operation of this robot, we analyzed the relationship between the gap width and the step height that can be traversed by the robot. The simulation results showed possible gap width and step height required for autonomous operation by the robot.

Keywords: Intelligent Process Automation, Sensors and Sensing Systems
Abstract: Humans perform complex tasks by using tools on a daily basis. These tasks can be divided into several simple movements (movement primitives). In order for the robot to be able to perform these tasks, a method that can autonomously detect primitive changes is needed. In particular, such tasks often involve contact state changes. In order to detect primitive changes, it is effective to refer to contact state changes. We propose a segmentation method that uses differential force and torque data caused by contact. In the experiment, we performed a segmentation of bottle lid opening task. We found that primitive changes can be detected by differential force and torque data.

Keywords: Planning and Navigation, Mobile Robots, Vehicle Control
Abstract: In recent years, many traffic accidents have occurred due to the increasing demand for electric wheelchairs. To solve this problem, many studies have been conducted on autonomous electric wheelchairs. Many of the factors that cause a burden to the user exist in dynamic environments where dynamic obstacles such as pedestrians exist. Therefore, a navigation method to realize a comfortable electric wheelchair for users has not been established. In this study, we proposed a strategy for safe and efficient avoidance of dynamic obstacles by performing global path planning while referring to a map, and by replacing dynamic obstacles to be avoided with virtual obstacles. Simulation experiments were conducted to verify the usefulness of the proposed method. The results showed that the proposed method provides smooth avoidance of single or multiple dynamic obstacles with little speed change and safe and efficient driving.

Keywords: Mobile Robots, Sensor Integration, Data Fusion
Abstract: SLAM using a monocular camera, has the problem of being unable to estimate the scale. This paper proposes a method to estimate the scale using a monocular camera and a shortsighted 2D lidar (short-range lidar). However, if a point cloud cannot be obtained for a short-range lidar, scale drift will occur during that time. Since the scale can be estimated when the point cloud is obtained at short-range lidar, the proposed method can eliminate scale drift and correctly estimate the scale as a whole.

Keywords: Control Application in Mechatronics, Human -Machine Interfaces, Rehabilitation Robots
Abstract: This study was a pilot study of a healthy subject to investigate the changes that occur with the use of a robotic hip prosthesis in patients with hip disarticulation. Patients with hip joint or near hip joint anomalies or amputations often have lower and limited mobility compared to healthy individuals as well as other lower limb amputees. The use of a robotic hip prosthesis may improve the burden of walking and moving in these patients. we performed gait training and measurements on non-amputee patients with a non-powered hip prosthesis, a robotic hip prosthesis, a motion sensor, and a VO2 sensor. As a result, the use of the robotic hip prosthesis reduced the burden of walking and improved gait, although the fatigue level in walking slightly increased. This indicates that the use of the robotic hip prosthesis has a positive effect on the patient's gait, but is associated with more fatigue than usual.

Keywords: Design/control of MEMS-nano devices, Micro/Nano Manipulation, Micro-Electro-Mechanical Systems
Abstract: In this article, we propose laser actuated microjoints which can be remotely actuated in both air and water. Their actuation relies on the optothermal response of a spiral bimaterial. The microjoints are fabricated using two-photon polymerization technology that offers the ability to tune the thermal and mechanical properties of the material by controlling the laser printing power. Modeling is first conducted to verify the parameters of the spiral that affect the rotational displacement and generated torque of the microjoint. Then, microjoints having a diameter of less than 200 um are characterized. The microjoints can realize a maximum deflection of approximately 8.5 degrees, a force in the uN-order using a 265-um long arm, an actuation repeatability of more than 100 times, and a time response of approximately 34 ms. Finally, the microjoints are implemented in a microgripper and an xy serial microarm. Successful micromanipulation of 40 um microbeads using the microgripper, and the simultaneous actuation of multiple microjoints of the xy serial microarm with two degrees of freedom are shown. This kind of rotational, compact, selective, and remotely actuated microjoints would allow the deployment of individually controlled mobile microrobots with several degrees of freedom for complex applications such as cell manipulation and microassembly.

Keywords: Applications of nano technology, Micro-Electro-Mechanical Systems
Abstract: Performing electrical Atomic Force Microscopy measurements in aqueous solutions is of paramount importance in a vast range of scientific fields. As with operation at ambient conditions, it is important to ensure that the force on the cantilever is purely of electrostatic origin. However, with the insertion of water and therefore mobile ions and polar molecules into the tip-sample system come several unwanted effects. Here, an experimental study is carried out, analyzing the influence of parameters such as drive-frequency, -amplitude and ionic concentration on the feasibility to perform electrical AFM measurements in aqueous solutions. To this end the system is theoretically modelled and parameter sweeps are preformed, leading to transition frequencies above which the electrostatic force has the predominant impact on the cantilever.

Keywords: Actuators, Modeling and Design of Mechatonic Systems, Actuators in Mechatronic Systems
Abstract: Active Combustion Control (ACC) has been proved to be an effective method for the suppression of pressure oscillation in aero engine combustion chambers. However, the actuators for ACC need to work with a bandwidth of 1000 Hz, a stroke up to submillimeter level and a high power density simultaneously. Existing electromagnetic actuators, or smart material actuators, are still unable to meet these requirements at the same time.To address this issue, in this paper, a Multi-dimensional Discrete Magnetostrictive Actuator (MDMA) was developed by adopting the multi-dimensional discrete configuration. Then, a Magnetic Equivalent Circuit (MEC) model was developed to analyze the magnetic field distribution characteristic of MDMA components. Next, an multiphysics comprehensive dynamic model was established to describe the complex dynamic properties of multi-dimensional discrete structures. Subsequently, a prototype of the MDMA was fabricated which is only 56 mm in diameter and only 71.5 mm high. Experimental results indicate that, the proposed MDMA reaches a stoke of 100 μm, and the working bandwidth can exceed 1000 Hz.

Keywords: Actuators in Mechatronic Systems, Modeling and Design of Mechatonic Systems, Design Optimization in Mechatronics
Abstract: Actuation of micro-robots in relation to the field of programmable matter has been the aim of most researches working on the topic. Most studies have concentrated on the ability of these robot modules to latch with one another and to be able to move around each other to form the desired configuration. These mostly have been based on Modular Self re-configurable Robots (MSR) which use mechanical, magnetic or pneumatic method in their manoeuvres. However, these have been faced by the challenges of miniaturization, power consumption and power transfer between modules. Electrostatic actuation has attracted more interest in recent works due to the ability to scale down the modules and to use the configuration for power transfer and communication. In this work, we propose to use electrostatic chuck principle to actuate modules having a cylinder form. Instead of using one array of electrodes, two columns of electrodes array, positive and negative, are considered to increase the force generated by the chuck effect. An array of electrodes prototype has been fabricated using lithography process for validating the actuation. A sufficient force and torque are generated for latching and making the cylinder rolling respectively, thus validating and showing the potential of electrostatic actuation for miniaturized programmable matter application.

Keywords: Actuators in Mechatronic Systems, Actuators, Modeling and Design of Mechatonic Systems
Abstract: The twisted string actuator (TSA) has been extensively studied as an alternative to metallic gear reducers due to its low-cost, lightweight, compliance, and highly adjustable actuation characteristics. Owing to these advantages, TSAs were utilized to realize lightweight human-robot interactive robots requiring safe interaction, such as wearables and prosthetics, at a relatively low cost. However, because its force transmission ratio is almost proportional to the length of the string, the string of the conventional TSA must be elongated to achieve high force output. Thus, it is difficult to obtain high force capacity while maintaining a small size. To overcome this limitation, we propose a novel TSA based on parallel bundle-driven actuation to significantly reduce the length of the TSA while achieving high force output. Furthermore, in order to boost the proposed TSA performance, we also present a passive mechanism of an asymmetric behavior compensator. The experimental results show that the force transmission ratio of the proposed TSA is increased by 21%, compared with that of the conventional TSA. Consequently, the proposed TSA has a 44% higher torque capacity with a 37% shorter string than those of the conventional TSA.

Keywords: Actuators in Mechatronic Systems, Design Optimization in Mechatronics, Modeling and Design of Mechatonic Systems
Abstract: Variable stiffness actuators (VSAs) can effectively protect humans in case of an accidental collision owing to the adjustable compliance. This paper proposes a variable radius approach to develop a variable stiffness actuator with a large stiffness range based on the antagonism principle. Unlike traditional antagonistic VSAs, the stiffness control and position control are decoupled, and stiffness adjustment is achieved through the simultaneous regulation of spring stiffness and joint radius. Specifically, the regulation of joint radius is achieved by using a variable configuration mechanism (VCM). A quadratic spring is constructed using a variable radius pulley (VRP). A novel profile function of the VRP is proposed, and mathematical models of joint stiffness are established. A VSA prototype is developed to evaluate the performance, and an experiment is carried out to verify the effectiveness of the proposed quadratic spring. Joint position and stiffness are driven by actuators with position control strategies. The results of the stiffness experiment show that the stiffness can be effectively adjusted in a large range by adjusting the joint radius and the stiffness of the quadratic spring.

Keywords: Flexible Manipulators and Structures, Novel Industry Applications of Mechatroinics, Actuators in Mechatronic Systems
Abstract: This paper presents a novel soft robotic system for a deformable mannequin that can be employed to physically realize the 3D geometry of different human bodies. The soft membrane on a mannequin is deformed by inflating several curved chambers using pneumatic actuation. Controlling the freeform surface of a soft membrane by adjusting the pneumatic actuation in different chambers is challenging as the membrane's shape is commonly determined by the interaction between all chambers. Using vision feedback provided by a structured-light based 3D scanner, we developed an efficient algorithm to compute the optimized actuation of all chambers which could drive the soft membrane to deform into the best approximation of different target shapes. Our algorithm converges quickly by including pose estimation in the loop of optimization. The time-consuming step of evaluating derivatives on the deformable membrane is avoided by using the Broyden update when possible. The effectiveness of our soft robotic mannequin with controlled deformation has been verified in experiments.

Keywords: Flexible Manipulators and Structures, Modeling and Design of Mechatonic Systems, Design Optimization in Mechatronics
Abstract: This paper presents the design and prototype of soft-fingered AI-enabled hand (SofIA) based on the Fin-Ray ® effect. The study proposes a material and method for fabricating soft Fin-Ray fingers by molding them entirely from urethane rubber. SofIA is equipped with a depth camera that provides visual feedback on the state of the fingers, which will be used in the development of a versatile sensing system based on deep learning. Flexible side supports were added to further improve the mechanical performance of the fingers. Using SofIA, a series of experiments were conducted with the original and modified Fin-Ray finger structures to test and validate the desired behaviour of the gripper. It was found that the hand is capable of manipulating objects ranging from 10 mm to 90 mm in diameter, objects up to 90 mm x 90 mm in length and width, and objects with a maximum mass of 400 g in a position parallel to the ground without any effect on the object material.

Keywords: Flexible Manipulators and Structures, Biomechatronics
Abstract: Robotic hand has vast application potential in home-serviced, medical care, surgery, maintenance, Etc. Various robotic hands made of rigid and flexible materials have been developed in the past decade. In this work, we present a rigid-flexible coupled underactuated hand with good agility and compliance for various kinds of grasping and safe operations. First, the dynamic simulation for finger bending and underactuated hand grasping was studied. Then, a prototype of the underactuated hand was fabricated. Next, we carried out a series of experiments based on the prototype. Single finger bending performance and blocking force were tested, respectively. We also tested the adaptive grasping ability of the hand. Finally, we performed collision experiments and compared the impact forces of the Be Bionic hand and our underactuated hand. Results showed that the underactuated hand could grasp objects with different shapes and possessed a higher degree of safety, reducing impact force by 30% to 50%.

Keywords: Flexible Manipulators and Structures, Design Optimization in Mechatronics, Modeling and Design of Mechatonic Systems
Abstract: Adaptive grasping is an important approach for robotic grippers to handle objects with irregular shapes. Compared to rigid-link-based adaptive grippers, the continuum-structure grippers benefit from their structural compliance and have thus a higher degree of adaptive grasping freedom. Based on this advantage, we have developed a continuum-structure-based double-finger gripper in this paper to achieve the adaptive grasping. To improve the design efficiency, a 3D-topology-optimization-based design method was adopted in this work, which realized the adaptive-grasping function of the robotic finger by introducing an additional spring into the design problem. The proposed robotic gripper was selective-laser-sintered (SLS) with the material polyamide (PA2200) and was actuated by a linear motor. Experiments were also conducted to evaluate the grasping performance and load capacity of the developed gripper. Results have shown that the gripper could successfully grasp objects of different shapes and materials. In addition, with a total weight of only 180 g, the developed gripper can achieve a maximum grasping payload of 8.8 kg, which is about 49 times of its self-weight. From the methodological point of view, this work has successfully demonstrated the feasibility of optimization-based automatic design of robotic grippers.

Keywords: Flexible Manipulators and Structures, Identification and Estimation in Mechatronics, Mechatronics in Manufacturing Processes
Abstract: Identifying changes in contact during contact-rich manipulation can detect discrete task states or errors, enabling improved robustness and autonomy. The ability to detect contact is affected by the mechatronic design of the robot, especially its physical compliance. Established design methods can design physical compliance for many aspects of contact performance (e.g. peak contact force, motion/force control bandwidth), but are based on time-invariant dynamic models. A change in contact mode is a discrete change in coupled robot-environment dynamics, not easily considered in existing design methods.

Towards robots which can robustly detect changes in contact mode online, this paper investigates how mechatronic design can improve contact estimation, with a focus on designing the location and degree of compliance. A design metric of information gain is proposed which measures how much position/force measurements reduce uncertainty in contact mode estimate. This information gain is developed for fully- and partially-observed systems, as partial observability can arise from joint flexibility in the robot or environmental inertia. Hardware experiments with various compliant setups validate that information gain predicts the speed and certainty with which contact is detected in (i) monitoring of contact-rich assembly and (ii) collision detection.

Keywords: Flexible Manipulators and Structures, Parallel Mechanisms, Fixture and Grasping
Abstract: The design of physical compliance -- its location, degree, and structure -- affects robot performance and robustness in contact-rich tasks. While compliance is often used in the robot's joints, flange, or end-effector, this paper proposes compliant structures in the environment, allowing safe and robust contact while keeping the higher motion control bandwidth and precision of high impedance robots. Compliance is here realized with flexures and viscoelastic materials, which are integrated to several mechanisms to offer structured compliance, such as a remote center of compliance. Additive manufacturing with fused deposition modeling is used, allowing faster design iteration and low-cost integration with standard industrial equipment. Mechanical properties, including the total stiffness matrix, stiffness ratio, and rotational precision, are analytically determined and compared to experimental results. Three remote center of compliance (RCC) devices are prototyped and tested in high-speed assembly tasks.

Keywords: Control Application in Mechatronics, Modeling and Design of Mechatonic Systems, Identification and Estimation in Mechatronics
Abstract: Direct-driven electro-hydraulic systems have a wide range of applications owing to their advantages of energy-savings and relatively high control flexibilities in comparison with classic variable displacement pumpcontrolled hydraulic systems. However, the control accuracy is limited by the inherent nonlinear hydraulic dynamics. Additionally, the pump flow rate may become nonlinear at low pump speeds, causing large pressure-related flow deviations; thereby, limiting the improvement of motion control accuracy. Unfortunately, the pump flow nonlinearity has been ignored or oversimplified without effective modeling in most studies so far. To improve the control accuracy of direct-drive hydraulic systems, a high-precision control strategy must be designed to deal with the nonlinear characteristics and resolve the issue of nonlinear pump flow at low speeds. This article proposes an adaptive robust motion control strategy for a direct-driven electro-hydraulic system with adaptive pump flow rate model compensation. A backstepping integrated direct/indirect adaptive robust controller is designed to deal with the dynamic nonlinearities and uncertainties, which guarantees the stability of the entire hydraulic system. Furthermore, a parameterized polynomial fitting modeling strategy is proposed to accurately describe the nonlinear characteristics of the pump flow rate. Therefore, the uncertain parameters are adjusted in real-time, achieving satisfactory parameter estimations and model compensation for asymptotic motion tracking. Theoretical proof and comparative experiments demonstrate the advantages of the proposed control strategy with adaptive polynomial fitting model compensation for high-precision motion control.

Keywords: Control Application in Mechatronics, Machine Learning, Intelligent Process Automation
Abstract: The productivity of CNC machining and milling in specific is limited by the endurable forces of the working tools. An increased feed velocity also induces higher active forces on the tool, such that the feed velocity has to be maximized within dynamical limits for an optimal operation of the CNC machining center. Model predictive control (MPC) strategies, together with an according target selector enable this kind of optimized operation. The optimization is possible due to models, which describe the feed dynamics and the relationship between the feed and maximum force. However in such a scenario, an accurate modeling and control of the feed dynamics become more crucial, as the built-in routines within the machining center induce unknown nonlinearities. Thus, to achieve the aforementioned objectives, the authors propose a practical nonlinear model predictive control (PNMPC) strategy for milling, incorporating Support Vector Machines (SVM) for both, the target selector and the identification of the unknown nonlinearities. The results show an improved overall control performance with PNMPC, compared to a linear time-varying MPC (LTV-MPC) with a successively linearized model.

Keywords: Control Application in Mechatronics, Artificial Intelligence in Mechatronics, Vehicle Control
Abstract: In this paper we outline some of the numerical heuristics used in existing sample-based MPC techniques and present a generic sample-based MPC algorithm for nonlinear optimal control. Compared to most of the existing techniques our generic algorithm does not place any restrictions on the form of the cost functions and dynamics used in the control problem formulation. We apply the numerical heuristics to the presented algorithm and compare their effectiveness individually by evaluating the control on an autonomous racing environment.

Keywords: Control Application in Mechatronics, Identification and Estimation in Mechatronics
Abstract: For economic reasons, machine builders are increasingly challenged to make single-axis driven machines perform rest-to-rest movements as fast as possible over time. In terms of control, a time-optimal motion is performed by injecting a bang-bang torque profile, characterized by a discrete switching point. However, the state-of-the-art to obtain the ideal application dependent switching point is often computationally demanding and lacks robustness, hampering smooth system implementation. Moreover, machine builders invariably design their machines using CAD software, which automatically provides good knowledge about, e.g., the machine’s load torque profile. Therefore, in this work, based on Newton’s work-energy principle, a framework for variable inertia systems is derived and used as a starting point to estimate the optimal bang-bang switching point efficiently, employing CAD extracted data. In addition, a self-learning control structure is proposed to correct for the initial switching point so that the application continues to move time-optimally, regardless of system influences such as temperature variation. A case study is used to validate the proposed methodology and associated control structure through simulation.

Keywords: Control Application in Mechatronics, Hybrid intelligent systems, Learning and Neural Control in Mechatronics
Abstract: Many advanced methods of process control are based on underlying process models. Therefore, the control performance usually depends strongly on the model accuracy, e.g. the extent of deviations between the real plant behavior and its mathematical representation. Such an advanced method is model predictive control, which has been successfully established in various fields. A model predictive control is based on a sufficiently accurate model of the system to optimize performance and ensure constraints are met. In practice, model descriptions can be subject to significant inaccuracies due to, e.g. insufficient data, unknown dynamics, or system changes. In addition, most processes in industry show time variant dynamics, e.g. due to wear and tear. Modeling such varying characteristics is usually impractical or requires a disproportional amount of modeling effort. Furthermore, the resulting models are usually too complex for real-time process control. In this work, a model predictive control based on a grey-box model is presented. While the model’s basic structure is given by a linearized physical model it is enhanced by a data-driven part using a long short-term memory network. An approach to formulate the linear, time-varying, model predictive control problem that takes into account the model uncertainties of the nominal model predicted by the long short-term memory network model is described. The improvements in control performance are demonstrated on a real demonstrator, which is a coupled tank system. The approach presented reduces the average absolute error by 69.16%, and overshooting by 11.7% on average.

Keywords: Automotive Systems, Transportation Systems, Vehicle Control
Abstract: Advanced driver assistance functions benefit from accurate traffic predictions when optimizing vehicle behavior in the sense of energy consumption minimization. Therefore a velocity prediction is presented that considers live traffic speeds from cloud-based services on a horizon of two kilometers. Using a distributed parameter traffic flow model, an Extended Kalman Filter is implemented to reconstruct the underlying traffic states from uncertain traffic information in combination with the current vehicle speed. Based on the estimated traffic states, a model-based speed prediction is introduced and validated within 150 recorded traffic jam situations from field testing. In comparison with a constant speed prediction, the proposed setup yields an average improvement of 18% over a prediction horizon of 90 seconds. Hence an improved in-vehicle speed prediction is enabled that reveals further potential as cloud-based traffic information accuracy is expected to improve in the coming years.

Keywords: Legged Robots, Robot Dynamics and Control
Abstract: We introduce a unique central pattern generator (CPG) that can synchronize the phases of the legs on quadruped robots without interfering with their oscillation and use it as an alternative to finite state machines in optimal controllers. For all we know, this is the first ever attempt to combine a CPG and an optimal controller in quadruped robots. The proposed central pattern generator is able to eliminate disturbances in gait patterns and maintain the support polygon for the optimal controller to stabilize the robot. With the advantage of both sides, the robot is able to sufficiently increase the chance of passing through randomly generated terrains compared to the robot without the proposed CPG in simulations.

Keywords: Legged Robots, Walking Machines, Modeling and Design of Mechatonic Systems
Abstract: Quadruped robots manifest great potential to traverse rough terrains with payload. Current model-based controllers, which are extensively adopted in quadruped robot locomotion control, rely on accurate estimation of parameters and will significantly deteriorate in severe disturbance, e.g., adding heavy payload. This study introduces an online identification method, which is named as Adaptive Control for Quadruped Locomotion (ACQL), to address model uncertainties. The newly proposed algorithm could achieve estimating and compensating the external disturbances induced by the payload online. The tracking accuracy of the robot's Center of Mass (CoM) and orientation trajectories for the identification task is highly improved. The locomotion task can be incorporated in inverse-dynamics-based Quadratic Programming (QP), realizing a trotting gait. The proposed method is verified in a real quadruped robot platform. Experiments prove the estimation efficacy for the payload weighing from 20 kg to 75 kg and loaded at different locations of the robot's torso.

Keywords: Legged Robots, Robot Dynamics and Control, Control Application in Mechatronics
Abstract: This paper studies the full-body motion generation of a quadruped robot for pace gait. A motion planning algorithm is designed based on the centroidal dynamics of the robot. The motion planning algorithm generates both position and force reference trajectories. These reference trajectories serve as a guide for the swing motion of feet during the swing phase, while they also serve as a guide for the ground contact forces during the stance phase. A hybrid force-motion control framework is constructed using the operational space formulation (OSF) in order to track generated reference trajectories. We contribute further to the OSF of floating-base robots by decoupling the dynamics of the right and left leg pairs to facilitate pace gait. The proposed motion generation method for pace gait is validated using a full-dynamics simulation environment. The results reveal the competence of the proposed whole-body pace gait control for a quadruped robot.

Keywords: Walking Machines, Identification and Estimation in Mechatronics, Legged Robots
Abstract: This paper aims at sensorless observation of contact force and friction for the heavy-legged robot (HLR) driven by electric cylinders. It is known that the acquisition of contact force is challenging but significant for HLR due to the high self-weight. Additionally, as the main driving device, the electric cylinder contains considerable friction and disturbances, which greatly influence the accuracy of observed contact force. As a significant novelty, a method for contact force and friction observation through permanent magnet synchronous motor (PMSM) load torque observer is proposed in this paper. The proposed approximate PMSM model containing dynamics of PMSM, the electric cylinder, and the HLR is established. The motor speed is utilized to observe the friction disturbance and the external torque based on the sliding mode observer (SMO) theory. Centroidal dynamics and the information of inertial measurement unit (IMU) are introduced to distinguish the contact force included in the observation result of the proposed model. The entire algorithm is validated in the MATLAB-SIMSCAPE environment, and the observation results show that this algorithm is reliable and accurate in the utilization of HLR.

Keywords: Humanoid Robots, Legged Robots, Robot Dynamics and Control
Abstract: Robust locomotion is a challenging task for humanoid robots, especially when considering dynamic disturbances. This paper proposes a disturbance observer-based cascaded model predictive control (MPC) approach for bipedal locomotion, with the capability of exploiting ankle, stepping, hip and height variation strategies. Specifically, based on the variable-height inverted pendulum model, a nonlinear MPC that is run at a low frequency is built for 3D locomotion (i.e., with height variation) while accounting for the footstep modulation as well. Differing from previous works, the nonlinear MPC is formulated as a convex optimization problem by semidefinite relaxation. Subsequently, assuming a flywheel at the pelvis center, a linear MPC that is run at a high frequency is proposed to regulate angular momentum (e.g., through rotating the upper body), which is solved by convex quadratic programming. To run the cascaded MPC in a closed-loop manner, a high order sliding mode observer (HOSMO) is designed to estimate system states and dynamic disturbances simultaneously. Simulation and hardware experiments demonstrate the walking robustness in real-world scenarios, including 3D walking with varying speeds, walking across non-coplanar terrains and push recovery.