Imitation Learning for Variable Speed Contact Motion for Operation up to Control Bandwidth

Robots have several requirements, including environmental adaptability, to operate in the real-world environment. Moreover, the desired success rate for task completion must be achieved. In this regard, end-to-end learning for autonomous operation is currently being investigated. However, the issue of the operating speed has not been investigated in detail. Therefore, in this study, we propose a method for generating variable operating speeds while adapting to perturbations in the environment. When the work speed changes, a nonlinear relationship occurs between the operating speed and force (e.g., inertial and frictional forces). However, the proposed method can be adapted to nonlinearities by utilizing a small amount of motion data. We experimentally evaluated the proposed method by erasing a line using an eraser fixed to the tip of the robot. Furthermore, the proposed method enables a robot to perform a task faster than a human operator and is capable of operating at the control bandwidth. INDEX TERMS Imitation learning, bilateral control, motion planning, fast-forward, machine learning

[1]  Mamoru Mitsuishi,et al.  Online Trajectory Planning and Force Control for Automation of Surgical Tasks , 2018, IEEE Transactions on Automation Science and Engineering.

[2]  Kouhei Ohnishi,et al.  A Novel Motion Equation for General Task Description and Analysis of Mobile-Hapto , 2013, IEEE Transactions on Industrial Electronics.

[3]  Makoto Iwasaki,et al.  Sensorless Torsion Control of Elastic-Joint Robots With Hysteresis and Friction , 2016, IEEE Transactions on Industrial Electronics.

[4]  Tie Zhang,et al.  Robotic constant-force grinding control with a press-and-release model and model-based reinforcement learning , 2020 .

[5]  Aude Billard,et al.  On Learning, Representing, and Generalizing a Task in a Humanoid Robot , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[6]  Ken Goldberg,et al.  Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperation , 2017, ICRA.

[7]  Zhou-Ping Yin,et al.  Hand–Eye Calibration in Visually-Guided Robot Grinding , 2016, IEEE Transactions on Cybernetics.

[8]  Kouhei Ohnishi,et al.  Motion control for advanced mechatronics , 1996 .

[9]  Toshiaki Tsuji,et al.  Motion Generation Using Bilateral Control-Based Imitation Learning With Autoregressive Learning , 2021, IEEE Access.

[10]  Sergey Levine,et al.  Learning force-based manipulation of deformable objects from multiple demonstrations , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[11]  Maria Kyrarini,et al.  Robot learning of industrial assembly task via human demonstrations , 2019, Auton. Robots.

[12]  Kouhei Ohnishi,et al.  Multi-DOF Micro-Macro Bilateral Controller Using Oblique Coordinate Control , 2011, IEEE Transactions on Industrial Informatics.

[13]  Y Hori,et al.  Robust Tracking Controller Design With Uncertain Friction Compensation Based on a Local Modeling Approach , 2010, IEEE/ASME Transactions on Mechatronics.

[14]  Hermann Ney,et al.  From Feedforward to Recurrent LSTM Neural Networks for Language Modeling , 2015, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[15]  Marco Forgione,et al.  Robot control parameters auto-tuning in trajectory tracking applications , 2020 .

[16]  Toshiyuki Murakami,et al.  Torque sensorless control in multidegree-of-freedom manipulator , 1993, IEEE Trans. Ind. Electron..

[17]  Kensuke Harada,et al.  Deep Learning Scooping Motion Using Bilateral Teleoperations , 2018, 2018 3rd International Conference on Advanced Robotics and Mechatronics (ICARM).

[18]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[19]  Francesco Braghin,et al.  Learning Continuous Control Actions for Robotic Grasping with Reinforcement Learning , 2020, 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[20]  S. Katsura,et al.  Motion copying system based on real-world haptics in variable speed , 2008, 2008 13th International Power Electronics and Motion Control Conference.

[21]  Toshiaki Tsuji,et al.  Imitation Learning Based on Bilateral Control for Human–Robot Cooperation , 2020, IEEE Robotics and Automation Letters.

[22]  Toshiaki Tsuji,et al.  Estimation and Kinetic Modeling of Human Arm using Wearable Robot Arm , 2017 .

[23]  Shigeki Sugano,et al.  Repeatable Folding Task by Humanoid Robot Worker Using Deep Learning , 2017, IEEE Robotics and Automation Letters.

[24]  Martin Jägersand,et al.  A Geometric Perspective on Visual Imitation Learning , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[25]  Zhijun Zhang,et al.  Three Recurrent Neural Networks and Three Numerical Methods for Solving a Repetitive Motion Planning Scheme of Redundant Robot Manipulators , 2017, IEEE/ASME Transactions on Mechatronics.

[26]  Toshiaki Tsuji,et al.  Time Series Motion Generation Considering Long Short-Term Motion , 2019, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[27]  You Zhou,et al.  Vision-Based Online Adaptation of Motion Primitives to Dynamic Surfaces: Application to an Interactive Robotic Wiping Task , 2018, IEEE Robotics and Automation Letters.

[28]  Toshiaki Tsuji,et al.  Dynamic Object Manipulation Considering Contact Condition of Robot With Tool , 2016, IEEE Transactions on Industrial Electronics.

[29]  Sergey Levine,et al.  Residual Reinforcement Learning for Robot Control , 2018, 2019 International Conference on Robotics and Automation (ICRA).

[30]  Masayoshi Tomizuka,et al.  Learning Variable Impedance Control via Inverse Reinforcement Learning for Force-Related Tasks , 2021, IEEE Robotics and Automation Letters.

[31]  Sergey Levine,et al.  Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning , 2019, CoRL.

[32]  Darwin G. Caldwell,et al.  Imitation Learning of Positional and Force Skills Demonstrated via Kinesthetic Teaching and Haptic Input , 2011, Adv. Robotics.

[33]  Carme Torras,et al.  A robot learning from demonstration framework to perform force-based manipulation tasks , 2013, Intelligent Service Robotics.

[34]  Chen Lu,et al.  Surface Following using Deep Reinforcement Learning and a GelSightTactile Sensor , 2019, ArXiv.

[35]  Cristian Alejandro Vergara Perico,et al.  Learning robust manipulation tasks involving contact using trajectory parameterized probabilistic principal component analysis , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[36]  Toshiaki Tsuji,et al.  Reinforcement Learning for Robotic Assembly Using Non-Diagonal Stiffness Matrix , 2021, IEEE Robotics and Automation Letters.

[37]  Yu Zhang,et al.  Highway long short-term memory RNNS for distant speech recognition , 2015, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[38]  Leonel Rozo.  Interactive Trajectory Adaptation through Force-guided Bayesian Optimization , 2019, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[39]  Toshiaki Tsuji,et al.  Bilateral Control Between Electric and Hydraulic Actuators Using Linearization of Hydraulic Actuators , 2017, IEEE Transactions on Industrial Electronics.

[40]  Jun Tani,et al.  Self-organization of behavioral primitives as multiple attractor dynamics: A robot experiment , 2003, IEEE Trans. Syst. Man Cybern. Part A.

[41]  Danwei Wang,et al.  A High-Bandwidth End-Effector With Active Force Control for Robotic Polishing , 2020, IEEE Access.

[42]  Tsuyoshi Adachi,et al.  Imitation Learning for Object Manipulation Based on Position/Force Information Using Bilateral Control , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[43]  Rouhollah Rahmatizadeh,et al.  From Virtual Demonstration to Real-World Manipulation Using LSTM and MDN , 2016, AAAI.

[44]  Darwin G. Caldwell,et al.  Learning optimal controllers in human-robot cooperative transportation tasks with position and force constraints , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).