로봇 매니퓰레이터의 동적 모델 학습을 위한 순환 신경망 기반 학습 과정
Recurrent Neural Network-based Learning Process of Dynamic Model in Robot Manipulators
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This paper proposes a new learning process for dynamic model of robot manipulators, which is based on a recurrence neural network (RNN) algorithm with feedback structure. For a model-learning process, we first chose well-known activity functions that Sigmoid, ReLU, and Hyperbolic tangent functions were employed to confirm the suitability of the dynamic model in the RNN. Next, to make input/output variables suitable for closed-loop systems, they were set as joint angles, angular velocities, and torques for each axis of the virtual robot manipulators. Then, these variables are directly related to reference trajectories that have several changes in moving speed of end-effector while drawing circles with continuously changing radius in task space. The effectiveness of the learned dynamic model were proved by using some of collected variables. Also, it were demonstrated to verify the feasibility of the proposed process in a closed-loop robot manipulator simulation, which were compared to tracking performance in a virtual robot manipulator.
Abstract
This paper proposes a new learning process for dynamic model of robot manipulators, which is based on a recurrence neural network (RNN) algorithm with feedback structure. For a model-learning process, we first chose well-known activity functions that Sigmoid, ReLU, and Hyperbolic tangent functions were employed to confirm the suitability of the dynamic model in the RNN. Next, to make input/output variables suitable for closed-loop systems, they were set as joint angles, angular velocities, and torques for each axis of the virtual robot manipulators. Then, these variables are directly related to reference trajectories that have several changes in moving speed of end-effector while drawing circles with continuously changing radius in task space. The effectiveness of the learned dynamic model were proved by using some of collected variables. Also, it were demonstrated to verify the feasibility of the proposed process in a closed-loop robot manipulator simulation, which were compared to tracking performance in a virtual robot manipulator.
참고문헌 (33)
- J. Qin, "Real-time Trajectory Compensation in Robotic Friction Stir Welding Using State Estimators", IEEE Transactions on Control Systems Technology, 24(6), pp.2207~2214 (2016)[DOI]
- H. Park, "Compliance-based Robotic Peg-in-hole Assembly Strategy Without Force Feedback", IEEE Transactions on Industrial Electronics, 64(8), pp.6299~6309 (2017)[DOI]
- X. Yu, "Adaptive-constrained Impedance Control for Human-robot Co-transportation", IEEE Transactions on Cybernetics, 52(12), pp.13237~13249 (2022)[DOI]
- F. Santoso, "Trusted Operations of a Military Ground Robot in the Face of Man-in-the-Middle Cyber-Attacks Using Deep Learning Convolutional Neural Networks : Real-Time Experimental Outcomes", IEEE Transactions on Dependable and Secure Computing, pp.1~12 (2023)[DOI]
- M. Cai, "Grasping Marine Products with Hybrid-driven Underwater Vehicle-manipulator System", IEEE Transactions on Automation Science and Engineering, 17(3), pp.1443~1454 (2020)[DOI]
- F. Aghili, "Robust Impedance-matching of Manipulators Interacting with Uncertain Environments : Application to Task Verification of the Space Station’s Dexterous Manipulator", IEEE, 2019, pp.24(4), 1565~1576
- J. Baek, "A Synthesized Sliding-mode Control for Attitude Trajectory Tracking of Quadrotor UAV Systems", IEEE, 2023, pp.28(4), 2189~2199
- M. Kobayashi, "BSL : Navigation Method Considering Blind Spots Based on ROS Navigation Stack and Blind Spots Layer for Mobile Robot", IEEE Transactions on Industry Applications, 60(1), pp.1695~1704 (2024)[DOI]
- H. -C. Huang, "Evolutionary Optimization of Fuzzy Reinforcement Learning and its Application to Time-Varying Tracking Control of Industrial Parallel Robotic Manipulators", IEEE Transactions on Industrial Informatics, 19(12), pp.11712~11720 (2023)[DOI]
- W. Li, "Force Sensing and Compliance Control for a Cable-driven Redundant Manipulator", IEEE, 2024, pp.29(1), 777~788
- Z. Xie, "A Bi-criteria Kinematic Strategy for Motion", Force Control of Robotic Manipulator, 2023, ~
- J. Baek, "A New Adaptive Sliding-mode Control Scheme for Application to Robot Manipulators", IEEE Transactions on Industrial Electronics, 63(6), pp.3628~3637 (2016)[DOI]
- J. Baek, "Practical Time-delay Control with Adaptive Gains for Trajectory Tracking of Robot Manipulators", IEEE Transactions on Industrial Electronics, 65(7), pp.5682~5692 (2018)[DOI]
- J. Baek, "A Widely Adaptive Time-delayed Control and its Application to Robot Manipulators", IEEE Transactions on Industrial Electronics, 66(7), pp.5332~5342 (2019)[DOI]
- J. Baek, "A New Widely and Stably Adaptive Sliding-mode Control with Nonsingular Terminal Sliding Variable for Robot Manipulators", IEEE Access, 8, pp.43443~43454 (2020)[DOI]
- J. Duan, "On the Optimization Landscape of Dynamic Output Feedback Linear Quadratic Control", IEEE Transactions on Automatic Control, 69(2), pp.920~935 (2024)[DOI]
- M. Makarov, "Modeling and Preview H∞ Control Design for Motion Control of Elastic-joint Robots with Uncertainties", IEEE Transactions on Industrial Electronics, 63(10), pp.6429~6438 (2016)[DOI]
- R. H. Middletone, "Adaptive Computed Torque Control for Rigid Link Manipulators", 1986 25th IEEE Conference on Decision and Control, pp.68~73 (1986)
- "Multilayer Perceptrons, Handbook of Neural Computation" (1997)[URL]
- D. Cagigas-Muñiz, "Artificial Neural Networks for Inverse Kinematics Problem in Articulated Robots", Engineering Applications of Artificial Intelligence, 126 (2023)[DOI]
- M. Defdaf, "A New Transform Discrete Wavelet Technique Based on Artificial Neural Network for Induction Motor Broken Rotor Bar Faults Diagnosis", International Transactions on Electrical Energy Systems, 31(4), pp.e12807~ (2021)[DOI]
- A. Eskandarian, "Dynamics Modeling of Robotic Manipulators Using An Artificial Neural Network", Journal of Robotic Systems, 11(1), pp.41~56 (1993)[DOI]
- M. Schuster, "Bidirectional Recurrent Neural Networks", IEEE Transactions on Signal Processing, 45(11), pp.2673~2681 (1997)[DOI]
- L. Jin, "Dynamic Recurrent Neural Networks for Modeling Flexible Robot Dynamics", Proceedings of the 10th IEEE International Symposium on Intelligent Control, pp.105~110 (1995)
- Y. Bengio, "The Problem of Learning Long-term Dependencies in Recurrent Networks", IEEE International Conference on Neural Networks, pp.1183~1188 (1993)
- M. Hüsken, "Recurrent Neural Networks for Time Series Classification", Neurocomputing, 50, pp.223~235 (2003)[DOI]
- A. De Myttenaere, "Mean Absolute Percentage Error for Regression Models", Neurocomputing, 192, pp.38~48 (2015)[DOI]
- G. Cohen, "Spatial and Temporal Downsampling in Eventbased Visual Classification", IEEE Transactions on Neural Networks and Learning Systems, 29(10), pp.5030~5044 (2018)[DOI]
- Y. Yu, "Bshift : a Low Cost Deep Neural Networks Accelerator", International Journal of Parallel Programming, 47, pp.360~372 (2019)[DOI]
- J. Bergstra, "Random search for hyper-parameter optimization", Journal of machine learning research, 13(2), pp.281~305 (2012)
- L. Yang, "On hyperparameter optimization of machine learning algorithms : Theory and practice", Neurocomputing, 415, pp.295~316 (2020)[DOI]
- M. W. Spong, "Dynamics,”Robot modeling and control", pp.198~208 (2020), John Wiley & Sons