[ICRA2020 video] Motor Synergy Development in High-performing Deep Reinforcement Learning algorithms
Motor Synergy Development in High-performing Deep Reinforcement Learning algorithms, IEEE Robotics and Automation Letters, April 2020, 5(2):1271-1278, Jiazheng Chai, M. Hayashibe, (ICRA2020 Paris)
URL: https://ieeexplore.ieee.org/document/8966298
As human motor learning is hypothesized to use the motor synergy concept, we investigate if this concept could also be observed in deep reinforcement learning for robotics. From this point of view, we carried out a joint-space synergy analysis on multi-joint running agents in simulated environments trained using two state-of-the-art deep reinforcement learning algorithms. Although a synergy constraint has never been encoded into the reward function, the synergy emergence phenomenon could be observed statistically in the learning agent. To our knowledge, this is the first attempt to quantify the synergy development in detail and evaluate its emergence process during deep learning motor control tasks. We then demonstrate that there is a correlation between our synergy-related metrics and the performance and energy efficiency of a trained agent. Interestingly, the proposed synergy-related metrics reflected a better learning capability of SAC over TD3. It suggests that these metrics could be additional new indices to evaluate deep reinforcement learning algorithms for motor learning. It also indicates that synergy is required for multi-joints robots to move energy-efficiently.
Видео [ICRA2020 video] Motor Synergy Development in High-performing Deep Reinforcement Learning algorithms канала Neuro-Robotics Lab
URL: https://ieeexplore.ieee.org/document/8966298
As human motor learning is hypothesized to use the motor synergy concept, we investigate if this concept could also be observed in deep reinforcement learning for robotics. From this point of view, we carried out a joint-space synergy analysis on multi-joint running agents in simulated environments trained using two state-of-the-art deep reinforcement learning algorithms. Although a synergy constraint has never been encoded into the reward function, the synergy emergence phenomenon could be observed statistically in the learning agent. To our knowledge, this is the first attempt to quantify the synergy development in detail and evaluate its emergence process during deep learning motor control tasks. We then demonstrate that there is a correlation between our synergy-related metrics and the performance and energy efficiency of a trained agent. Interestingly, the proposed synergy-related metrics reflected a better learning capability of SAC over TD3. It suggests that these metrics could be additional new indices to evaluate deep reinforcement learning algorithms for motor learning. It also indicates that synergy is required for multi-joints robots to move energy-efficiently.
Видео [ICRA2020 video] Motor Synergy Development in High-performing Deep Reinforcement Learning algorithms канала Neuro-Robotics Lab
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Tower of Hanoi (Occupational Therapy Task) with Microsoft Hololens 2ニューロロボティクス、東北大学 学祭模擬講義 2021Towards Robust Wheel-Legged Biped Robot System: Combining Feedforward and Feedback Control[EMBC2020] Personalized Fall Risk Estimation with Kinect TwoIndividual deformability compensation of soft hydraulic actuators through iterative learning[EMBC21] Deep Reinforcement Learning with Gait Mode Specification for Trot-Gallop AnalysisAI-CPG: Adaptive Imitated Central Pattern GeneratorsMotor Synergy Development in Symmetric Gait ofWhole-body Locomotion Learning (ICRA2023 video)[ICRA2020] Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian反復学習制御によるFiber-Reinforced Soft Actuatorの個体差補償Synergetic synchronized oscillation by distributed neural integratorsJoint Elasticity Produces Energy Efficiency in Underwater LocomotionSpiking Neural Network Discovers Energy-Efficient Hexapod Motion in Deep Reinforcement Learning[EMBC2011] Joint Angle Estimation with Inertial Sensors and its Integration with Kinect[EMBC2012] Real-time Simulation of Volumetric Muscle Deformation with EMG signals[ICRA21] Deep Reinforcement Learning Framework for Underwater Locomotion of Soft RobotSynergetic Learning Control Paradigm for Redundant Robot to Enhance Error-Energy IndexElectromyographically Controlled Snake RobotSelf-Organizing Neural Network for Reproducing Human Postural Mode Alternation through LearningSimultaneous On-line Motion Discrimination and Evaluation of Whole-body Exercise by Synergy Probes