Robot Policy Improvement With Natural Evolution Strategies for Stable Nonlinear Dynamical System

• Published in: IEEE transactions on cybernetics Vol. 53; no. 6; pp. 4002 - 4014
• Main Authors: Hu, Yingbai, Chen, Guang, Li, Zhijun, Knoll, Alois
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.06.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Cloning; Dynamic stability; Dynamical system; Dynamical systems; Evolution; exponential natural evolution strategies (NESs); Heuristic algorithms; imitation learning; Impedance; Motion planning; Motion stability; Obstacle avoidance; Parameters; policy improvement of robustness and adaptability; Probabilistic models; Robot dynamics; Robots; Robustness; Stability analysis; Stability criteria; Stiffness; Supervised learning; Task analysis; Task complexity

Robot learning through kinesthetic teaching is a promising way of cloning human behaviors, but it has its limits in the performance of complex tasks with small amounts of data, due to compounding errors. In order to improve the robustness and adaptability of imitation learning, a hierarchical learni...