Stochastic Optimal Control for Robot Manipulation Skill Learning Under Time-Varying Uncertain Environment
- 18 October 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 54 (4) , 2015-2025
Abstract
In this article, a novel stochastic optimal control method is developed for robot manipulator interacting with a time-varying uncertain environment. The unknown environment model is described as a nonlinear system with time-varying parameters as well as stochastic information, which is learned via the Gaussian process regression (GPR) method as the external dynamics. Integrating the learned external dynamics as well as the stochastic uncertainties, the complete interaction system dynamics are obtained. Then the iterative linear quadratic Gaussian with learned external dynamics (ILQG-LEDs) method is presented to obtain the optimal manipulation control parameters, namely, the feedforward force, the reference trajectory, as well as the impedance parameters, subject to time-varying environment dynamics. The comparative simulation studies verify the advantages of the presented method, and the experimental studies of the peg-hole-insertion task prove that this method can deal with complex manipulation tasks.Keywords
Funding Information
- National Natural Science Foundation of China (61725303, 62103334)
- China Postdoctoral Science Foundation (2021M702669)
This publication has 27 references indexed in Scilit:
- Admittance-Based Controller Design for Physical Human–Robot Interaction in the Constrained Task SpaceIEEE Transactions on Automation Science and Engineering, 2020
- A sensorimotor reinforcement learning framework for physical Human-Robot Interaction2016
- Safe and robust learning control with Gaussian processes2015
- Impedance adaptation for optimal robot–environment interactionInternational Journal of Control, 2013
- Model-Free Reinforcement Learning of Impedance Control in Stochastic EnvironmentsIEEE Transactions on Autonomous Mental Development, 2012
- Variable admittance control of a four-degree-of-freedom intelligent assist device2012
- Learning motion primitive goals for robust manipulation2011
- Reinforcement learning of impedance control in stochastic force fields2011
- Adaptive Optimal Feedback Control with Learned Internal Dynamics Models2010
- A generalized iterative LQG method for locally-optimal feedback control of constrained nonlinear stochastic systems2005