Online inverse reinforcement learning for nonlinear systems
Systems and Control
2018-11-27 v1
Abstract
This paper focuses on the development of an online inverse reinforcement learning (IRL) technique for a class of nonlinear systems. The developed approach utilizes observed state and input trajectories, and determines the unknown cost function and the unknown value function online. A parameter estimation technique is utilized to allow the developed IRL technique to determine the cost function weights in the presence of unknown dynamics. Simulation results are presented for a nonlinear system showing convergence of both unknown reward function weights and unknown dynamics.
Keywords
Cite
@article{arxiv.1811.10471,
title = {Online inverse reinforcement learning for nonlinear systems},
author = {Ryan Self and Michael Harlan and Rushikesh Kamalapurkar},
journal= {arXiv preprint arXiv:1811.10471},
year = {2018}
}
Comments
arXiv admin note: text overlap with arXiv:1801.07663