Animals have remarkable abilities to adapt locomotion to different terrains and tasks. However, robots trained by means of reinforcement learning are typically able to solve only a single task and a transferred policy is usually inferior to that trained from scratch. In this work, we demonstrate that meta-reinforcement learning can be used to successfully train a robot capable to solve a wide range of locomotion tasks. The performance of the meta-trained robot is similar to that of a robot that is trained on a single task.
@article{arxiv.2107.03741,
title = {Adaptation of Quadruped Robot Locomotion with Meta-Learning},
author = {Arsen Kuzhamuratov and Dmitry Sorokin and Alexander Ulanov and A. I. Lvovsky},
journal= {arXiv preprint arXiv:2107.03741},
year = {2021}
}