We propose a model-free algorithm for learning efficient policies capable of returning table tennis balls by controlling robot joints at a rate of 100Hz. We demonstrate that evolutionary search (ES) methods acting on CNN-based policy architectures for non-visual inputs and convolving across time learn compact controllers leading to smooth motions. Furthermore, we show that with appropriately tuned curriculum learning on the task and rewards, policies are capable of developing multi-modal styles, specifically forehand and backhand stroke, whilst achieving 80\% return rate on a wide range of ball throws. We observe that multi-modality does not require any architectural priors, such as multi-head architectures or hierarchical policies.
@article{arxiv.2003.14398,
title = {Robotic Table Tennis with Model-Free Reinforcement Learning},
author = {Wenbo Gao and Laura Graesser and Krzysztof Choromanski and Xingyou Song and Nevena Lazic and Pannag Sanketi and Vikas Sindhwani and Navdeep Jaitly},
journal= {arXiv preprint arXiv:2003.14398},
year = {2020}
}
Comments
V2: new URL of supplementary video. 8 pages, 4 figures