We present task-oriented Koopman-based control that utilizes end-to-end reinforcement learning and contrastive encoder to simultaneously learn the Koopman latent embedding, operator, and associated linear controller within an iterative loop. By prioritizing the task cost as the main objective for controller learning, we reduce the reliance of controller design on a well-identified model, which, for the first time to the best of our knowledge, extends Koopman control from low to high-dimensional, complex nonlinear systems, including pixel-based tasks and a real robot with lidar observations. Code and videos are available \href{https://sites.google.com/view/kpmlilatsupp/}{here}.
Cite
@article{arxiv.2309.16077,
title = {Task-Oriented Koopman-Based Control with Contrastive Encoder},
author = {Xubo Lyu and Hanyang Hu and Seth Siriya and Ye Pu and Mo Chen},
journal= {arXiv preprint arXiv:2309.16077},
year = {2023}
}
Comments
Accepted by the 7th Annual Conference on Robot Learning (CoRL), 2023 (oral spotlight)