Task-Oriented Koopman-Based Control with Contrastive Encoder

Xubo Lyu, Hanyang Hu, Seth Siriya, Ye Pu, Mo Chen

Keywords: Learning and control, Koopman-based control, Represention learning

TL;DR: 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

Abstract: 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: https://sites.google.com/view/kpmlilatsupp/.

Student First Author: yes

Instructions: I have read the instructions for authors (https://corl2023.org/instructions-for-authors/)

Video: https://sites.google.com/view/kpmlilatsupp/

Website: https://sites.google.com/view/kpmlilatsupp/

Code: https://sites.google.com/view/kpmlilatsupp/