Without relying on any privileged or visual information, we can train a robust locomotion policy. In real-world deployment, the quadruped robot is capable of traversing various challenging terrains.
Abstract
The recent mainstream reinforcement learning control for quadruped robots often relies on privileged information, demanding meticulous selection and precise estimation, thereby imposing constraints on the development process. This work proposes a Self-learning Latent Representation (SLR) method, which achieves high-performance control policy learning without the need for privileged information. To enhance the credibility of the proposed method's evaluation, SLR was directly compared with state-of-the-art algorithms using their open-source code repositories and original configuration parameters. Remarkably, SLR surpasses the performance of previous methods using only limited proprioceptive data, demonstrating significant potential for future applications. Ultimately, the trained policy and encoder empower the quadruped robot to traverse various challenging terrains.
Training Framework
The SLR training framework leverages the Markov Decision Process, guiding the latent's self-learning based on state transitions (transition model), state distinctions (random sampling), and cumulative rewards (critic), without relying on manually set privileged information constraints.