Descriptive Model-based Learning and Control for Bipedal Locomotion

TL;DR

This paper introduces a novel control framework for bipedal robots that uses a minimal descriptive model to maintain balance, enabling more natural and robust walking by allowing high-dimensional motion to evolve freely.

Contribution

It proposes a new control approach that avoids prescribing a full low-dimensional model, instead using a minimal descriptive model to improve gait efficiency and robustness.

Findings

Achieves more human-like walking gait.

Enhances robustness of bipedal balance.

Allows high-dimensional motion to evolve freely.

Abstract

Bipedal balance is challenging due to its multi-phase, hybrid nature and high-dimensional state space. Traditional balance control approaches for bipedal robots rely on low-dimensional models for locomotion planning and reactive control, constraining the full robot to behave like these simplified models. This involves tracking preset reference paths for the Center of Mass and upper body obtained through low-dimensional models, often resulting in inefficient walking patterns with bent knees. However, we observe that bipedal balance is inherently low-dimensional and can be effectively described with simple state and action descriptors in a low-dimensional state space. This allows the robot's motion to evolve freely in its high-dimensional state space, only constraining its projection in the low-dimensional state space. In this work, we propose a novel control approach that avoids prescribing a low-dimensional model to the full model. Instead, our control framework uses a descriptive model with the minimum degrees of freedom necessary to maintain balance, allowing the remaining degrees of freedom to evolve freely in the high-dimensional space. This results in an efficient human-like walking gait and improved robustness.