Learning Stable Manoeuvres in Quadruped Robots from Expert Demonstrations

TL;DR

This paper presents a neural network-based method for generating stable leg trajectories in quadruped robots, enabling smooth transitions between velocities and turning radii by learning from expert demonstrations, with less training data needed.

Contribution

It introduces a two-stage approach combining discrete policies and a neural filter for smooth trajectory transitions, improving stability and reducing demonstration requirements.

Findings

Effective in generating stable maneuvers in quadruped robots

Enables novice users to perform expert-level trajectories

Requires fewer expert demonstrations than standard methods

Abstract

With the research into development of quadruped robots picking up pace, learning based techniques are being explored for developing locomotion controllers for such robots. A key problem is to generate leg trajectories for continuously varying target linear and angular velocities, in a stable manner. In this paper, we propose a two pronged approach to address this problem. First, multiple simpler policies are trained to generate trajectories for a discrete set of target velocities and turning radius. These policies are then augmented using a higher level neural network for handling the transition between the learned trajectories. Specifically, we develop a neural network-based filter that takes in target velocity, radius and transforms them into new commands that enable smooth transitions to the new trajectory. This transformation is achieved by learning from expert demonstrations. An application of this is the transformation of a novice user's input into an expert user's input, thereby ensuring stable manoeuvres regardless of the user's experience. Training our proposed architecture requires much less expert demonstrations compared to standard neural network architectures. Finally, we demonstrate experimentally these results in the in-house quadruped Stoch 2.