A Population-Level Analysis of Neural Dynamics in Robust Legged Robots

TL;DR

This study applies computational neuroscience methods to analyze neural population dynamics in robust legged robots, revealing differences in stability and reliance on sensory input between fragile and robust controllers.

Contribution

It introduces a novel approach to understanding robot locomotion controllers by analyzing their population-level neural activity and topological structure.

Findings

Fragile controllers have more fixed points with unstable directions.

Recurrent state dynamics are low-dimensional during walking.

Fragile agents rely more on sensory input when perturbed.

Abstract

Recurrent neural network-based reinforcement learning systems are capable of complex motor control tasks such as locomotion and manipulation, however, much of their underlying mechanisms still remain difficult to interpret. Our aim is to leverage computational neuroscience methodologies to understanding the population-level activity of robust robot locomotion controllers. Our investigation begins by analyzing topological structure, discovering that fragile controllers have a higher number of fixed points with unstable directions, resulting in poorer balance when instructed to stand in place. Next, we analyze the forced response of the system by applying targeted neural perturbations along directions of dominant population-level activity. We find evidence that recurrent state dynamics are structured and low-dimensional during walking, which aligns with primate studies. Additionally, when recurrent states are perturbed to zero, fragile agents continue to walk, which is indicative of a stronger reliance on sensory input and weaker recurrence.