Severe Damage Recovery in Evolving Soft Robots through Differentiable Programming

TL;DR

This paper introduces a neural cellular automata system enabling evolved soft robots to regenerate over 80% of their functionality after severe damage, combining evolution and gradient-based training for robustness.

Contribution

It presents a novel approach that integrates evolution and differentiable programming to enable damage recovery in soft robots.

Findings

Robots can regain over 80% functionality after damage

Neural cellular automata effectively grow resilient robot morphologies

Gradient-based training enhances robustness in evolved robots

Abstract

Biological systems are very robust to morphological damage, but artificial systems (robots) are currently not. In this paper we present a system based on neural cellular automata, in which locomoting robots are evolved and then given the ability to regenerate their morphology from damage through gradient-based training. Our approach thus combines the benefits of evolution to discover a wide range of different robot morphologies, with the efficiency of supervised training for robustness through differentiable update rules. The resulting neural cellular automata are able to grow virtual robots capable of regaining more than 80\% of their functionality, even after severe types of morphological damage.