A minimal approach to distributed control for soft robots.

One of the main challenges in robotics is developing systems that can adapt to their environment and achieve autonomous behavior.

Current approaches typically aim to achieve this by increasing the complexity of centralized controllers, unlike biological systems, where complex and adaptive behavior often emerges from simple local interactions with the environment. The current research designs and analyzes a decentralized control strategy that harnesses emergence to create adaptive control with minimal resources. This control strategy's properties are investigated using a modular robotic system consisting of coupled computational units with limited memory. Each executes a fully decentralized, local, and asynchronous algorithm to solve the system-wide problem of moving to the light (phototaxis). With this research, we aim to explore how distributing a robot's "brain" over its body could achieve more robust behavior during a locomotion task.