L-systems driven self-reconfiguration of modular robots

Abstract

In the domain of modular self-reconfigurable robotic systems, self-reconfiguration is known to be a highly challenging task. This article presents a novel algorithm for distributed self-reconfiguration by combining cellular automata and L-systems. Cellular automata is used to handle the relative motion planning of decentralized modules. L-systems are introduced to provide a topological description for the target configuration. The turtle interpretation is extended to modular robotics to generate local predictions for distributed modules from global description. Local predictions spread out in the system through gradient propagation. Modules, using cellular automata rules managing local motion, climb gradient to the expanding fronts for constructing global configurations. Both simulations and experiments have demonstrated the practical effectiveness of the proposed algorithm.