A distributed and parallel control mechanism for self-reconfiguration of modular robots using L-systems and cellular automata

For distributed self-reconfiguration of Modular Self-Reconfigurable (MSR) robots, one of the main difficulties is the contradiction between limited information of decentralized modules and well-organized global structure. This paper presents a distributed and parallel mechanism for decentralized self-reconfiguration of MSR robots. This mechanism is hybrid by combining Lindenmayer systems (L-systems) describing the topological structure as configuration target and Cellular Automata (CA) for local motion planning of individual modules. Turtle interpretations are extended to modular robotics for generating module-level predictions from global description. According to local information, independent modules make motion planning by Cellular Automata in parallel. This distributed mechanism is robust to failure of modules, scalable to varying module numbers, and convergent to predefined reconfiguration targets. Simulations and statistical results are provided for validating the proposed algorithm.