Collective Robot Navigation Using Diffusion Limited Aggregation

Many applications of swarm robotics require autonomous navigation in unknown environments. We describe a new collective navigation strategy based on diffusion limited aggregation and bacterial foraging behaviour. Both methods are suitable for typical swarm robots as they require only minimal sensory and control capabilities. We demonstrate the usefulness of the strategy with a swarm that is capable of autonomously finding charging stations and show that the collective search can be significantly more effective than individual-based search.