Mapping of unknown environments using minimal sensing from a stochastic swarm

Swarms consisting of cyborg-insects or millirobots can be used for mapping and exploration of unstructured environments in emergency-response situations. Under extreme conditions, traditional localization techniques may fail to provide reliable position estimates. Instead, we propose a robust approach to obtain a topological map of an unknown environment using encounter information from a swarm of agents following a stochastic motion model via the use of tools from topological data analysis. A classification approach is introduced to determine the persistent topology features of the space. The approach is analyzed using simulation and experimental data using a swarm robotic platform called the WolfBot. For all experiments, the agents are programmed to follow a stochastic motion model and only rely on encounter information between agents to construct a map of the environment. The results indicate that the proposed approach can identify robust topological features with high accuracy.