Foraging Algorithms for Swarm Robotics

Abstract

Foraging is a benchmark problem for swarm robotics. It is inspired by swarms of insects cooperating to locate and/or transport food items that a single individual can not move. The challenge is to program a swarm of simple robots, with minimal communication and individual capability, to search the environment for a food target and return it to their base collectively. In robotics, foraging is important for several reasons. It is a metaphor for a broad class of problems integrating exploration, navigation and object identification, manipulation and transport. In multi-robot systems, foraging is a canonical problem for the study of robot-robot cooperation. Moreover, many actual or potential real-world applications for robotics are instances of foraging robots, for instance cleaning, harvesting, search and rescue, land-mine clearance or planetary exploration. This thesis focuses on foraging as a multi-robot task. The robots start at a home location, explore the world in search of their target, and return from the target incrementally to the home. Such task is difficult in swarm robotic systems because of the lack of global localization, communication, and odometry, which make it impossible for the robots to acquire or build maps, for example. Once a robot loses contact with the other robots, it is effectively lost and has no way to return home. This work focuses on robots with very simple hardware capabilities. The robots are not assumed to have global position information, global communication, and their only