MORE | Spring 2021

Control and Coordination of Multi-Robot Swarm Systems

Robot swarms show great promise in performing complex tasks in areas ranging from search and rescue to interplanetary exploration and yet controlling the behaviors of these swarms effectively is an open research problem. This research investigates techniques to control a drone swarm where the drones learn to act using demonstrations obtained from artificially generated swarm motion data and then utilize Reinforcement Learning(RL) to improve performance in the face of real-world challenges such as sensor noise. The use of Graph Neural Networks (GNNs) in the controller for the swarm system provides the drones with contextual information about their environment and serves as an inductive bias in the learning process.