Engineering Collective Intelligence: Methods in Multi-Agent Coordination

Catalog description

Decentralized, self-organizing systems are the future of autonomous technology. Large numbers of self driving cars, ground robots, delivery drones, and mobile sensors are being rapidly deployed into the open world. It is imperative that these systems are capable of proactive collaboration to operate safely and cohesively in dynamic environments. This course will provide an overview of state-of-the-art methods for the design of decentralized coordination in groups of communicating agents, with a focus on mathematical fundamentals and code implementation. A particular emphasis will be placed on nonlinear methods grounded in biological and cognitive principles that give rise to collectively intelligent behaviors in nature. Possible topics covered include collective decision-making, synchronization, and task allocation, with applications to collaborative navigation and collision avoidance.