Department of Mechanical Engineering
Multi-Agent Coordination by Decentralized Estimation and Control
I will describe a framework for the design of collective behaviors for groups of identical mobile agents. The approach is based on decentralized simultaneous estimation and control, where each agent communicates with neighbors and estimates the global performance properties of the swarm needed to make a local control decision. Challenges of the approach include designing a control law with desired convergence properties, assuming each agent has perfect global knowledge; designing an estimator that allows each agent to make estimates of the global properties needed to implement the controller; and possibly modifying the controller to recover desired convergence properties when using estimates of global performance. I will describe this framework applied to two different problems, formation control and cooperative target localization.
The stability of networked mobile agents using decentralized estimate-and-control is not ensured by any simple separation principle. I will discuss how to derive small-gain conditions guaranteeing stability. These conditions are typically expressed as bounds on the aggressiveness of control gains as a function of the communication network. We believe such conditions are fundamental to a theory of decentralized cooperative control.
This is joint work with my student Peng Yang and my colleague Randy Freeman.