Formation control of multi-agent systems with switching topologies: An ILC-based distributed algorithm design

This paper deals with iterative formation control problems for multi-agent systems with switching topologies. Our approach combines consensus protocol algorithms conducted with an iterative learning control update. This leads to networks dynamically changing with evolution along two directions: a finite time axis and an infinite iteration axis. A distributed algorithm is constructed based on nearest neighbor information, and its convergence is established via a Roesser system-based two-dimensional approach. It is shown that a desired relative formation between agents can be achieved along a finite time trajectory if and only if the union of the interaction graphs between the agents over iteration intervals of finite length has a spanning tree for every time step. Simulations are included to illustrate the effectiveness of the derived results.