Cooperative Control of Multi-Channel Linear Systems with Self-Organizing Private Agents

Abstract: Cooperative behavior design for multi-agent systems with collective tasks is a critical issue in promoting swarm intelligence. This paper investigates cooperative control for a multi-channel system, where each channel is managed by an agent expected to self-organize a controller to stabilize the system collaboratively by communicating with neighbors in a network. Integrating a state decomposition technique and a fusion approach, a fully distributed privacy-preserving mechanism is proposed to shield agents' private information from neighbors' eavesdropping. Moreover, the cost of introducing the privacy-preserving mechanism and the benefit of adding more channels to the system are quantitatively analyzed. Finally, comparative simulation examples are provided to demonstrate the effectiveness of the theoretical results.