An Integrated Localization and Control Framework for Multi-Agent Formation

High-accuracy formation is essential for multi-agent systems to accomplish certain tasks, and the accuracy of the formation is determined jointly by the network localization and formation control procedures. Existing studies commonly treat the two procedures separately in the system design, leading to suboptimal formation performance. This paper establishes a general framework for high-accuracy multi-agent formation via integrated localization and control. In particular, we propose a new metric called the formation error to characterize the minimum squared distance between two formations for arbitrary translation and rotation, and develop an integrated localization and control scheme to minimize the mean formation error (MFE). Theoretical bounds for the MFE are derived in a closed form, which guides the integrated design of the sensing strategy and control policy. In the case study, we develop efficient integrated algorithms for multi-agent formation under spectrum resource constraints. Numerical results validate the performance gain of the proposed algorithms over existing ones as well as demonstrate the effects of the network parameters on formation performance.