Joint Relative Localization and Formation Control for Multiagent Systems With Nonpersistent Excitation

This article investigates the joint relative localization and formation control problem of leader–follower multiagent systems in GPS-denied environments. Existing results normally require persistently excited inter-agent relative motion to achieve accurate relative localization, which causes inevitable formation errors. To address this issue, an adaptive distributed relative localization scheme without persistent excitation (PE) motions is integrated into the leader–follower formation framework, using inter-agent distance and onboard velocity measurements. Specifically, by employing the parameter-estimation-based observer (PEBO) technique and introducing the dynamic extension operation, the time-varying relative position estimation is reformulated as an online constant parameter identification problem. Furthermore, we develop a relative localization-based formation control strategy for stationary target formations and further extend it to scenarios involving dynamic target formations. It is proven that the proposed method guarantees the asymptotic convergence to the desired formations. Finally, both numerical simulations and physical experiments are presented to illustrate the effectiveness of our theoretical results.