Decentralized Formation Control With Prescribed Distance Constraints and Shape Uniqueness
Yuqi Pan, Binglin He, Junyi Geng, Yang Wang · 2023
For Multi-Agent Systems (MAS) formation problem, this paper presents a novel decentralized control protocol that can guarantee the uniqueness of formation shape and restrain the distance between neighboring agents within a prescribed range. The proposed control law is decentralized, in the sense that each agent merely employs local relative information regarding its neighbors to obtain the control input. Using a delicately designed gain matrix, we avoid the problem of non-uniqueness of shape that is existed in the majority of the decentralized formation methods, especially the distance-based methods. Meanwhile, the distance constraints, like connectivity maintenance and collision avoidance, required in many practical scenarios are also addressed. Furthermore, the convergence of formation is rigorously proved under a less restrictive assumption on the communication graph. Finally, a comparative simulation verifies the effectiveness and superiority of proposed approach.