Distributed SE(d) Formation Control of Multi‐Agent Systems Using Relative Measurements

ABSTRACT This study develops a solution to the distributed formation control problem of multi‐agent systems using only relative measurements between agents. In this problem, poses, that is, orientations and positions, of the agents are expected to achieve desired ones. We assume that each agent can only measure its neighbors' relative poses, and that communication between agents is unavailable. To address this problem, a fully distributed and relative controller is proposed to ensure achieving the control objective. We employ the gradient‐flow method with a clique‐based objective function to achieve the best performance of the distributed, gradient‐based controller. Subsequently, a network condition for achieving this objective is the connectedness. The proposed method ensures the stationary center of the agents and the minimum undesired equilibrium set, yielding short travel distances and a large attraction region, respectively. Furthermore, the proposed method outperforms the existing methods as evidenced by simulations in a two‐dimensional space. In addition, the effectiveness of the proposed method is verified through a three‐dimensional simulation. The proposed method shows potential for applications in drone swarm coordination, autonomous underwater vehicle formation, and mobile robot collaborative transportation when only local sensing is available while communication is limited.