Collective circular and translational motion of multiple unicycles using relative position and heading information in local coordinates

Motivated by practical scenarios such as cooperative surveillance and target circumnavigation of multi-agent systems in GPS-denied environments, we need to consider collective circular motion with formation-translation (CFT) motion of multiple nonholonomic vehicles using locally available relative position and heading information. We aim to make the whole group of agents move along their desired circles respectively and make the agents maintain a desired formation with translational motion. In this paper, we assume that the circular center information is only available to one agent (the leader agent), whereas every other agent can only obtain relative position and heading information of neighbors in their respective local coordinates, as well as their local absolute heading information. We model the information flow of the agents as a weighted digraph, which contains a directed spanning-tree with a root node that has no incoming edges. We propose a distributed controller for collective CFT motion of the vehicles with information in their respectively local coordinates, and prove asymptotic convergence of such group motion. © 2026 The Franklin Institute. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.