Global Attitude Synchronization for Heterogeneous Multi-agent Systems on SO(3)
Mouaad Boughellaba, Soulaimane Berkane, Abdelhamid Tayebi
TL;DR
This work addresses global attitude synchronization for heterogeneous rigid bodies evolving on $SO(3)$ under a tree-structured interaction graph. It first develops a continuous distributed scheme with almost global asymptotic stability and then introduces two global hybrid schemes: one with angular velocity measurements and one velocity-free that relies on auxiliary dynamics. A generic potential on $SO(3)^M\times\mathbb{R}^M$ is used to design the hybrid controllers, ensuring a unique global attractor and eliminating undesired equilibria; the velocity-free version compensates the lack of velocity information via an auxiliary system. Simulations on a seven-satellite network illustrate faster convergence and robustness of the hybrid controllers, highlighting practical benefits for sensor-bearing and scalable multi-agent platforms. The results advance global synchronization on $SO(3)$ for heterogeneous agents and lay groundwork for extending to time-varying/topology-delayed networks.
Abstract
In this paper, we address the problem of attitude synchronization for a group of rigid body systems evolving on SO(3). The interaction among these systems is modeled through an undirected, connected, and acyclic graph topology. First, we present an almost global continuous distributed attitude synchronization scheme with rigorously proven stability guarantees. Thereafter, we propose two global distributed hybrid attitude synchronization schemes on SO(3). The first scheme is a hybrid control law that leverages angular velocities and relative orientations to achieve global alignment to a common orientation. The second scheme eliminates the dependence on angular velocities by introducing dynamic auxiliary variables, while ensuring global asymptotic attitude synchronization. This velocity-free control scheme relies exclusively on attitude information. The proposed schemes are applicable to heterogeneous multi-agent systems, where agents may have distinct inertia matrices. Simulation results are provided to illustrate the effectiveness of the proposed distributed attitude synchronization schemes.
