Bearing-based distributed pose estimation for multi-agent networks

Abstract

In this paper, we address the distributed pose estimation problem for multi-agent systems, where the agents have unknown static positions and time-varying orientations. The interaction graph is assumed to be directed and acyclic with two leaders that have access to their position and orientation. We propose a nonlinear distributed pose estimation scheme relying on individual angular velocity measurements and local relative (time-varying) bearing measurements. The proposed estimation scheme consists of two cascaded distributed observers, an almost globally asymptotically stable (AGAS) attitude observer and an input-to-state stable (ISS) position observer, leading to an overall AGAS distributed localization scheme. Numerical simulation results are presented to illustrate the performance of our proposed distributed pose estimation scheme.