Cooperative Relative Localization for Mobile Robot Teams: An Ego-Centric Approach

Abstract

Abstract This paper describes a cooperative relative localization method for mobile robot teams. That is, it describes a method whereby each robot may determine the pose of every other robot in the team, relative to itself. This method does not require GPS, landmarks, or any kind of environment model. Instead, robots make direct measurements of the relative pose of nearby robots, and broadcast this information to the team as a whole; each robot processes this information to generate ego-centric estimates of the pose of other robots, including those robots that they cannot observe directly. The method makes use of a Bayesian formalism and particle filter implementation, and is, as a result, very robust. The system described in this paper will both self-initialize (i.e., it does not require a priori pose estimates) and self-correct (it can recover from tracking failures). The method is well suited to applications involving unstructured, unknown or non-stationary environments.