Fusing Odometry, UWB Ranging, and Spatial Detections for Relative Multi-Robot Localization

TL;DR

This paper introduces a flexible Monte Carlo-based method for multi-robot relative localization that fuses odometry, UWB ranging, and spatial detections, demonstrating improved accuracy and real-time deployment in experiments.

Contribution

A novel particle filter approach that integrates UWB, odometry, and spatial detections for enhanced multi-robot localization accuracy.

Findings

Single UWB range suffices with odometry for accurate relative states.

Our method outperforms traditional multilateration in accuracy.

Integration of spatial detections further improves localization precision.

Abstract

This letter presents a cooperative relative multi-robot localization design and experimental study. We propose a flexible Monte Carlo approach leveraging a particle filter to estimate relative states. The estimation can be based on inter-robot Ultra-Wideband (UWB) ranging and onboard odometry alone or dynamically integrated with cooperative spatial object detections from stereo cameras mounted on each robot. The main contributions of this work are as follows. First, we show that a single UWB range is enough to estimate the accurate relative states of two robots when fusing odometry measurements. Second, our experiments also demonstrate that our approach surpasses traditional methods, namely, multilateration, in terms of accuracy. Third, to further increase accuracy, we allow for the integration of cooperative spatial detections. Finally, we show how ROS 2 and Zenoh can be integrated to build a scalable wireless communication solution for multi-robot systems. The experimental validation includes real-time deployment and autonomous navigation based on the relative positioning method. It is worth mentioning that we also address the challenges for UWB-ranging error mitigation for mobile transceivers. The code is available at https://github.com/TIERS/uwb-cooperative-mrs-localization.