Robust Multi-Robot Global Localization with Unknown Initial Pose based on Neighbor Constraints

TL;DR

This paper introduces a neighbor-constraint-based data association algorithm for multi-robot global localization, significantly enhancing robustness in large-scale environments with unknown initial poses by leveraging semantic graph invariance.

Contribution

The paper proposes a novel neighbor-constraint data association method that improves robustness over existing semantic graph approaches in multi-robot localization.

Findings

Demonstrates improved robustness on three datasets

Outperforms previous methods in large-scale environments

Effective in scenarios with low map overlap

Abstract

Multi-robot global localization (MR-GL) with unknown initial positions in a large scale environment is a challenging task. The key point is the data association between different robots' viewpoints. It also makes traditional Appearance-based localization methods unusable. Recently, researchers have utilized the object's semantic invariance to generate a semantic graph to address this issue. However, previous works lack robustness and are sensitive to overlap rate of maps, resulting in unpredictable performance in real-world environments. In this paper, we propose a data association algorithm based on neighbor constraints to improve the robustness of the system. We demonstrate the effectiveness of our method on three different datasets, indicating a significant improvement in robustness compared to previous works.