Radar-Informed 3D Multi-Object Tracking under Adverse Conditions

TL;DR

RadarMOT is a radar-informed 3D multi-object tracking framework that explicitly uses radar data to improve robustness and accuracy in adverse conditions and at long ranges.

Contribution

It introduces a novel radar-informed approach that explicitly incorporates radar point clouds for enhanced multi-object tracking performance.

Findings

Improves AMOTA by 12.7% at long range

Enhances robustness in adverse weather conditions

Explicit radar integration outperforms implicit methods

Abstract

The challenge of 3D multi-object tracking is achieving robustness in real-world applications, for example under adverse conditions and maintaining consistency as distance increases. To overcome these challenges, sensor fusion approaches that combine LiDAR, cameras, and radar have emerged. However, existing multimodal methods usually treat radar as another learned feature inside the network. When the overall model degrades in difficult environments, the robustness advantages that radar could provide are also reduced. In this paper we propose RadarMOT, a radar-informed 3D multi-object tracking framework that explicitly uses radar point clouds as additional observations to refine state estimation and recover objects missed by the detector at long ranges. Evaluations on the MAN-TruckScenes dataset show that RadarMOT consistently improves the Average Multi-Object Tracking Accuracy (AMOTA) by 12.7\% at long range and up to 10.3\% in adverse weather. The code will be available at https://github.com/bingxue-xu/radarmot