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

2026-04-15

AI Summary

This work addresses the degraded robustness and trajectory consistency of 3D multi-object tracking (MOT) in adverse environmental conditions and long-range scenarios. To this end, we propose RadarMOT, a novel framework that explicitly incorporates radar point clouds into the tracking pipeline rather than treating them merely as embedded features within a network. By integrating radar with LiDAR and camera modalities, explicitly modeling radar point clouds, and refining state estimation, RadarMOT effectively compensates for missed detections at long ranges and enhances tracking stability. Evaluated on the MAN-TruckScenes dataset, our method achieves a 12.7% improvement in AMOTA under long-range conditions and a 10.3% gain in adverse weather, significantly outperforming existing approaches.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Tracking, profiling, and countermeasures against them
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The challenge of 3D multi-object tracking (3D MOT) 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 multi-modal fusion methods usually treat radar as another learned feature inside the network. When the overall model degrades in difficult environmental conditions, the robustness advantages that radar could provide are also reduced. We propose RadarMOT, a radar-informed 3D MOT framework that explicitly uses radar point cloud data as additional observation to refine state estimation and recover detector misses at long ranges. Evaluations on the MAN-TruckScenes dataset show that RadarMOT consistently improves the Average Multi-Object Tracking Accuracy (AMOTA) with absolute 12.7% at long range and 10.3% in adverse weather. The code will be available at https://github.com/bingxue-xu/radarmot

Problem

• 3D multi-object tracking
• adverse conditions
• long-range tracking
• robustness
• sensor fusion

Innovation

• Radar-Informed Tracking
• 3D Multi-Object Tracking
• Sensor Fusion
• Radar Point Cloud
• Adverse Conditions