An Efficient 3D Multi-Object Tracking Algorithm for Low-Cost UGV Using Multi-Level Data Association

Highlights: What are the main findings? An efficient multi-object tracking algorithm is proposed, which integrates 2D object detection with unsupervised learning to achieve 3D object detection. A multi-object tracking algorithm using multi-level data association strategy is proposed, which effectively mitigates the impact of detection errors on tracking and prevents trajectory fragmentation caused by occlusion and other factors. What is the implication of the main finding? The proposed detection and tracking algorithm is independent of high-performance GPU devices, making it advantageous for practical deployment. On low-cost unmanned system platforms, our algorithm can maintain strong stability. The proposed algorithm employs multi-sensor fusion to effectively address the problem of trajectory fragmentation in the absence of 3D information, maintaining high accuracy in complex scenarios. 3D object detection and tracking technology are increasingly being adopted in unmanned ground vehicles, as robust perception systems significantly improve the obstacle avoidance performance of a UGV. However, most existing algorithms depend heavily on computationally intensive point cloud neural networks, rendering them unsuitable for resource-constrained platforms. In this work, we propose an efficient 3D object detection and tracking method specially designed for deployment on low-cost vehicle platforms. For the detection phase, our method integrates an image-based 2D detector with data fusion techniques to coarsely extract object point clouds, followed by an unsupervised learning approach to isolate objects from noisy point cloud data. For the tracking process, we propose a multi-target tracking algorithm based on multi-level data association. This method introduces an additional data association step to handle targets that fail in 3D detection, thereby effectively reducing the impact of detection errors on tracking performance. Moreover, our method enhances association precision between detection outputs and existing trajectories through the integration of 2D and 3D information, thereby further mitigating the adverse effects of detection inaccuracies. By adopting unsupervised learning as an alternative to complex neural networks, our approach demonstrates strong compatibility with both low-resolution LiDAR and GPU-free computing platforms. Experiments on the KITTI benchmark demonstrate that our tracking framework achieves significant computational efficiency gains while maintaining detection accuracy. Furthermore, experimental evaluations on the real-world UGV platform demonstrated the deployment feasibility of our approach.