LiDAR-Based Vehicle Detection and Tracking for Autonomous Racing

2025-01-24

AI Summary

This work addresses the challenges of real-time 3D detection, high-precision vehicle pose estimation, and robust multi-object tracking in dynamic, high-speed autonomous racing scenarios—specifically at speeds exceeding 275 km/h. To this end, we propose three key innovations: (1) a lightweight geometric point cloud segmentation method to accelerate preprocessing; (2) a vehicle pose estimation algorithm that jointly optimizes 3D bounding box regression with iterative closest point (ICP)-based refinement for enhanced orientation and position accuracy; and (3) a variable-step joint probabilistic data association filter (JPDAF) framework for multi-target tracking, balancing real-time performance and association robustness. The end-to-end system latency is under 30 ms, with detection accuracy exceeding 92%. Evaluated in the Indy Autonomous Challenge, our system enabled the PoliMOVE team to achieve fully autonomous overtaking maneuvers and secure multiple championship victories.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Autonomous racing provides a controlled environment for testing the software and hardware of autonomous vehicles operating at their performance limits. Competitive interactions between multiple autonomous racecars however introduce challenging and potentially dangerous scenarios. Accurate and consistent vehicle detection and tracking is crucial for overtaking maneuvers, and low-latency sensor processing is essential to respond quickly to hazardous situations. This paper presents the LiDAR-based perception algorithms deployed on Team PoliMOVE's autonomous racecar, which won multiple competitions in the Indy Autonomous Challenge series. Our Vehicle Detection and Tracking pipeline is composed of a novel fast Point Cloud Segmentation technique and a specific Vehicle Pose Estimation methodology, together with a variable-step Multi-Target Tracking algorithm. Experimental results demonstrate the algorithm's performance, robustness, computational efficiency, and suitability for autonomous racing applications, enabling fully autonomous overtaking maneuvers at velocities exceeding 275 km/h.

Problem

• Autonomous Racing
• Vehicle Detection and Tracking
• Safety and Performance under Extreme Conditions

Innovation

• Lidar-based Detection
• High-speed Vehicle Tracking
• Autonomous Racing Application