DTR: Delaunay Triangulation-based Racing for Scaled Autonomous Racing

2025-05-30

AI Summary

Reactive controller FTG suffers from the “FTG trap”—loss of control when track boundaries are missing—due to its reliance on boundary detection. Method: This paper proposes DTR, a Delaunay Triangulation-based Reactive controller that directly processes raw LiDAR point clouds. It segments boundaries and constructs local geometric structure via Delaunay triangulation to robustly fit a centerline, enabling high-speed autonomous racing without localization, mapping, or global planning. Contribution/Results: DTR is the first application of Delaunay triangulation to real-time racecar control; it systematically avoids the FTG trap while ensuring high safety and ultra-low latency. Real-world evaluation shows a 70% lap-time improvement over FTG, approaching map-dependent methods; end-to-end latency is only 8.95 ms with 38.85% CPU utilization. The system has been successfully deployed and validated on a physical racecar platform.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Reactive controllers for autonomous racing avoid the computational overhead of full ee-Think-Act autonomy stacks by directly mapping sensor input to control actions, eliminating the need for localization and planning. A widely used reactive strategy is FTG, which identifies gaps in LiDAR range measurements and steers toward a chosen one. While effective on fully bounded circuits, FTG fails in scenarios with incomplete boundaries and is prone to driving into dead-ends, known as FTG-traps. This work presents DTR, a reactive controller that combines Delaunay triangulation, from raw LiDAR readings, with track boundary segmentation to extract a centerline while systematically avoiding FTG-traps. Compared to FTG, the proposed method achieves lap times that are 70% faster and approaches the performance of map-dependent methods. With a latency of 8.95 ms and CPU usage of only 38.85% on the robot's OBC, DTR is real-time capable and has been successfully deployed and evaluated in field experiments.

Problem

• Reactive controllers for autonomous racing avoid computational overhead
• FTG strategy fails in scenarios with incomplete boundaries
• DTR combines Delaunay triangulation to avoid FTG-traps

Innovation

• Uses Delaunay triangulation for LiDAR processing
• Segments track boundaries to avoid FTG-traps
• Achieves real-time performance with low CPU usage