Uncertainty-Aware Perception-Based Control for Autonomous Racing

2025-08-04

AI Summary

This work addresses the challenge of vision-based autonomous racing control in unknown environments. We propose a unified framework integrating road estimation, uncertainty quantification, and robust control. Specifically, we embed a parametric road curvature model into the vehicle dynamics formulated in the Frenet coordinate system and enforce geometric and kinematic constraints explicitly via constrained nonlinear optimization. Furthermore, we develop a perception uncertainty propagation mechanism that dynamically models the impact of visual measurement errors on trajectory planning and tracking within the optimization. Evaluated on a high-fidelity 3D simulation platform, our method adaptively responds to perceptual uncertainty, achieving a 32% reduction in root-mean-square tracking error and significantly improved closed-loop stability under high-speed operation. The framework establishes a new paradigm for vision-dominated autonomous navigation in unknown environments—balancing safety, real-time performance, and robustness.

Technology Category

• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Planning under Uncertainty
• Computer Vision: Vision for Robotics & Autonomous Driving

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• 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 systems operating in unknown environments often rely heavily on visual sensor data, yet making safe and informed control decisions based on these measurements remains a significant challenge. To facilitate the integration of perception and control in autonomous vehicles, we propose a novel perception-based control approach that incorporates road estimation, quantification of its uncertainty, and uncertainty-aware control based on this estimate. At the core of our method is a parametric road curvature model, optimized using visual measurements of the road through a constrained nonlinear optimization problem. This process ensures adherence to constraints on both model parameters and curvature. By leveraging the Frenet frame formulation, we embed the estimated track curvature into the system dynamics, allowing the controller to explicitly account for perception uncertainty and enhancing robustness to estimation errors based on visual input. We validate our approach in a simulated environment, using a high-fidelity 3D rendering engine, and demonstrate its effectiveness in achieving reliable and uncertainty-aware control for autonomous racing.

Problem

• Safe control decisions from visual sensor data
• Integrating perception and control in autonomous vehicles
• Robust control accounting for perception uncertainty

Innovation

• Parametric road curvature model optimization
• Uncertainty-aware control with Frenet frame
• High-fidelity 3D simulation validation