Closing Speed Computation using Stereo Camera and Applications in Unsignalized T-Intersection

2024-12-30

AI Summary

Autonomous vehicles navigating uncontrolled T-junctions face collision risks due to depth estimation errors from stereo cameras. Method: This paper proposes a safety-critical decision-making framework integrating uncertainty-aware modeling and geometric guarantees. First, it constructs a closed-loop velocity uncertainty model and introduces an adaptive depth sampling mechanism to rigorously bound its upper confidence limit. Second, it innovatively embeds the bounded closing-velocity constraint into the convex hull property of quadratic Bézier curves for trajectory planning, enabling analytically verifiable collision avoidance. Results: Evaluated on NGSIM-based realistic traffic simulations, the method significantly improves both trajectory safety and real-time performance, demonstrating substantial robustness enhancement for uncontrolled intersection navigation under measurement noise.

Technology Category

• Planning, Routing, and Scheduling: Planning under Uncertainty
• Reasoning under Uncertainty: Sequential Decision Making
• Intelligent Robots: State Estimation

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Large-scale security measurements
• Semantics and Knowledge: Provenance, trust, security and privacy, and ethical issues in managing semantic data

Abstract

This letter presents a conflict resolution strategy for an autonomous vehicle mounted with a stereo camera approaching an unsignalized T-intersection. A mathematical model for uncertainty in stereo camera depth measurements is considered and an analysis establishes the proposed adaptive depth sampling logic which guarantees an upper bound on the computed closing speed. Further, a collision avoidance logic is proposed that utilizes the closing speed bound and generates a safe trajectory plan based on the convex hull property of a quadratic B'ezier curve-based reference path. Realistic validation studies are presented with neighboring vehicle trajectories generated using Next Generation Simulation (NGSIM) dataset.

Problem

• Autonomous Vehicles
• T-intersections
• Collision Avoidance

Innovation

• T-intersection autonomous driving
• Dual-camera system
• Collision prevention system