Autonomous Intersection Crossing With Vehicle Location Uncertainty

Abstract

To date, a large body of the literature has looked into the problem of autonomous intersection crossings facilitated by Connected Autonomous Vehicles (CAVs). Nevertheless, existing approaches assume that CAVs know their exact location and system state. This work presents a novel framework that allows for an optimized intersection management, which considers vehicle location uncertainties for linear-Gaussian systems. Building upon the proposed framework, a family of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$0-1$ </tex-math></inline-formula> integer linear programming optimizations are presented that can set, sequentially or simultaneously, the acceleration profiles of all vehicles in the intersection. Extensive simulation results are presented, proving that the proposed framework represents a real-time near-optimal approach that maximizes intersection throughput with probabilistic collision avoidance guarantees.