A Feasibility-Enforcing Primal-Decomposition SQP Algorithm for Optimal Vehicle Coordination

TL;DR

This paper presents a real-time, distributed optimization algorithm for autonomous vehicle coordination at intersections, validated through simulations and real-world experiments with cars on a test track.

Contribution

It introduces a primal-decomposition SQP algorithm tailored for vehicle coordination, extending prior work with new algorithmic solutions and real-world testing.

Findings

Algorithm is feasible for real-time implementation.

Successful coordination demonstrated in simulations.

Real car experiments confirm practical applicability.

Abstract

In this paper we consider the problem of coordinating autonomous vehicles approaching an intersection. We cast the problem in the distributed optimisation framework and propose an algorithm to solve it in real time. We extend previous work on the topic by testing two alternative algorithmic solutions in simulations. Moreover, we test our algorithm in experiments using real cars on a test track. The experimental results demonstrate the applicability and real-time feasibility of the algorithm.