Human-Usable and Emergency Vehicle-Aware Control Policies for Autonomous Intersection Management (2006)
Traffic congestion and automobile accidents are two of the leading causes of decreased standard of living and lost productivity in urban settings. Recent advances in artificial intelligence and, specifically, intelligent vehicle technology suggest that vehicles driven entirely by autonomous agents will be possible in the near future. In previous work, we presented a novel reservation-based approach for governing interactions of multiple autonomous vehicles, specifically at intersections. This approach alleviated many traditional problems associated with intersections, in terms of both safety and efficiency. However, such a system relies on all vehicles being equipped with the requisite technology --- a restriction that would make implementing such a system in the real world extremely difficult. In this paper, we augment the system such that it is able to accomodate traditional human-operated vehicles using existing infrastructure. Furthermore, we show that as the number of autonomous vehicles on the road increases, traffic delays decrease monotonically toward the levels exhibited in the system involving only autonomous vehicles. Additionally, we demonstrate how the system can be extended to allow high-priority vehicles such as ambulances, police cars, or fire trucks through more quickly without placing undue burden on other vehicles. Both augmentations are fully implemented and tested in our custom simulator, and we present detailed experimental results attesting to their effectiveness.
In AAMAS 2006 Workshop on Agents in Traffic and Transportation, May 2006.

Kurt Dresner Ph.D. Alumni kurt [at] dresner name
Peter Stone Faculty pstone [at] cs utexas edu