A Multiagent Approach to Autonomous Intersection Management

Dresner, K.; Stone, P. · 2008 · OpenAlex-citations

DOI: 10.1613/jair.2502

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the inefficiency and safety limitations of current intersection control mechanisms, such as traffic lights and stop signs, in the context of a future dominated by autonomous vehicles. The authors argue that existing systems are designed to compensate for human driver limitations, such as slow reaction times and inconsistent behavior, and thus fail to leverage the precision and speed of autonomous agents. As autonomous vehicles become prevalent, the bottleneck for roadway efficiency shifts from driver capability to the coordination mechanism itself. The research proposes a multiagent system where vehicles and intersections act as autonomous agents, utilizing a reservation-based approach to coordinate movement through intersections more efficiently than traditional methods. The authors developed a custom time-based simulator to evaluate their proposed framework. The simulation models a 250m x 250m area with a central intersection, incorporating realistic vehicle kinematics, sensor ranges, and physical constraints like maximum steering angles and acceleration limits. The core of the system is a communication protocol allowing driver agents to request reservations for specific space-time blocks within the intersection. An intersection manager processes these requests using a "First Come, First Served" (FCFS) policy. This policy divides the intersection into a grid of reservation tiles and runs an internal simulation of each vehicle's trajectory to ensure no spatial or temporal conflicts exist. If a conflict is detected, the request is rejected; otherwise, it is confirmed. The system is designed to be robust against message loss, ensuring safety even if communication fails, and includes extensions to handle mixed traffic with human drivers and prioritize emergency vehicles. Experimental results from the simulation demonstrate that the reservation-based mechanism significantly outperforms traditional traffic lights and stop signs in terms of efficiency. The FCFS policy allows vehicles to traverse intersections with minimal delay by enabling simultaneous, conflict-free movement from multiple directions, effectively subsuming the functionality of existing control methods. The system successfully avoids deadlocks and starvation, ensuring all vehicles eventually cross the intersection. Furthermore, the mechanism maintains safety guarantees, preventing collisions as long as vehicles adhere to the protocol. The extensions for human-driven vehicles and emergency priority were also tested, showing that the system can integrate with non-autonomous traffic and provide emergency access without imposing significant delays on civilian vehicles. The significance of this work lies in its demonstration that autonomous intersection management can dramatically improve roadway safety and efficiency by eliminating the inefficiencies inherent in human-centric traffic controls. By treating intersections as multiagent coordination problems, the proposed framework offers a scalable, distributed solution that does not rely on centralized control, thereby avoiding single points of failure. The research provides a foundational protocol and policy for future Intelligent Transportation Systems, suggesting that as autonomous vehicle adoption increases, replacing legacy intersection controls with reservation-based systems will be essential for maximizing the benefits of autonomous driving technology.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• ehmi external hmi