Computational model for continuous flow of autonomous vehicles at road intersections

Area of Science:

• Intelligent Transportation Systems
• Autonomous Vehicle Technology
• Traffic Management

Background:

• Advancements in autonomous vehicle (AV) technology are increasing their reliability and adoption.
• Widespread AV use necessitates more efficient traffic management systems, particularly at intersections.
• Current traffic light systems are ill-suited for a future dominated by autonomous vehicles.

Purpose of the Study:

• To develop and evaluate a computational model for managing autonomous vehicle crossings at intersections.
• To enable continuous traffic flow for autonomous vehicles, minimizing the need for stops.
• To assess the impact of controller range and vehicle characteristics on intersection management efficiency.

Main Methods:

• Development of a computational model for autonomous vehicle intersection management.
• Implementation of a simulation algorithm to control AV behavior at intersections.
• Conducting 600,000 simulations with varying controller action distances and vehicle group sizes.

Main Results:

• A zero-collision rate was observed when the intersection controller's action distance was 2300 meters or greater.
• The efficiency of the management method correlated positively with the controller's range.
• Autonomous vehicle speeds at intersections closely matched their initial average speeds, indicating smooth flow.

Conclusions:

• The proposed computational model effectively manages autonomous vehicle traffic at intersections.
• A sufficient controller range (≥2300m) is critical for ensuring collision-free autonomous vehicle movement.
• The system facilitates continuous traffic flow, maintaining near-initial speeds for autonomous vehicles.