A test method and device for urban intersection unmanned vehicle passing decision

By constructing a simulation environment for a four-way, two-lane urban intersection and an intelligent driver model, randomly generating information about oncoming vehicles, and combining this with a bicycle model for kinematic modeling, the problem of the simulation environment being disconnected from the real-world scenario in existing technologies is solved, achieving efficient closed-loop training and improved robustness of the decision-making model.

CN122242304APending Publication Date: 2026-06-19ADVANCED TECH RES INST OF BEIJING UNIV OF TECH +3
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ADVANCED TECH RES INST OF BEIJING UNIV OF TECH
Filing Date
2026-05-22
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing technologies struggle to construct a simulation testing environment that can realistically simulate the diverse interactive behaviors of drivers on actual roads while achieving efficient closed-loop training with relatively low computing power. This results in a disconnect between the simulation environment and real-world scenarios, leading to low training efficiency and insufficient generalization ability of the decision-making model.

Method used

A simulation environment for a four-way, two-lane urban intersection is constructed. The number, location, and driver type of oncoming vehicles are randomly generated. A kinematic modeling method combining an intelligent driver model and a bicycle model is used to generate environmental vehicle state information. Finally, an autonomous vehicle traffic decision is output through a deep reinforcement learning model.

Benefits of technology

It improves the adaptability and robustness of the autonomous vehicle traffic decision-making model in real-world scenarios, ensuring traffic safety and efficiency while significantly reducing computational complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122242304A_ABST
    Figure CN122242304A_ABST
Patent Text Reader

Abstract

This invention discloses a testing method and device for autonomous vehicle traffic decision-making at urban intersections, belonging to the field of autonomous driving testing technology. The invention includes: constructing a four-way, two-lane urban intersection simulation environment and setting a fixed reference trajectory for left-turning vehicles; randomly generating the number, initial position, initial speed, and driving intention of oncoming vehicles, and randomly assigning aggressive, cautious, or calm driver types; calling the corresponding decision logic according to the driver type, calculating the acceleration of oncoming vehicles by combining a bicycle model and an intelligent driver model, updating their position and speed, and generating environmental state information; inputting the state information into the decision model to be tested, outputting the vehicle's acceleration, and updating the vehicle's state; collecting traffic result data, calculating a comprehensive score, and evaluating or optimizing the decision model. This invention can accurately fit actual driver behavior, achieving traffic decision training and testing in unprotected left-turn scenarios with low computing power, improving the safety and traffic efficiency of the decision model.
Need to check novelty before this filing date? Find Prior Art