|
|
|
| Single Intersection Path Planning Based on Local Game |
| JIANG Nan1a, ZHAO Qinghai1a,2, XU Chong1a, DU Siyu1b
|
1. a. School of Mechanical and Electrical Engineering; b. School of Automation, Qingdao University, Qingdao 266071,China;
2. Electric vehicle intelligent power integration technology national local joint engineering research center, Qingdao 266071,China |
|
|
|
|
Abstract Aiming at the problems of high collision rate and low traffic efficiency caused by conflict risk of intelligent vehicles in single intersection environment, a path planning algorithm based on local game algorithm is proposed. By establishing the intersection conflict risk model, the conflict risk is analyzed, and the conflict resolution method between vehicles and pedestrians is proposed. The local game theory is introduced to establish the profit function to evaluate the feasible decision, and the solution of the pure strategy Nash equilibrium under the constraint condition is obtained. The vehicle speed is planned, and the path planning of the intelligent vehicle is finally realized. The simulation results show that the proposed path planning algorithm improves the interaction success rate by 9.32% compared with the evolutionary game and cooperative game algorithm under different traffic flows, and the traffic efficiency increases by 33% on average, which can effectively alleviate the traffic pressure.
|
|
Received: 20 January 2023
Published: 03 January 2025
|
|
|
|
|
|
|
|
|
