The prediction of delay time at intersection and route planning for autonomous vehicles

• Main Authors: GOU, Genwang, ZHAO, Yongxin, LIANG, Jiawei, SHI, Ling
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2020
• Subjects: intersection management; autonomous vehicle; SMOTE algorithm; route planning; Theory and Algorithms; Transportation
• Online Access: https://ink.library.smu.edu.sg/sis_research/7835; https://ink.library.smu.edu.sg/context/sis_research/article/8838/viewcontent/paper018.pdf
• Institution: Singapore Management University

Intelligent Intersections (roundabout and crossroads) management is considered as one of the challenges to significantly improve urban traffic efficiency. Recent researches in artificial intelligence suggest that autonomous vehicles have the possibility of forming intelligent intersection management, and likely to occupy the leading role in future urban traffic. If route planning method can be used for route decision of autonomous vehicle, the urban traffic efficiency can be further improved. In this paper, we propose an Intelligent Intersection Control Protocol (IICP) for controlling autonomous vehicles cross intersection, and recommend route for autonomous vehicles to reduce travel time and improve urban traffic efficiency. Firstly, we run IICP to obtain the original data, use SMOTE algorithm to synthesize balance data, and use RF, GBDT algorithms to predict delay time. Secondly, we use the iEigenAnt algorithm to find multiple short routes in traffic network. Finally, we recommend route for autonomous vehicles based on the minimum of driving time on the route and all delay time at each intersection to improve urban traffic efficiency.