TY - GEN
T1 - A Lane-Changing Decision Model of Structured Roads Based on Optimized XGBoost Algorithm
AU - Zhao, Zhen
AU - Ren, Xuemei
AU - Wang, Haoyuan
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022
Y1 - 2022
N2 - In the field of autonomous driving, lane-changing decision is an important content of driving decision mechanism. To solve the problem that autonomous vehicles in the driving process, due to environmental factors, are difficult to accurately make lane-changing decisions, this paper proposes a XGBoost prediction model using Bayesian optimization, and based on NGSIM project data, analysis of the vehicles lane-changing decisions under influence of the environment. The experimental results showed that the established SMAC-XGBoost model has better performance compared with other models. In the experiment of lane-changing decisions recognition and prediction, the recognition accuracy of SMAC-XGBoost model can reach more than 95%, which has a good prediction effect.
AB - In the field of autonomous driving, lane-changing decision is an important content of driving decision mechanism. To solve the problem that autonomous vehicles in the driving process, due to environmental factors, are difficult to accurately make lane-changing decisions, this paper proposes a XGBoost prediction model using Bayesian optimization, and based on NGSIM project data, analysis of the vehicles lane-changing decisions under influence of the environment. The experimental results showed that the established SMAC-XGBoost model has better performance compared with other models. In the experiment of lane-changing decisions recognition and prediction, the recognition accuracy of SMAC-XGBoost model can reach more than 95%, which has a good prediction effect.
KW - Bayesian optimization
KW - Data processing
KW - Lane-changing decision
KW - XGBoost algorithm
UR - http://www.scopus.com/inward/record.url?scp=85118160644&partnerID=8YFLogxK
U2 - 10.1007/978-981-16-6324-6_56
DO - 10.1007/978-981-16-6324-6_56
M3 - Conference contribution
AN - SCOPUS:85118160644
SN - 9789811663239
T3 - Lecture Notes in Electrical Engineering
SP - 548
EP - 557
BT - Proceedings of 2021 Chinese Intelligent Systems Conference - Volume II
A2 - Jia, Yingmin
A2 - Zhang, Weicun
A2 - Fu, Yongling
A2 - Yu, Zhiyuan
A2 - Zheng, Song
PB - Springer Science and Business Media Deutschland GmbH
T2 - 17th Chinese Intelligent Systems Conference, CISC 2021
Y2 - 16 October 2021 through 17 October 2021
ER -