Research on lane change prediction model based on GBDT

Research on lane change prediction model based on GBDT
复制标题

基于GBDT的换道预测模型研究

DOI:
10.1016/j.physa.2022.128290
复制
发表时间:
2022-11
期刊:
Physica A: Statistical Mechanics and its Applications
影响因子:
--
通讯作者:
Changxi Ma
Changxi Ma
中科院分区:
其他
文献类型:
--
作者:
Dong Li;Changxi Ma

文献摘要

参考文献

相似文献

车辆换乘是车辆行驶过程中极其重要的环节,对交通安全有很大影响。特别是在自动驾驶汽车中,有可能准确预测驾驶安全的发展和自动驾驶汽车的发展。本文利用NGSIM自然数驾驶数据集,对小型车行驶数据进行筛选,然后对变道车辆数据和非变道车辆数据进行标注,最后利用集成学习中的随机森林模型和梯度提升树模型对NGSIM数据进行集中,对US-101段的中小型车辆数据进行训练。最后,对两种模型的训练精度和测试精度进行了比较,发现梯度Boost树模型对车辆变异性的预测效果较好,可用于自动驾驶对后备箱的预测和判断。
Vehicle Change is an extremely important part of the vehicle driving process, which has great impact on traffic safety. Especially in the automatic driving vehicle, it is possible to accurately predict the development of driving safety and the development of automatic driving vehicles. This paper uses NGSIM natural number driving data set, screening the driving data of the small car, and then the lane-changing vehicle data and non-lane-changing vehicle data are tagged, and finally using the random forest model and gradient lift tree model in integrated learning to NGSIM data concentration The data of small and medium-sized vehicles in the US-101 section will be trained. Finally, the training accuracy and test accuracy of the two models are compared, and the gradient boost tree model exhibits a good effect on the prediction result of the vehicle variability, and can be used for automatic driving of the trunk prediction and judgment.
DOI: 10.1016/j.physa.2015.06.028
发表时间: 2015-11
影响因子: 3.3
作者:
Xiang Li;Jian-Qiao Sun
通讯作者: Xiang Li;Jian-Qiao Sun
DOI: 10.1177/0361198120922210
发表时间: 2020-06-12
影响因子: 1.7
作者:
Mahajan, Vishal;Katrakazas, Christos;Antoniou, Constantinos
通讯作者: Antoniou, Constantinos
DOI: 10.1007/978-1-4419-9326-7_5
发表时间: 2012-01-01
期刊: ENSEMBLE MACHINE LEARNING: METHODS AND APPLICATIONS
影响因子: --
作者:
Cutler, Adele;Cutler, D. Richard;Stevens, John R.
通讯作者: Stevens, John R.
DOI: 10.1109/ms.2016.31
发表时间: 2016-03
期刊: IEEE Software
影响因子: 3.3
作者:
Weishan Zhang;Liang Xu;Zhongwei Li;Q. Lu;Yan Liu
通讯作者: Weishan Zhang;Liang Xu;Zhongwei Li;Q. Lu;Yan Liu
DOI: 10.1007/978-0-387-77501-2_5
发表时间: 2020
期刊: Statistical Learning from a Regression Perspective
影响因子: --
作者:
Richard A. Berk
通讯作者: Richard A. Berk