重庆理工大学学报2024，Vol.38Issue(17)：1-13,13.DOI:10.3969/j.issn.1674-8425(z).2024.09.001
智能网联汽车轨迹预测研究现状与趋势
Research status and trend of trajectory prediction for intelligent connected vehicles
摘要
Abstract
As a key technology on intelligent connected vehicles,trajectory prediction plays an important role in ensuring vehicle safety and preventing traffic accidents.Based on the existing trajectory prediction methods,the mathematical model,data set and evaluation index of trajectory prediction are presented in this paper.Then,the current research of trajectory prediction is introduced,and the data-driven trajectory prediction methods such as deep learning and reinforcement learning are reviewed.Meanwhile,the network model structure,advantages and applicable scenarios of existing research methods are compared and analyzed with the view of space-time modeling,interactive information and scene context.Finally,the research trend of multi-agent and multi-modal trajectory prediction and the future trend of trajectory prediction combined with large language model are further discussed.关键词
自动驾驶/轨迹预测/数据驱动/数据集/评价指标Key words
automatic driving/trajectory prediction/agent/data set/evaluation index分类
信息技术与安全科学引用本文复制引用
杨智勇,杨俊,欧明辉,周瑜..智能网联汽车轨迹预测研究现状与趋势[J].重庆理工大学学报,2024,38(17):1-13,13.基金项目
重庆市自然科学基金项目(cstc2021ycjh-bgzxm0088) （cstc2021ycjh-bgzxm0088）
重庆市高校创新研究群体项目(CXQT21032) （CXQT21032）
重庆市教育委员会科学技术研究计划项目(KJZD-M202303401) （KJZD-M202303401）
