作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2023, Vol. 49 ›› Issue (5): 1-11. doi: 10.19678/j.issn.1000-3428.0065627

• 热点与综述 • 上一篇    下一篇

自动驾驶场景下的轨迹预测技术综述

李雪松, 张锲石, 宋呈群, 康宇航, 程俊   

  1. 中国科学院深圳先进技术研究院, 广东 深圳 518055
  • 收稿日期:2022-08-30 修回日期:2022-11-21 发布日期:2023-01-12
  • 作者简介:李雪松(1999-),男,硕士研究生,主研方向为智能交通、计算机视觉;张锲石、宋呈群,高级工程师、博士;康宇航,助理研究员、博士;程俊(通信作者),研究员、博士。
  • 基金资助:
    国家自然科学基金(U21A20487,U1913202);深圳市科技计划基础研究项目(JCYJ20180507182610734)。

Review of Trajectory Prediction Technology in Autonomous Driving Scenes

LI Xuesong, ZHANG Qieshi, SONG Chengqun, KANG Yuhang, CHENG Jun   

  1. Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, Guangdong, China
  • Received:2022-08-30 Revised:2022-11-21 Published:2023-01-12

摘要: 轨迹预测是自动驾驶和智能交通领域的关键技术,对于车辆和移动行人轨迹的准确预测可提升自动驾驶系统对周围环境变化的感知能力,保障自动驾驶系统的安全性。数据驱动轨迹预测方法可捕捉智能体之间的交互特征,对场景内智能体历史运动和静态环境信息进行分析,准确预测智能体的未来轨迹。介绍轨迹预测的数学模型并将其分为传统轨迹预测方法和数据驱动轨迹预测方法2类,阐述主流数据驱动轨迹预测方法所面临的智能体交互建模、运动行为意图预测、轨迹多样性预测、场景内静态环境信息融合等4个主要挑战,从轨迹预测数据集使用、性能评价指标、模型特点等方面出发对典型数据驱动轨迹预测方法进行分析与对比,总结归纳这些典型数据驱动轨迹预测方法针对上述挑战的解决思路和应用场景,并对自动驾驶场景下轨迹预测技术的未来发展方向进行展望。

关键词: 轨迹预测, 深度学习, 自动驾驶, 数据驱动, 多智能体交互

Abstract: Trajectory prediction is a key technology in the fields of autonomous driving and intelligent transportation. The accurate prediction of trajectories for vehicles and moving pedestrians can improve the perception of environmental changes in autonomous driving systems,thereby ensuring overall safety.The data-driven trajectory prediction method accurately captures interaction characteristics between agents,analyzes the historical motion and static environment information of all agents within a scene,and predicts the agents' future trajectories.The mathematical models of trajectory prediction are introduced and categorized as traditional and data-driven trajectory prediction methods.The four main challenges faced by mainstream data-driven trajectory prediction methods include intelligent agent interaction modeling,motion behavior intention prediction,trajectory diversity prediction,and static environmental information fusion within a scene.Herein,starting from the use of trajectory prediction datasets,the performance evaluation indicators,model characteristics,and other aspects of typical data-driven trajectory prediction methods are analyzed and compared. On this basis,the solutions and application scenarios of the said methods to address the abovementioned challenges are summarized,and future development directions of trajectory prediction technology in autonomous driving are proposed.

Key words: trajectory prediction, deep learning, autonomous driving, data-driven, multi-agent interaction

中图分类号: