基于深度学习的行人轨迹预测方法综述
作者:
作者单位:

青岛科技大学 信息科学技术学院,山东 青岛 266061

作者简介:

通讯作者:

E-mail: lihui@qust.edu.cn.

中图分类号:

TP391

基金项目:

国家自然科学基金项目(61702295,61672305).


Survey of pedestrian trajectory prediction methods based on deep learning
Author:
Affiliation:

College of Information Science and Technology,Qingdao University of Science and Technology,Qingdao 266061,China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为了规划合理的路径以规避行人,针对行人轨迹预测的研究具有广泛的应用价值.基于手工特征的传统方法难以预测复杂场景下的行人轨迹.深度学习以人工神经网络为架构,具有强大的学习能力,在各个领域取得了显著的效果.基于深度学习的行人轨迹预测方法已逐渐发展为一种趋势.为了宏观把握基于深度学习的行人轨迹预测的研究状况,首先,对不同方法进行组织与分类,比较不同方法的优缺点,讨论不同方法在行人轨迹预测领域的应用与发展;其次,根据行人轨迹预测模型的设计差异,对比不同算法对模型性能产生的影响;最后,针对行人轨迹预测中存在的问题,对基于深度学习的行人轨迹预测方法的未来发展进行了展望.

    Abstract:

    In order to plan a reasonable path to avoid pedestrians, the research on pedestrian trajectory prediction has a wide range of application value. Traditional methods based on manual features are difficult to predict pedestrian trajectory in complex scenes. Deep learning is based on artificial neural networks, which has strong learning ability and has achieved remarkable results in various fields. The pedestrian trajectory prediction method based on deep learning has gradually developed into a trend. In order to grasp the research status of pedestrian trajectory prediction based on deep learning, firstly, different methods are organized and classified, their advantages and disadvantages are compared, and the application and development of these methods in the field of pedestrian trajectory prediction are discussed. Then, according to the design differences of pedestrian trajectory prediction models, effects of different algorithms on the model performance are compared. Finally, in view of existing problems in pedestrian trajectory prediction, the future development of pedestrian trajectory prediction method based on deep learning is prospected.

    参考文献
    相似文献
    引证文献
引用本文

孔玮,刘云,李辉,等.基于深度学习的行人轨迹预测方法综述[J].控制与决策,2021,36(12):2841-2850

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2021-11-18
  • 出版日期: 2021-12-20