基于时空数据融合技术的人员轨迹预测系统设计
Design of Personnel Trajectory Prediction System Based on Spatio-temporal Data Fusion Technology
李洪杰 1董经纬1
作者信息
- 1. 中国人民解放军陆军工程大学,江苏 南京 210007
- 折叠
摘要
由于人类活动的随机性和主观性,当前预测一个人的未来轨迹仍然是一个具有挑战性的问题.文章针对目标多源历史轨迹数据存储、融合预测等问题,设计一种基于时空数据融合技术的人员轨迹预测系统.结合需求分析提出了系统的基本框架,构建了底层数据库;基于犹豫模糊集实现了多源轨迹数据的融合;基于多尺度轨迹空间LSTM集成预测模型实现了目标轨迹的预测.
Abstract
Due to the randomness and subjectivity of human movement,predicting a person's future trajectory remains a challenging problem.This paper designs a personnel trajectory prediction system based on spatio-temporal data fusion technology to address the storage,fusion,and prediction of target multi-source historical trajectory data.Based on requirement analysis,the basic framework of the system is proposed and the underlying database is constructed;the fusion of multi-source trajectory data is achieved based on hesitant fuzzy sets;the prediction of target trajectory is achieved based on a multi-scale trajectory space LSTM integrated prediction model.
关键词
轨迹预测/犹豫模糊集/LSTMKey words
trajectory prediction/hesitant fuzzy set/LSTM引用本文复制引用
基金项目
2021年度江苏省重点研发计划(BE2021729)
出版年
2024
国家科技期刊平台
NSTL
万方数据