The invention discloses a robust trajectory prediction method and system based on a decoupling space-time network. The method comprises the steps that a vehicle trajectory data set is acquired and preprocessed; constructing a decoupling space-time network model; and inputting the preprocessed vehicle track data set into the decoupling space-time network model to obtain a vehicle track future signal. According to the method, a reliable prediction result can be provided no matter whether historical signals are disturbed or not.
本发明公开了一种基于解耦时空网络的鲁棒轨迹预测方法及系统,包括:获取车辆轨迹数据集并进行预处理;构建解耦时空网络模型;将预处理后的所述车辆轨迹数据集输入所述解耦时空网络模型,获得车辆轨迹未来信号。本发明不论在历史信号受到扰动与否,都能提供可靠的预测结果。
Robust trajectory prediction method and system based on decoupling space-time network
一种基于解耦时空网络的鲁棒轨迹预测方法及系统
2025-01-14
Patent
Electronic Resource
Chinese
European Patent Office | 2025
|Traffic index prediction method based on decoupling space-time neural network
European Patent Office | 2024
|Distributed traffic flow prediction method and system based on space-time decoupling
European Patent Office | 2023
|Space-time heterogeneous decoupling method based on multi-mode traffic flow prediction
European Patent Office | 2024
|Vehicle trajectory prediction method based on time-space fusion attention
European Patent Office | 2025
|