The invention discloses a vehicle track prediction method considering a vehicle interaction relationship transfer characteristic under a CPS, and the method comprises the steps: S1, enabling an intelligent vehicle to jointly sense the information of surrounding vehicles, and mapping a vehicle interaction relationship in a physical space to an information space for representation; s2, taking the dynamic graph as the input of a prediction model, and carrying out position coding on input elements; s3, collaborative extraction of spatial and temporal features is carried out on the encoded feature sequence through a multi-head attention network; s4, track prediction is carried out through a coder-decoder based on a gating circulation unit; and S5, feeding a prediction result as an information feature back to the dynamic graph, fusing the physical feature and the information feature, and realizing track prediction considering vehicle interaction relationship transmission. According to the method, the interaction relationship transfer characteristics between the vehicles are comprehensively considered from the perspective of information physics, the interaction influence between the vehicles can be dynamically captured, and the accuracy of trajectory prediction is effectively improved.
本发明公开了一种CPS下考虑车辆交互关系传递特性的车辆轨迹预测方法,包括:S1.智能汽车联合感知周围车辆信息,将物理空间的车辆交互关系映射到信息空间进行表征;S2.以动态图作为预测模型的输入,对输入元素进行位置编码;S3.通过多头注意力网络对编码后的特征序列进行时空特征的协同提取;S4.通过基于门控循环单元的编码解码器进行轨迹预测;S5.将预测结果作为信息特征反馈到到动态图中,融合物理特征和信息特征,实现考虑车辆交互关系传递的轨迹预测。本发明通过以信息物理视角综合考虑车辆间的交互关系传递特性,可动态捕捉车辆间的交互影响,有效提高了轨迹预测的准确性。
Vehicle trajectory prediction method considering vehicle interaction relationship transfer characteristics under CPS
一种CPS下考虑车辆交互关系传递特性的车辆轨迹预测方法
2024-10-01
Patent
Electronic Resource
Chinese
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06V |
European Patent Office | 2023
|Peripheral vehicle trajectory prediction method based on vehicle trajectory interaction dynamics
European Patent Office | 2024
|Vehicle Lane-Changes Trajectory Prediction Model Considering External Parameters
BASE | 2021
|Vehicle Lane-Changes Trajectory Prediction Model Considering External Parameters
DOAJ | 2021
|Vehicle lane changing intention prediction method considering vehicle complex interaction
European Patent Office | 2024
|