The invention discloses a multi-modal trajectory prediction method, and the method specifically comprises the following steps: S1, carrying out the preprocessing of multi-modal data, and unifying the historical trajectory of an automatic driving vehicle, the historical trajectory of a target, and the information of nearby lanes into a target coordinate system; s2, encoding the preprocessed data, and mapping multi-modal input to a hidden layer; and S3, lane sensing selection is carried out on the coded hidden layer. According to the trajectory prediction method considering the interaction of the autonomous vehicle and the scene obstacle, the autonomous vehicle and the obstacle in the scene are unified as an intelligent moving body, and the safety of target trajectory prediction in a complex scene is improved; a lane perception estimation method is provided, the incidence relation between a target and a lane in a map is recognized, a high-possibility lane is extracted, the method is beneficial to guiding a target prediction track to the most probable lane, and the accuracy of target track prediction is improved.
本发明公开了多模态轨迹预测方法,具体包括以下步骤:步骤S1,对多模态数据进行预处理,将自动驾驶车辆的历史轨迹、目标的历史轨迹、附近车道信息统一至目标坐标系;步骤S2,对预处理后的数据进行编码,将多模态输入映射到隐藏层;步骤S3,对编码后的隐藏层进行车道感知选择。本发明考虑自动驾驶车辆与场景障碍物交互的轨迹预测方法,将自动驾驶车辆与场景中的障碍物统一为智能运动体,提高了复杂场景下目标轨迹预测的安全性;提出了车道感知估计方法,识别目标与地图中车道的关联关系、提取高可能性车道,这种方法有助于将目标预测轨迹引导至最具可能性的车道上,提高目标轨迹预测的准确性。
Multi-modal trajectory prediction method
一种多模态轨迹预测方法
2023-12-29
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 |
Multi-modal vehicle trajectory prediction method based on hierarchical order network
European Patent Office | 2023
|Interactive vehicle multi-modal trajectory prediction method based on GRU-GCN
European Patent Office | 2023
|