PHEV energy management method and system based on visual traffic information

The invention relates to a PHEV energy management method and system based on visual traffic information, and the system comprises an upper layer perception module which is used for extracting effective traffic information in collected image information through employing a YOLO image recognition model with a pre-trained visual traffic information weight; the speed planning module is used for carrying out speed trajectory prediction according to the effective traffic information; and the lower-layer energy management module is used for performing optimal control quantity prediction based on the planned target vehicle speed track to realize energy real-time management. Compared with the prior art, the method has the advantages of accurate traffic information acquisition and efficient PHEV power flow distribution. 本发明涉及一种基于视觉交通信息的PHEV能量管理方法及系统，所述系统包括：上层感知模块，用于采用预训练好视觉交通信息权重的YOLO图像识别模型提取出采集到的图像信息中的有效交通信息；速度规划模块，用于依据有效交通信息进行速度轨迹预测；下层能量管理模块，用于基于规划得到的目标车速轨迹进行最优控制量预测，实现能量实时管理。与现有技术相比，本发明具有交通信息获取精准、PHEV功率流分配高效的优点。