The invention relates to a multi-parameter cooperative control locomotive running and positioning method, which comprises the following steps of: S1, acquiring a traffic flow picture data set in a mixed traffic environment, and then preprocessing the traffic flow picture data set; s2, dividing the picture data set preprocessed in the step S1 into m blocks according to the number of subqueues, and generating a corresponding number estimation density map by using a residual multi-scale neural network; s3, integration is carried out based on the density map obtained in the step S2 to obtain the finally estimated number n-1 of the manually driven vehicles in each sub-queue; and S4, collecting a traffic flow video data set in a mixed traffic environment, and then performing preprocessing. The invention relates to the technical field of locomotive control. According to the method, the vehicle motion state on the mixed road where the manual driving vehicle and the unmanned vehicle coexist can be analyzed, so that the control performance of a motorcade is improved, the driving efficiency and the driving safety are improved, the number of parameters in the model and the calculation complexity can be reduced, and the prediction efficiency is improved.
本发明涉及一种多参协同控制的机车走行与定位方法,包括以下步骤:S1、采集混合交通环境下的车流图片数据集,然后进行预处理;S2、将步骤S1预处理的图片数据集按照子队列的个数划分为m块,使用残差多尺度神经网络生成对应的数量估计密度图;S3、基于步骤S2得到的密度图进行积分得到最终估计的每个子队列中人工驾驶车辆的数量n‑1;S4、采集混合交通环境下的车流视频数据集,然后进行预处理。本发明涉及机车控制的技术领域。本发明能够分析人工驾驶车辆与无人驾驶车辆共同存在的混合道路上的车辆运动状态,从而提高了车队的控制性能,进而提高行驶效率和行驶安全性,且能够减少模型中的参数数量和计算复杂度,提高预测效率。
Multi-parameter cooperative control locomotive running and positioning method
一种多参协同控制的机车走行与定位方法
2024-06-11
Patent
Electronic Resource
Chinese
LOCOMOTIVE RUNNING STATE DETECTION METHOD BASED ON SATELLITE POSITIONING
European Patent Office | 2025
|LOCOMOTIVE RUNNING STATE DETECTION METHOD BASED ON SATELLITE POSITIONING
European Patent Office | 2022
|Engineering Index Backfile | 1915
|Engineering Index Backfile | 1910