Research Of The Fleet Vehicles Driving Position On Real-time Monitoring

The party team is an important form of presentation,often appearing in review ceremonies performance.Fleet vehicles is a common type of team party that is used in many situations.In the process of displaying the fleet vehicles,maintaining the neat and orderly arrangement of the team vehicles will present a spectacular and majestic sight.Especially at the military parade ceremonies,ensuring that the overall equipment team formation is orderly will show the image of the Chinese army’s solemn and well-trained strong army.Therefore,it is very important to ensure that the distance between the vehicles of the platoon is certain and the linearity of the vehicle is running,that is,to keep the vehicle running on a predetermined trajectory.This imposes higher requirements on the driving skills of the driver of the other team’s vehicle,and therefore effective training of the driver is required.To maintain the formation of the fleet vehicle,it is by planning the route and ensuring that the distance between the vehicle and the vehicle will be realized,which requires the driver to conduct effective training early.During the actual training driving of the vehicle,the vehicle may deviate from the intended trajectory and the offset(angle and distance)will occur.The driver should calibrate the track of the vehicle in time according to the deviation.At present in the offset for mainly by the human eye visual estimation by driver,which leads to the deviation information is not accurate,so that the driver also can’t effectively adjust the travel of the vehicle state and timely correction of vehicle position to make it run on the right track.The training process of the fleet vehicle driver is the process of continuously manipulating the vehicle according to the deviation information and correcting the vehicle trajectory,which is a process of accumulating experience.When the obtained offset information is inaccurate,the driver cannot accumulate effective experience during the training process,thus reducing the training effect and affecting the training efficiency.This article aims at the above problems,researches and designs a real-time monitoring method for vehicle driving position based on machine vision technology.It can monitor the angle and distance of the vehicle from the predetermined trajectory in real time,in order to feedback the driver to the offset information that drivers can use to adjust the vehicle’s running trajectory in time.This method can effectively assist driver training and improve training results.The main work of this article is as follows:1.By analyzing the queue relationship between the team vehicles,aiming at the driver’s difficulty in maintaining the array formation during the driving of the vehicle,based on the principle of machine vision,a real-time monitoring scheme for the vehicle’s driving position was designed.The hardware and software planning of the monitoring system implementation platform was also carried out.2.By comparing and analyzing the distance measurement models commonly used in the monocular vision system and combining the needs of the monitoring program,a small-hole imaging model and a ranging model based on sequence images were comprehensively used to establish a ranging model suitable for this study,and Zhang’s calibration camera method was determined.3.Combining related experiments,the image processing techniques related to the monitoring scheme were studied,including image graying,image noise reduction,image segmentation,edge detection,etc.The image processing scheme of this paper was designed.According to the characteristics of the image processed in this study,a new threshold-based image segmentation method was proposed and the feasibility and superiority of the method were verified through experiments.Combined with common body decoration patterns,the features of the mark images were analyzed.And Hough transform and quadrilateral automatic detection methods were used for mark recognition.4.The static measurement experiment of the monitoring system was designed to verify the accuracy of the monitoring algorithm.The measurement error was within an acceptable range,and the effectiveness of the image processing scheme and offset calculation method was verified.An experiment bench was set up to carry out dynamic experiments of the monitoring system to simulate the dynamic state of the vehicle.The accuracy of the overall scheme of the system was tested.The measurement error was within an acceptable range,time-consuming algorithms can meet real-time monitoring requirements It has proved that the monitoring program established in this paper has certain feasibility.