Research And Development Of Intelligent Vehicle Brake Test System

In order to make the car in an emergency being braked in time to reduce casualties,the national standards provide for the car braking performance testing,measuring the car braking performance of the criteria include the average deceleration(MFDD)issued sufficiently,braking coordination time,braking initial speed,braking distance and other parameters.In this paper,on the basis of fully studying the design principle of existing automobile braking detection system,we adopt image processing technology to design an intelligent automobile braking detection system that independently collects,processes and transmits data,and the main research contents are as follows.(1)Design a research plan for the automobile braking detection system.Based on the full study of the measurement principle of the fifth wheel meter and the design of the portable car brake detector,the camera is used to take photos of the road surface,and the displacement between two adjacent images is calculated by the image processing algorithm so as to measure braking parameters such as braking distance,which is transmitted to the cell phone client through Bluetooth communication technology to realize the storage,management and uploading of data to the remote service.(2)Theoretical Research on Displacement Measurement Recognition Method Based on Computer Vision.The feature optical flow technology based on feature points is studied to solve the basic equation of optical flow.The SURF feature point algorithm is combined with the optical flow method.The SURF feature point algorithm is used to purify the multi-constraint feature points,and the singular value decomposition method is used to effectively eliminate the wrong matching points.Research based on the characteristic points recognition of sub-pixel corner detection algorithm of tracking technology,the close neighbor ratio purification,positive and negative two-way purification and direction Angle of purification methods optimize the Harris corner matching algorithm,the more accurate angular point matching,and combined with the least squares fitting ellipse center can enhance the accuracy of angular point,quadratic interpolation points are extracted for screening the sub-pixel coordinates,through coordinate transformation formula for measuring the braking distance and braking parameters.(3)Automotive brake detection system software and hardware design.The system consists of a hardware part and a software part.The hardware part includes a microprocessor(STM32F107VCT6),a Bluetooth communication module(HC-05)and a CCD visual inspection camera.The microprocessor receives the signal of brake pedal closing,sends the command of image acquisition to the camera and transmits the data of braking parameters through the Bluetooth module.The Bluetooth communication module realizes the data transmission between the microprocessor and the cell phone client.The CCD visual detection camera acquires the images of braking process.The software part includes the embedded system software and the mobile phone client software.The embedded system software includes image processing algorithm to process images,Bluetooth to transmit data and control the camera to capture images.The cell phone client software provides a good interface to realize the functions of data storage,management and remote service to upload data.(4)System test and result analysis.The image processing algorithm was tested in different environments by setting the type of road,changing the distance of the camera from the ground and braking at different initial speeds,and it was concluded that the braking parameters should be measured at a normal road without cracks,a camera distance of 20 CM from the ground and braking at 20Km/h-40Km/initial speed and using the sub-pixel corner point detection algorithm to be more accurate.