Research On Vehicle Tracking Algorithm UKF And Wavelet Transform

Today,in the highly developed economy,the car as a kind of more high-speeding transport has gradually replace the original battery car.It becomes a travel tool for people to travel.But there is a serious problem in front of us that is the traffic problems.As a result of the vehicle is various,led to the complexity of the road traffic,and in such a complex road traffic it is easily occurred the traffic accidents.Therefore the establishment of the intelligent transportation system becomes an imminent thing in today's society.But the target tracking of the vehicle as part of intelligent transportation system have a great significance to reduce traffic congestion and traffic accidents.This paper uses the algorithm of based on combination of unscented Kalman filter and wavelet transform to realize the vehicle tracking.Firstly,the advantages and disadvantages of several Kalman tracking algorithms are compared in theory.The algorithm of tracking the Kalman filter is chosen as the target tracking.And through the simulation of MATLAB is used to verify the superiority of the unscented Kalman filter algorithm.In order to improve the accuracy of vehicle tracking,this paper adds multi-scale analysis of wavelet transform on the basis of the unscented Kalman filter algorithm.The principle is that the processed signal is decomposed on the scale N,and the smoothing signal on the different scale i is decomposed by the unscented Kalman filter to obtain the optimal estimation of the smoothing signal.Then the optimal estimation signal of the obtained smoothing signal is reconstructed to obtain the optimal estimate on the scale i,and then the optimal estimate on different scales is fused to obtain the optimal estimate of the whole system.And it is simulated by MATLAB.Finally,the results of two kinds of tracking algorithms are analyzed,which proves that the algorithm of combination of unscented Kalman filter and wavelet transform has less error in vehicle tracking and higher accuracy.In order to adapt to the radar signal high-speed acquisition,the system use FPGA as the system hardware control chip on hardware.The hardware design of the whole system includes three modules:data acquisition module,data processing module and display module.The flow of each module is:first through the laser radar to obtain the vehicle's position and speed information,and then through the A/D conversion circuit to achieve the conversion signal from the analog signal to digital signal,then converted the digital signal into the FPGA Data processing,finally showing the results of after processing optimal estimate.On the basis of the hardware design of the system,the Verilog language is used to program for each hardware interface of the system,and the algorithm is programmed in software.Finally,the building hardware platform use Quartus ii software to download the program and debugging.