Font Size: a A A

Research And Design Of The Embedded-Based Location For Vehicle

Posted on:2016-04-20Degree:MasterType:Thesis
Country:ChinaCandidate:Z L GaoFull Text:PDF
GTID:2322330542973749Subject:Precision instruments and machinery
Abstract/Summary:PDF Full Text Request
With the production of Global Position System(GPS),vehicle navigation and location system developed rapidly,and become a great technological revolution in the vehicle navigation field.Based on the background of GPS/DR vehicle navigation system,this thesis researches the related technologies to achieve the vehicle location under VxWorks operating system for the purpose of improving the vehicle location accuracy.Conventional Dead Reckoning composed by gyroscope,accelerometer,odometer,electronic compass and so on.With the development of Micro Electro-mechanic System s,MEMS inertial sensors used in vehicle location are keen on people.In this thesis,the part of DR navigation uses three MEMS sensors(L3G4200D,three axis gyroscope;H MC5883 L,three axis magnetometer;AXDL345,three axis accelerometer)to compute t he attitude and position.Because the MEMS sensor's error is large,the thesis first to ana lyze the error and build modes,and then the deterministic error was corrected,the rando m error removed by the wavelet denoising technology to get the processed data.These methods effectively reduce the positioning error in DR part and improve the accuracy of DR part.The traditional attitude determination method is achieved by the gyroscope and accelerometer.The error of the pitch angle and roll angle is small,but there is a big error in the heading angle.This thesis has joined the magnetometer,effectively reducing the heading angle's error.It uses accelerometer to determine the pitch angle and roll angle and uses magnetometer to determine the heading angle,thus calculate the real-time updates of the attitude angle and attitude matrix.The final realization of positioning adopts The Federal Kalman filtering method.This system Composed of 3 sub filters and 1 main filter.The part of GPS uses ordinary Kalman filtering method as the filter 1;the part of DR constituted by gyroscope and accelerometer uses UKF as the sub filter 2,because the measurement equation is nonlinear;the second part of DR constituted by magnetometer and accelerometer makes up the filter 3.Finally it realizes information fusion using the main filter,realizes the optimal estimation of the positioning of the system.The unique design of the DR part,enhance the fault tolerance,also improve the accuracy of the system.Finally the hardware rescources of the system are given,the Vx Works operating syst em is transplanted to the Samsung S3C2410 processor,and designed programs according to the flow chart of software.Finally,the feasibility and superiority of the positioning a ccuracy of the system is verified through the field experiment.
Keywords/Search Tags:GPS, DR, VxWorks Embedded Operating System, MEMS Sensor, the Federal Kalman Filter
PDF Full Text Request
Related items