Research On The Key Technology Of Intelligent Electric Vehicle Positioning Based On GFS/SINS

Intelligent vehicles include intelligent electric vhiccles,They mainly include three aspects: environmental perception,target decision and behavior control.The technology of positioning navigation and path tracking are respectively an indispensable part of sensing and decisionmaking module,the accuracy of which will directly affect the driving safety and economy.Due to the reason that the update frequency is low and the signals are easily blocked in GPS and that the error is easy to accumulate over time despite with high frequency in SINS,the combined navigation of GPS and SINS are mostly applied in practical application.However,there is little literature on the design of the trajectory simulator in the integrated navigation and the construction of the SINS simulation model.In addition,although many path tracking algorithms have achieved better path tracking effects,most of them are in costant spped,and few can automatically change the speed with the road curvature during turning.Besides,while driving in higher speeds,accidents are easy to happen in the corner without slowing down.Therefore,in order to better realize the adaptive cruise of intelligent electric vehicles,this paper go on the following researches for those problems by taking an intelligent electric vehicle as the experimental carrier.(1)In order to obtain the plane coordinates of GPS in BJ-54 coordinate system in real time,analyze the navigation principle,related coordinate systems and the protocol of NMEA0183 at first.Then based on the dual-antenna differential GPS-Trimble BD982,design a method to covert latitude and longitude coordinates in WGS-84 coordinate system directly acquired by serial port to plane rectangular coordinates in BJ-54 coordinate system.Besides,the two attitude angles involved in the dual antenna are also calculated.(2)On the basis of studying the commonly used coordinate systems and the relationship between those in SINS,the SINS simulation model is constructed by direction cosine method and quaternion method respectively.Besides,analyze the relavent principles and the determination of navigation parameters in detail.A method to design trajectory simulator based on Gaussian projection inverse calculation principle is proposed,and the corresponding acceleration and attitude angle in three axes are input into the SINS model for error analysis,which lays the foundation for the design of GPS/SINS integrated navigation system.(3)The Kalman filter algorithm,loose combination and indirect method are used to design the GPS/SINS integrated navigation system model.Besides,based on the trajectory simulator,the GPS,the accelerometer in SINS and gyroscope in SINS are simulated respectively.Then,the simulated three-dimensional acceleration and angular velocity values are input into the SINS simulation model to obtain the SINS navigation parameters,and the navigation parameters of GPS with the failure condition,SINS and integrated navigation system are compared and analyzed based on the Kalman filter algorithm.Then,instead of the simulator of trajectory,collecting real-time information of GPS and SINS while driving along a road to verify the effectiveness of the integrated navigation system in real vehicle application.(4)Propose an improved Pure Pursuit algorithm to track the trajectory based on the kinematic bicycle model of the vehicle.It takes the look-ahead distance,velocity and curvature into consideration.Besides,it can adjust the velocity and the look-ahead distance adaptively according to the trajectory curvature to improve the precision of trajectory tracking.