| With the development of automatic driving and other technologies,the demand for accurate vehicle positioning in indoor scenes is increasing day by day.Indoor parking lot is one of the main scenes of vehicle driving at present.However,in this scene,there are many problems,such as limited space,dense built-in facilities,and unstable light source signal.Even if the positioning scene and location scene are greatly modified,the overall positioning accuracy of the system is not high because the positioning process is vulnerable to environmental noise interference,which seriously affects the precise positioning of vehicles.In order to overcome the shortcomings of existing indoor positioning technology,the study chooses the platform of vehicle inertial navigation technology based on autonomous positioning,which integrates the vehicle odometer and the inherent environmental characteristics of the indoor parking lot to achieve the correction of the accumulated error of the inertial navigation positioning system,so as to improve the parking positioning accuracy of the vehicle in the indoor parking lot environment.The project research has completed the following works.(1)In order to facilitate the study of vehicle positioning characteristics,based on the built vehicle positioning reference coordinate system,it studied the basic principle of INS positioning on the track estimation model,constructed INS attitude calculation model based on relative positioning,made the initial calibration of INS,finally also tested and analyzed the pure INS positioning effect.(2)It constructed the kinematic model based on vehicle odometer,designed the scheme of integrated positioning of INS and odometer,and used EKF algorithm to integrate the dead reckoning information of INS and vehicle odometer to further improve the positioning accuracy of the system.At the same time,in order to calibrate the slippage error of the odometer system,it designed an adaptive odometer slip error compensation algorithm.The high-precision angular velocity information from the inertial navigation system was used to realize the self-tuning of the corresponding compensation parameters.(3)It designed a global position correction scheme based on ultrasonic sensor.In view of the divergence of the relative positioning results in the long-term positioning process,it proposed to extract the inherent environmental characteristics of the indoor parking lot by using a simple vehicle borne ultrasonic sensor device,and used split merge algorithm to fit the extracted features with line segments.With the help of the course angle information and feature matching results obtained in the process of relative positioning,the global position of the vehicle can be updated,finally verified the feasibility of the positioning scheme by experiment.(4)In order to solve the problem of accurate decision-making in the process of multi-source location information fusion,it studied the multi parameter model building strategy of different precision location information in the process of fusion,and proposed an improved UKF algorithm to ensure the accuracy and efficiency of information fusion,so as to realize the data fusion of multi source location information,and complete the optimal location state estimation of positioning system,finally also verified the effectiveness of the applied fusion algorithm.Finally,it used the scheme designed in this paper to test and analyze the vehicle location in the actual indoor parking environment.The test results show that the scheme designed in this paper can better improve the positioning accuracy of vehicles in the indoor parking lot. |
