| Positioning,Navigation and Timing(PNT)system is an important development direction of positioning,navigation and timing technologies.Global Navigation Satellite System(GNSS)is the most widely used PNT technology,which can provide users positioning,navigation and timing services with all-weather,all-day and highprecision.However,in the complex environment,satellite navigation signals suffer from the shield by surrounding obstacles,multipath effect or electromagnetic interference,which may lead to the severely attenuated power,declined signal quality or outage.As a result,GNSS measurement values will be abnormal or even unavailable.Multi-sensor fusion is an effective method to enhance the performance of navigation system.Aiming at the application scenario of complex GNSS environment,this paper describes the development and research status of GNSS navigation technology,strapdown inertial navigation technology,chip atomic clock technology and integrated navigation technology.Based on Microelectro-mechanical System Inertial Measurement Unit(MEMS IMU),GNSS and Chip Scale Atomic Clock(CSAC),the research on key technologies of vehicle-mounted integrated navigation has carried out.The main research contents and innovations include:1.In view of the possible interference or shield of GNSS signals in navigation scenes,the INS/GNSS integrated navigation adaptive robust filtering method is studied.The conventional adaptive filtering model of tightly coupled integrated navigation was improved to construct a robust Kalman filtering model.According to the changes of GNSS observation conditions,the adaptive factor vector was calculated in real time to adjust the measurement noise of the filter system and improve the integrated navigation solution accuracy.In the case of redundant GNSS observation,the function of error detection and identification realizes the error detection and separation of pseudo-range and pseudo-range rate to improve the accuracy of measurement input of integrated navigation system.The effectiveness of the proposed method is tested by experimental data.2.In view of the defect that the MEMS INS/GNSS integrated navigation system is degraded to an independent INS navigation system and the speed of error accumulation is faster due to the complete outage of GNSS signals,a navigation error constraint method based on vehicle kinematics and operating environment is studied.Based on the knowledge of dynamic or physical conditions of land vehicles,two error constraint models are constructed: The one is the velocity constraint model based on vehicle motion status;the other is the position constraint model based on the vehicle operating environment.Each model can be used independently or combined together,which can effectively reduce the divergence rate of the solution error of inertial navigation system without adding the external sensor.3.Based on the high stability of CSAC in time and frequency domain in a short term,the CSAC aided integrated navigation method is studied,and the model based on this method is constructed.The error detection and correction of GNSS measurement information of integrated navigation system can be carried out in the case of redundant observation,so as to reduce the adverse effects of abnormal measurement values on navigation solutions.In the case of insufficient observation satellites,the prediction value of clock bias by CSAC clock model is used to improve the accuracy of measurement input of the integrated navigation filter,so as to suppress the error accumulation of MEMS INS.4.To overcome the challenge of integrated navigation system in complex GNSS environment,a tightly coupled multi-sensor integrated navigation method was proposed,based on which an integrated navigation system model was designed and a vehicle-mounted experiment platform was developed.At the hardware level,the highly stable frequency reference of CSAC is used to improve the observation accuracy of GNSS pseudo-range and pseudo-range rate.At the algorithm level,based on the Kalman filter model of tightly coupled integrated navigation,the error detection and identification method is used to reduce the influence of gross errors on the integrated navigation solution in the condition of observation redundancy.The velocity constraint model based on motion status is used to suppress the accumulated velocity of navigation solution error when the the number of visual satellites is insufficient or GNSS signal is completely interrupted.A Kalman adaptive filter was constructed to adjust the filtering parameters in dynamic environment in real time and improve the reliability of navigation solutions.The CSAC receiver clock error model is used to monitor and correct the measurement value of clock bias,so as to improve the accuracy of receiver clock bias.The effectiveness of the multi-sensor tightly coupled integrated navigation method was tested by designing and carrying out a field experiment.The experimental results show that the method has better availability and reliability in both complex GNSS environment and conventional environment.Compared with the conventional tightly coupled integrated navigation method,the integrated navigation solution via multi-sensor tightly coupled integrated navigation method has a better accuracy in position,velocity,attitude and precision time measurement,which effectively enhance the PNT performance. |
PDF Full Text Request