Research On The Land Vehicle Navigation With Single MEMS IMU

Micro-electro-mechanical-system(MEMS)inertial measurement unit(IMU)has been widely used in land vehicle navigation.However,the navigation errors of a low-cost inertial system with the MEMS IMU will accumulate quickly over time in the absence of external reference information.It is of great value to reduce inertial navigation error in many applications.Therefore,an inertial navigation system with MEMS IMU installed on wheels for dead reckoning is proposed in this paper to reduce the inertial navigation error without external reference information.The traditional rotating modulated inertial navigation system needs a high-precision turntable to make the inertial measurement unit rotate at a uniform speed,but the advantages of low cost,low power consumption,small size and light weight of MEMS IMU installed on the high-precision turntable will not exist.MEMS IMU is installed on the wheel,the sensor will be along with the wheel rotation,to study the wheel as the effectiveness of the turntable for MEMS IMU error modulation,this paper studied the MEMS IMU rotation for the effect of different types of sensor error,such as zero bias and scale factor error,nonorthogonal error,zero bias instability,and the angular rate gyro random walk.Secondly,the inertial navigation error of non-uniform rotation and speed data containing Gaussian white noise error is studied.The results show that IMU rotation can reduce the navigation errors caused by bias,bias instability and angular rate random walk.And the inertial navigation error can be reduced moderately,when the non-uniform rotation and rotation speed data contain errors.When the MEMS IMU is mounted on the wheels for navigation,the vehicle azimuth error will be modulated.As the wheel rotates,the projection of gravity on the accelerometer’s non-rotating axis changes sinusoidally.In this paper,the wheel rotation Angle model is established,the wheel rotation Angle is calculated with the accelerometer data,the wheel rotation angular velocity is measured with the gyroscope,and then the extended Kalman filter(EKF)is used to integrate the two to estimate the wheel rotation Angle more accurately,and the virtual speedometer based on MEMS IMU is realized.In traditional inertial navigation system,roll Angle and pitching Angle errors will make the acceleration of gravity in the horizontal plane projection error and position error over time lead to speed rapid growth,and the speed of virtual odometer error and range error depends mainly on EKF estimate of the wheel rotation Angle and angular velocity error,and avoid this problem,so as to reduce the influence of posture error of position error.In this paper,the dead-reckoning navigation system in two-dimensional plane is realized by combining rotation modulation technology,virtual odometer and vehicle attitude constraint,and its error is analyzed theoretically.The 480-second car experiment shows that the proposed navigation scheme reduces the position RMS by 98% compared with the traditional inertial navigation method of SINS combined with vehicle attitude constraint.