An improved square root unscented Kalman filter for projectile's attitude determination

摘要

In this paper, considering the performance of short-time working, inertial flight and high impulse of launch for projectile, an improved square root unscented Kalman filter (UKF) is used. According to the special motional rule of projectile, only use three low-cost MEMS accelerometers as inertial measurement unit to measure the specific force, and construct the linear state equation and nonlinear measurement equation of nonlinear system. Use the square root UKF to estimate the state of system, which effectively enhances the stability of filter and restrains the divergence of filter. Reduce the number of sigma points from 2n+1 to n+2, which improves the computational efficiency and real-time performance of filter. The experimental results in a three-axis flight test rotary table show that the improved square root UKF obtains the good precision of estimation and satisfies the request of real-time response.