Research On Attitude Estimation Method For Micro Unmanned Aerial Vehicles

Civilian micro unmanned aerial vehicles due to the driving force and load ability is limited, needs to carry attitude and heading reference system of small volume, light quality, low power consumption, and appropriate precision. This paper designed and developed a microminiature low-cost attitude and heading reference system based on ARM embedded microcontroller processor and MEMS inertial measurement module, and in view of the problem of MEMS sensors due to the effect of noise and drift, which were used alone will cause large measuring error, the pose estimation method was put forward based on the steepest descent theory and the complementary filter. The gyroscope, accelerometer and magnetometer multi-sensor data fusion simulation experiment was designed. The effects of the extended kalman filter and the steepest descent method for data correction were compared, respectively from the amount of calculation and convergence, convergence speed, convergence precision. The experimental results show that the EKF and STDM can ensure that the data has good convergence and the convergence precision of both is similar, but the amount of calculation for STDM is only about1/8of EKF, while the convergence speed is nearly four times faster than EKF. So the STDM has higher application value in spacecraft attitude estimation.