一种旋翼无人机组合导航系统设计及应用

针对传统组合导航算法相对复杂,对系统硬件要求太高,实时性不好的缺点,设计了一种基于线性卡尔曼滤波器的简单实用组合导航算法.考虑到传统卡尔曼滤波器在系统噪声变化时滤波精度变差甚至发散以及模糊控制器计算量大的缺陷,设计了线性迭代调整观测噪声参数的方法对滤波器进行自适应改进,提高了算法的鲁棒性和实时性.利用STM32微控制器和MEMS惯性单元以及UBLOXGPS定位模块搭建硬件平台进行实验验证.结果表明:在168MHz时钟频率下,一次姿态数据读取和解算共耗时3.27ms,一次组合导航滤波耗时2.18ms,二者分别运行在100Hz和5Hz频率下.最终组合导航结果可以满足无人机在1m的精度范围内自动悬停的需求,验证了所设计的组合导航算法的可靠性和实用性.

A Kind of Design and Application of Integrated Navigation Systems for UAVs

Aiming at the disadvantages that the traditional integrated navigation algorithm has high complexity,high hardware requirements,and poor real-time performance,a simple and practical integrated navigation algorithm based on linear Kalman filter is designed.Considering the problems that the filtering accuracy will become terrible or even divergence when the system noise varies,and great computational burden of the fuzzy controller,a method to adjust observation noise parameters adaptively based on linear iteration is proposed to improve the adaptive performance and robustness of the filter.Then,the proposed method is verified in an experimental system that consists of the STM32 microcontroller,the MEMS inertial unit and the UBLOX GPS positioning module.The experimental results indicate that,the time of reading attitude data and attitude solution costs 3.27 ms in total running at 100 Hz,and the time of integrated navigation filtering costs 2.18 ms in total running at 5 Hz,both under 168 MHz clock frequency.The final integrated navigation results can meet the needs of the UAVs automatic hover in the precision of 1 m and verify the reliability and practicability of the designed integrated navigation algorithm.