基于四元数的四旋翼无人机扩展卡尔曼滤波算法

Quaternion-based Extended Kalman Filtering Algorithm for Quadrotor UAVs

四旋翼无人机因体积小、灵活机动、成本低等诸多优势而得到广泛应用。但因其是一个欠驱动、强耦合的非线性系统,执行任务时会受到加速度和磁场干扰,对姿态解算产生严重影响。为了提高无人机姿态的控制精度,提出一种适用于小型无人机姿态估计的四元数扩展Kalman滤波算法,该算法通过建立四元数姿态运动模型和航姿传感器测量模型,解决了微机电系统用于载体姿态测量时精度低、易被干扰的问题,有效提高了姿态估计的精度;最后,通过仿真实验验证了所提算法的有效性。

Quadrotor UAVs are widely used due to their small size, flexibility and low cost. However, because it is an under-driven, strongly coupled nonlinear system, it will be disturbed by acceleration and magnetic field when performing tasks, which will have a serious impact on attitude calculation. In order to improve attitude control accuracy, a quaternion extended Kalman filtering algorithm for the attitude estimation of small UAVs is proposed. The algorithm solves the problem of the use of micro electro mechanical system(MEMS) by establishing a quaternion attitude motion model and a heading and attitude sensor measurement model. Due to solving the problem of low accuracy and easy interference during carrier attitude measurement, the accuracy of attitude estimation is effectively improved. Finally, the effectiveness of the proposed algorithm is verified by simulation experiments.