自适应混合滤波算法在微型飞行器姿态估计中的应用

Application of Adaptive Hybrid Filter Algorithm in the Estimation of the Micro Air Vehicle Attitude

针对低成本惯性测量单元(IMU)存在漂移和噪声干扰等问题,提出了一种具有自适应参数调节的混合滤波算法。采用四元数法进行系统模型的描述,用梯度下降法对加速度计测得的数据进行处理,再通过互补滤波器将其与陀螺仪测量值进行融合,形成混合滤波算法。同时,考虑到飞行姿态的复杂性,进行参数λ的自适应调节,因而改进后的混合滤波算法,能保证各种飞行姿态变化情况下实时姿态的最优估算。实际系统在线实时性能测试表明,提出的算法简单,估计精度高,易于在嵌入式系统中实现,具有较高推广应用价值。

Concerning the drift and noise interference of low cost inertial measurement unit, a hybrid filtering algorithm with adaptive adjustment of parameters was proposed. With the quaternion for describing the attitudes, the accelerometer data is processed using gradient descent algorithm. And then the results are fused with Gyro measurements through the complementary filter, which is called the mixed filter algorithm. At the same time, considering the complexity of flight attitude, the parameters can be adaptively adjusted. So the improved hybrid filter algorithm can guarantee the optimal attitude estimation in real time for various flight attitudes. The online test results of the real-time system show that the proposed algorithm simple to realize and has high estimation accuracy. It is especially suitable for implementation on embedded hardware which has high application value.