Multi-Rate Kalman Filter for UAV Attitude Estimation Based on Adaptive Disturbance Model

Weibin Feng,Caiqi Wang,Shiqian Wu,Qi Wu,Hongsen Zhu
DOI: https://doi.org/10.23919/ccc63176.2024.10661750
2024-01-01
Abstract:Attitude estimation is a pivotal issue in unmanned aerial vehicle (UAV) control system. Due to its high dynamics, the current attitude estimation methods suffer from drift and instability in dynamic situations. In this paper, multi-rate Kalman filter is proposed to estimate the attitude i.e., roll and pitch of UAVs using an Inertial Measurement Unit (IMU) consisting of a threeaxis gyroscope and accelerometer, in conjunction with an optical flow sensor. To mitigate the effects of external acceleration disturbances, a first-order Gauss-Markov process is employed to model the disturbances and combines it with a Kalman filter to eliminate the disturbances. A sigmoid function is introduced to map the relationship between the magnitude of the external disturbance and the model coefficient to achieve adaptive adjustment of the model coefficient. Furthermore, the differential information of the optical flow velocity is used as an additional observation, and a multi-rate Kalman filter is presented based on the different update frequencies of the observation sensors. The estimated gravity acceleration is then used to calculate the attitude via the fast complementary filtering (FCF) algorithm. Experimental results on STM32 microcontroller show that the proposed method can effectively improve the accuracy of attitude estimation and anti-interference ability of UAVs in motion state.
What problem does this paper attempt to address?