基于AKF的UWB与IMU融合定位示教/再现导航算法研究OA

A teaching/playback navigation algorithm for the ultra wide band(UWB)and the inertial measurement unit(IMU)fusion positioning based on the adaptive Kalman filter(AKF)is proposed in this paper.Firstly,the automated guided vehicle(AGV)kinematics and the UWB measurement models are constructed.Then,a dual adaptive noise covariance adjustment strategy is designed:the measurement noise covariance R is dynamically optimized through the chi-square test of the innovation sequence,and the process noise covariance Qis adaptively adjusted according to the motion state sensed by the IMU and the UWB signal loss status.This addresses the limitation of fixed parameters in traditional Kalman filter.A soft-ware and hardware system is built based on the above filtering algorithm,and a teaching/playback navigation algorithm is designed.In the teaching phase,the AGV is manually controlled to acquire the teaching trajectory.In the playback phase,the segmented PI control is used to achieve accurate tracking of the teaching trajectory.Experiments show that the improved teaching/playback algorithm has excellent robustness and trajectory tracking accuracy.