Sequential square root filtering for measuring tractor driving wheel slip rate

Abstract

Adaptive sequential square root Kalman filtering (ASSRKF) algorithm is purposed to measure slip rate of wheel tractor online. The filtering process is formulated as a process of recursive the Kalman state model, where signals from wheel speed sensors, angular acceleration, GPS and accelerometer are fused. The principal advantages of combining sequential processing with square root algorithm are enhancing numerical accuracy and lowering storage requirements, thus removing the limitation of the computing capabilities of the embedded control system on the Kalman filter algorithm. On the basis of the sequential square root algorithm, the paper further propose formulas for the parallel fusion of data and adaptive filtering, so that the phenomenon of covariance matrix being unable to be inversed is avoided and real-time wheel slip rate can be obtained without the statistical law of the prior error. Both the simulation and the experimental results indicate that those presented in this paper are efficient.