Decision tree assisted EKF for vehicle slip angle estimation using inertial motion sensors

Vehicle side slip angle is a critical variable used in car safety systems like Electronic Stability Control. Due to the practical difficulty in direct measurement of side slip angle, accurate estimation of vehicle side slip angle using available signals is becoming important. This paper presents a novel algorithm for estimating the side slip angle of a vehicle in real time using inertial motion sensors. The algorithm uses a J48 decision tree classifier to assist the Extended Kaiman Filter (EKF) predictions of the vehicle side slip angle. The decision tree classifies the inertial data into classes based on the condition the slip angle is expected to be in. Using the class information asserted by the classifier, the error covariance parameter of the EKF is adjusted to compensate for changes in disturbances and nonlinearities. The results show that the decision tree assisted EKF technique presented in this paper is capable of predicting the slip angle with sound accuracy using inertial motion data.