A novel fusion algorithm for estimation of the side-slip angle and the roll angle of a vehicle with optimized key parameters

Abstract

As is known, estimation of the dynamic states of a vehicle plays an important role in safety control, but it is difficult to obtain the vehicle states accurately without expensive measurement instruments because of the non-linear and huge hysteretic characteristics. Nowadays, many methods have been adopted to solve this problem, the results of which are not ideal because it is assumed that the key parameters are constant, in particular in severe manoeuvres. This paper develops a novel estimation method for the vehicle states using the extended Kalman filter with a fusion algorithm. First, the optimal key parameters (the equivalent roll stiffness and the equivalent roll damping) are identified by genetic algorithms using the data from the relationship between the key parameters and the vehicle real-time states. Then, a novel non-linear observer for the side-slip angle and the roll angle is established on the basis of a four-degree-of-freedom vehicle model by utilizing the identified key parameters and the sensors mounted on normal vehicles. The performance of the observer is investigated using both simulations and real-vehicle experiments. The results demonstrate the reasonable accuracy of the estimation method proposed in this paper.