EKF-Based Parameter Estimation for a Scaled Electrical Vehicle

João Victor A. P. Bezerra; Niederauer Mastelari; Rafael A. Cordeiro; Mauro F. Koyama; Ely C. de Paiva; Felipe W. Varga

doi:10.20906/SBAI-SBSE-2023/3887

João Victor A. P. Bezerra Faculty of Mechanical Engineering, University of Campinas (FEM/UNICAMP), Campinas, SP
Niederauer Mastelari Faculty of Mechanical Engineering, University of Campinas (FEM/UNICAMP), Campinas, SP
Rafael A. Cordeiro Department of Electrical Engineering, Federal University of Espírito Santo, Vitória, ES
Mauro F. Koyama Faculty of Mechanical Engineering, University of Campinas (FEM/UNICAMP), Campinas, SP
Ely C. de Paiva Faculty of Mechanical Engineering, University of Campinas (FEM/UNICAMP), Campinas, SP
Felipe W. Varga Faculty of Mechanical Engineering, University of Campinas (FEM/UNICAMP), Campinas, SP

DOI: https://doi.org/10.20906/SBAI-SBSE-2023/3887

Keywords: Nonlinear Kalman filters, parameters estimation, cornering stiffness, sideslip angle, tire-ground forces, vehicle dynamics

Abstract

Car-like robots play an important role on the research investigations and design of autonomous driving. This is the case of this work, which considers a four-wheeled 1:5-scaled electrical vehicle, with rear differential distribution.We present here the estimation of the vehicle parameters, namely the car sideslip angle and the cornering stiffness of the tires, using a Kalman filter based on a 2D bicycle nonlinear model of the vehicle.The vehicle parameters help to validate the mathematical model used in the vehicle simulator. Experimental results proves the efficiency of the estimation approach.