Study on state parameters estimation for commercial vehicle

Abstract

Vehicle mass and road gradient are the important parameters for engine torque control, transmission shift scheduling and vehicle longitudinal control. It will add manufacturing cost to use more sensors to obtain these values. Therefore, there is increasing concern on the estimation methods of vehicle mass and road gradient based on the vehicle model. In this paper, on the premise of no additional sensors, the engine torque, engine speed, velocity, acceleration/brake/clutch pedal signals and gear from the CAN bus are used as the original data. The estimation methods of vehicle mass and road gradient are studied by applying vehicle dynamic, Luenberger state observer and Recursive Least Square with varying forgetting factors. Furthermore, the real time estimation arithmetic is validated through dSPACE/MicroAutoBox system on FAW J5 commercial vehicle. © 2013 Springer-Verlag.