摘要：AbstractThis paper presents an integrated estimation scheme for the effective engine torque in automotive systems. This leads to a cascaded estimation structure, composed of an adaptive parameter estimator for the augmented wheel dynamics and the longitudinal motion, and an unknown input observer for the engine crankshaft dynamics. The adaptive parameter estimator has the ability to track time-varying parameters and can therefore provide an estimate of the driving torque for the wheels. Then this estimated torque is transmitted to the engine as the load torque through the drivetrain, and is used to design the unknown input observer. The standard models of driveline and tyre friction are modified for ease of parameter estimation. Only the engine crankshaft velocity, the wheel rotational velocity, and the vehicle longitudinal speed are needed. The convergence of these estimators is analyzed. Simulations based on a dynamic simulator built with commercial vehicular simulation software, IPG CarMaker, and Matlab/Simulink show promising results.
