A Speed and Acceleration Estimation Algorithm for Powertrain Control

This paper describes a speed and acceleration estimation algorithm based on the Kalman filtering technique. This real-time algorithm, which requires only speed measurement, is computationally simple and can be used in a variety of applications, such as powertrain controls. The dynamic equations of the filter, which has three states, are derived assuming random noise in the process and measurement equations. The filter design is accomplished first analytically and then numerically using a commercial control design package. The filter is implemented in a research vehicle to estimate transmission output speed and accleration, which are then used as feedback signals in powertrain control. In this implementation, the parking pawl (12 teeth) on the transmission output shaft is used as the pulse generator for the speed measurement. Simulation and vehicle road test results are presented to show that this filter is very effective in removing the noise from the estimated speed and acceleration signals. Issues relating to the trade-off between filter convergence and noise rejection are also discussed.