Electronic throttle control is an integral part of an engine electronic control unit (ECU) that directly affects vehicle fuel economy, drivability, and engine-out emissions by managing engine torque and air-fuel ratio through adjusting intake charge flow to the engine. The highly nonlinear dynamics of the throttle body call for nonlinear control techniques that can be implemented in real-time and are also robust to controller implementation imprecision. Discrete sliding mode control (DSMC) is a computationally efficient controller design technique which can handle systems with high degree of nonlinearity. In this paper, a generic robust discrete sliding mode controller design is proposed and experimentally verified for the throttle position tracking problem. In addition, a novel method is used to predict and incorporate the sampling and quantization imprecisions into the DSMC structure.
First, a nonlinear physical model for an electromechanical throttle body is derived. Parameters of the model are determined using techniques of model/parameter identification. Next, a DSMC is formulated for controlling the throttle position. The performance of the DSMC is examined under different sampling and quantization levels via the analog-to-digital converter (ADC). The experimental results show that the controller tracking performance is significantly affected by the ADC imprecisions. To this end, ADC effects are modeled and the DSMC control law is reformulated to consider and overcome the uncertainty due to ADC imprecisions. Real-time experimental validation results show that the proposed robust DSMC improves the throttle position tracking performance, under ADC imprecisions, by up to 70% compared to a conventional controller.
