Sliding Mode Idle Speed Ignition Control Strategies for Automotive Engines

In recent years it has become apparent that control system design based on conventional linear transfer function or state-variable feedback suffers from several serious disadvantages including: i) It requires an accurate model, which cannot be attained by current identification procedures ii) The control is not robust to unstructured uncertainties such as additive or multiplicative parameter variations However, an alternative and potentially improved class of control systems involves sliding mode control, in which the actuating signal is changed in a jump-wise manner, depending on the position of the state-trajectory in state space. Under certain conditions the state trajectory exhibits a stable high frequency sliding mode in which the state trajectory tracks a defined hyperplane towards the desired steady-state. Major advantages of the method over conventional linear controls are that i) The trajectory may be constrained to be within certain regions of the state space, thus preventing expensive or dangerous transients that commonly arise in high-gain linear systems ii) Actuation involves only a simple form of switching (on-off) control that is easily implemented using (say) a relay or stepper motor with locally-mounted integrated circuits iii) Unlike many traditional controller designs, it takes into account the fact that in practice, control action is always finite iv) The control is robust in the sense that the sliding mode is relatively insensitive t o model inaccuracy or external disturbances [1]; thus it is applicable to cases where the system model is poorly defined or time-varying. This affords the opportunity of simplifying the complex and uncertain time varying processes that occur within automotive engines and to develop new methods of controlling their performance.