A higher order sliding controller scheme for a steering vehicle control is proposed in this paper. Since the sys tem is uncertain, to overcome the constraint on the knowledge of the vehicle model, local models related to some operating points have been used to synthesis a nominal fuzzy type-2 global model.The controller is based on the integral sliding mode concept and it is composed of two parts. The first one leads to achieve finite time stabilization of the sliding manifold and its derivatives, it has two dynamics: a robust feedback and a forcing term. The second part has to reject the system uncertainties and noises.As a result, a higher order sliding mode is established and the time convergence is chosen in advance. Promising results have been obtained using real data acquired by a laboratory vehicle.
I. INTRODUCTIONThe design of robust vehicle controllers remains an inter esting research topic and has received more attention during the last decades. This, to design the driving aid systems and increase the road safety. The development of such systems, requires the knowledge of the models able to represent the behaviors of the vehicle. Unfortunately, a such requirement is ensured partially and the known nonlinear vehicle models are only used for simulations tests but not for control design. To overcome the models knowledge constraints, several solutions are propose: fuzzy systems [26], [28], [6], [20], switched systems [4], [15], [8] and LPV systems [2]. The first problem addressed in this work is related to the design of a global vehicle model which will be used to develop a robust vehicle controller. In fact, several vehicle controllers based on simple vehicle models are proposed [1], [24], [7], [23]. These controllers are developed to accomplish some maneuvers: platooning [24], stop-and-go control [17], [13], steering and braking control [23], coupled longitudinal and lateral vehicle control [21]. Unfortunately, in major part of these works, simple vehicle models are used in the design controllers procedure. Consequently, the performances and the validity domain of such controllers depend strongly of the vehicle models. To overcome the constraint on the knowledge of the vehicle model, some local vehicle linear models have been used to obtain a nominal type-2 fuzzy model.The second problem treated in this work concerns the design of a robust steering vehicle control using the type-2 fuzzy vehicle model. The proposed steering vehicle control uses higher order sliding mode approach [5], [12]. The sliding mode control (SMC) is among a widely used method because of its simplicity and robustness [30]. However, because of the