2014 年 80 巻 813 号 p. DR0130
One of the key factors affecting vehicle motion control is the nonlinearity of tire forces including coupling between longitudinal and lateral tire forces and saturation of the frictional force. In addition, uncertainties such as external disturbance and tire force modeling error affect state estimation accuracy. In this paper, we incorporate a nonlinear tire model in our previous method based on sliding mode control with repetitive numerical computation, in order to enhance the tracking performance in the nonlinear region of tire forces. To reduce the noise of velocity measurement of contact-type odometers, we apply an extended Luenberger type VSS observer which can cope with both nonlinearity and disturbance of the vehicle dynamics. The performance and computational effort of the proposed method are verified through numerical simulations and experiments using a small vehicle with an on-board embedded CPU.
