Real-time automotive slip angle estimation with extended H∞ circle criterion observer for nonlinear output system

Many active vehicle safety systems such as electronic stability control (ESC), rollover prevention, and lane departure avoidance could benefit from knowledge of the vehicle slip angle. However, it is a challenge to design an observer to estimate slip angle reliably under a wide range of vehicle maneuvers and operating conditions. This is due especially to nonlinear tire characteristics and system models which have nonlinear output equations. Hence this paper develops an extended H∞ circle criterion observer for state estimation in systems with nonlinear output equations. The observer design approach utilizes a modified Young's relation to include additional degrees of freedom in the linear matrix inequality (LMI) used for observer gain design. This enhanced LMI is less conservative than others proposed in the literature for Lipschitz nonlinear systems, both with and without nonlinear output equations. The observer is applied to slip angle estimation and utilizes inexpensive sensors available in all modern vehicles. Finally, experimental tests on a Volvo XC90 sport utility vehicle are used to evaluate the developed approach. The experimental results show that the slip angle estimates for a variety of test maneuvers on road surfaces with different friction coefficients are reliable.