Robust H∞ output-feedback control for lane keeping of autonomous ground vehicles considering the system uncertainties

This paper proposed a robust H-infinity static output-feedback controller for the lane keeping of autonomous ground vehicles (AGVs) while the environment/vehicle parameters are uncertain. The variations in vehicle mass, tire cornering stiffness, and steering ratio, which owing to model computational simplification, are estimated firstly and then handled as parameter uncertainties. Given the expensive cost of accessible equipment for vehicle lateral data measurement, a robust H-infinity static output-feedback controller which does not require accurate measurement mentioned above, was proposed in this study. As time-varying nonlinear terms, such as longitudinal velocity, can neither be transformed into a linear one, nor be eliminated by the alter of variables which is normally applied in regular state-feedback controllers, a set of Takagi-Sugeno (TS) based Bilinear matrix inequality (BMI) are solved by Yalmip to achieve the optimum gains. Simulation based on a Prescan-Simulink united platform and a Hardware in the loop (HIL) test have verified the validity of the proposed control approach.