Vehicle Stability Analysis by Zero Dynamics to Improve Control Performance

Abstract

Good stability performance is a prerequisite for the motion of vehicles controlled either by a human driver with advanced driver-assistance systems or by highly automated driving systems. Hence, vehicle motion control has become the core technology for vehicles driving everywhere in all conditions. However, vehicles might start to skid or spin, i.e., lose motion stability, even if some vehicle motion controllers are active. This motivates our work with a twofold contribution. First, we show that there are zero dynamics when input–output linearization is used in controller design. By analyzing the behavior of the zero dynamics, we identify why and under what conditions the vehicle starts to slide sideways and then spins. Second, based on the analysis results, we propose a cooperative control scheme to improve the motion stability of vehicles. A linear combination of the yaw rate and the sideslip angle is used as the control output, and we discuss which linear combinations are appropriate. Finally, simulation results, including a hardware-in-the-loop simulation, and experimental results with a refitted DongFeng E70 vehicle are given to illustrate the effectiveness of our method.