Steering of Autonomous Vehicles Based on Friction-Adaptive Nonlinear Model-Predictive Control

The vehicle steering-control behavior is highly dependent on the road surface. However, the road surface conditions are typically unknown a priori, and control actions that are safe to perform on asphalt may therefore lead to vehicle instability on low-friction surfaces. It is therefore important that the road surface is estimated, or at least detected, online, and that the vehicle dynamics control algorithms are adapted to the changing conditions. In this paper, we propose a nonlinear model-predictive control (NMPC) scheme that adapts its tire parameters in response to the estimated road surface. We show how estimating the initial slope of the tire-force curve can be used to change the full nonlinear tire-curve used by the NMPC and validate the method in simulation.