Independent Wheel Effects in Real Time Estimation of Tire-Road Friction Coefficient from Steering Torque

Improvement of control technologies and increased public acceptance of active systems on vehicles has led to greater implementation of safety critical assistance systems in production vehicles. However, the estimate of tire friction coefficient utilized within these systems is often obtained using only data from inertial sensors in which the tire friction is weakly represented. In contrast, skilled human drivers utilize steering torque to obtain a sense of the surface friction. In theory, this approach can also be utilized with electric power steering or steer-by-wire technologies, but there are significant challenges in obtaining and interpreting data from the steering system including the signal characteristics of steering torque measurements and the well-known problem of practical measurement of tire slip angles. More importantly, a problem often lost in modeling assumptions is in left/right asymmetry of the steering torques. The work presented in this paper illustrates the weak signal-to-noise ratio for steering torque in carefully controlled experiments and demonstrates that the lumped-axle assumption may lead to poor estimates of the front axle peak force value when attempting to predict it at slip angles less than those required for full saturation of both wheels.