Osamu Nishihara

A New Control Strategy to Reduce Steering Torque Without Perceptible Vibration for Vehicles Equipped With Electric Power Steering

This paper proposes a new electric, power steering control strategy, which significantly reduces the effort needed to change the steering direction of stationary vehicles. Previous attempts to reduce undesirable steering vibration have failed to reduce the steering torque because high-assist gains tend to produce oscillation or increase noise sensitivity Herein, to eliminate this vibration, a new control strategy was developed based on pinion angular velocity control using a newly developed observer based on a simplified steering model. Tests yielded excellent estimations of the pinion angular velocity, and this made it possible to eliminate vibration at all steering wheel rotation speeds. Experiments with a test vehicle confirmed significant steering torque reduction, over a wide range of steering wheel speeds, without vibration transmission to the driver The proposed control strategy allowed use of an assist gain more than three times higher than is conventional. Additionally, the proposed control strategy does not require supplemental sensors. [DOI: 10.1115/1.4001838]

Estimation of Road Friction Coefficient Based on the Brush Model

Road friction coefficients are highly effective for advanced vehicle control technologies, although the estimation at four individual tires has not been practically used for ordinary vehicles. This study describes the essential relation between the tire forces and the aligning torque that can be rearranged as an estimation equation for the grip margin. The grip margin is readily convertible into the friction coefficient. The brush model is reanalyzed, beginning from the conventional simple physical model, and intrinsic expressions are derived. The grip margin, which is defined as the residual tire force normalized by the radius of friction circle, was estimated using three components of the tire forces and the aligning torque. A simple cubic equation is obtained as a grip margin equation for an isotropic brush model. Previous studies assumed an anisotropic brush model and obtained an imperfect quintic equation. In the present study, a new term is added to the algebraic equation, which was shown to be consistent with the isotropic model. The solutions to the equations are approximated by Chebyshev polynomials. The estimation methods are tested by numerical simulations using CarSim, which is a popular vehicle simulation software application. The estimated friction coefficients agree well with the values that are set during each run of the simulations, especially for the cases of smaller grip margins and lower friction conditions.