Automatic steering control has a direct impact on satisfaction of various vehicle performance requirements including obstacle avoidance, lane tracking accuracy, and passenger comfort. The presence of unknown vehicle dynamics in steering applications requires a robust control design potentially reducing the unexpected effects of uncertain driving conditions and minimizing vehicle path deviations in case particularly strong lateral forces are experienced. Among the issues which remain to be addressed are possible design conflicts between steering angle limitations and vehicle stability. We present an adaptive steering controller for uncertain vehicle dynamics subject to input constraints without compromising system safety under changing road conditions and crosswind effects. The proposed yaw rate control strategy is shown to yield improvements in the steering performance despite the uncertain vehicle mass and velocity in addition to unknown road adhesion factor. Several maneuvering scenarios are simulated to verify the effectiveness of our control design approach.