The team of Stanford implemented the velocity and steering controllers separately for its vehicle “Stanley” at the DARPA Grand Challenge. The velocity control is implemented as a PI controller, while the steering angle is computed by a nonlinear feedback function of the cross-track error, which is known as the “Stanley method.” The performance of this controller degrades seriously at higher speeds.