Automatic guidance of farm tractors would improve productivity of many
field operations by reducing operator fatigue and improving machine p
erformance. A self-tuning steering controller, which can be used for a
ll noncontact types of positioning systems, was designed for tractor g
uidance. A simple two degrees-of-freedom model was chosen with which t
o develop a prediction model used in recursive least-squares parameter
estimation. A variable forgetting factor was implemented and its algo
rithm modified to cope with time-varying nonlinear systems. The self-t
uning steering controller based upon minimum variance control was test
ed and verified with a tractor dynamic simulator. Circular, lane chang
e, and sinusoidal paths were used. A model tractor was also used to ve
rify controller performance for straight-line and sinusoidal paths.