This paper describes a novel learning/adaptive state trajectory control method and its application to electronic hydraulic pressure control. The control algorithm presented herein learns the inverse input-state mapping of the plant at the same time this map is employed in the feedforward loop to force the state of the plant to asymptotically converge to a prescribed state trajectory. The algorithm accomplishes this task without requiring prior exact information about the state transition map of the plant. The novel controller is applied to an electrohydraulic poppet valve with the objective of tracking a desired supply pressure signal. In this application, the controller learns the inverse conductance characteristics of the valve. The supply pressure tracking performance subject to the proposed controller is validated through experimental data.