An adaptive inverse controller is developed for feedback linearizable nonlinear systems with nonsmooth actuator nonlinearities. The use of an actuator nonlinearity inverse and a feedback linearizing controller leads to an error equation suitable fur deriving an adaptive update law for the inverse parameters. Closed-loop signal boundedness is proved analytically, and system performance improvement is shown by simulation results. Such adaptive control schemes are also developed for multivariable nonlinear systems with actuator nonlinearities. For nonlinear systems that do not possess a relative degree, dynamic extension is employed to realize adaptive inverse compensation designs fur actuator nonlinearities. These adaptive designs ensure closed-loop stability in the presence of uncertain actuator nonlinearities.