This paper proposes an adaptive nonlinear sliding mode control for feed drive systems. The aim is to increase precision by minimizing tracking errors of each feed drive axis without additional energy consumption. The proposed adaptive sliding mode control law is used to adjust controller gains based on the tracking error at each sampling time. Therefore, an appropriate control input can be generated on the system state such as the beginning of the transient state or around the steady state. To verify the effectiveness of the proposed controller, computer simulations and experiments were conducted for a typical lead screw feed drive system using a typical circular reference trajectory. The tracking error results of the proposed controller are compared with those of a nonlinear sliding mode controller. The proposed controller has achieved a reduction of 25% of the tracking error with smaller control input variance.