In this paper, an algorithm for direct speed and flux adaptive control of induction motors using unknown time-varying rotor resistance and load torque is described and validated with experimental results. This method is based on the variable structure theories and is potentially useful for adjusting online the induction motor controller unknown parameters (load torque and rotor resistance). The presented nonlinear compensator provides voltage inputs on the basis of rotor speed and stator current measurements, and generates estimates for both the unknown parameters and the nonmeasurable state variables (rotor flux and derivatives of the stator current and voltage) that converge to the corresponding true values. Experiments show that the proposed method achieved very good tracking performance within a wide range of the operation of the induction motor with online variation of the rotor resistance: up to (87%). This high tracking performance of the rotor resistance variation demonstrates that the proposed adaptive control is beneficial for motor efficiency. The proposed algorithm also presented high decoupling performance and very interesting robustness properties with respect to the variation of the stator resistance (up to 100%), measurement noise, modeling errors, discretization effects, and parameter uncertainties (e.g., inaccuracies on motor inductance values). The other interesting feature of the proposed method is that it is simple and easily implementable in real time. Comparative results have shown that the proposed adaptive control decouples speed and flux tracking while standard field-oriented control does not.