In order to provide patients with a stable, comfortable and active compliant rehabilitation training environment, an active compliant interactive control scheme for a lower limb rehabilitation robot based on adaptive adjustment of impedance parameters is proposed. It consists of two parts:the outer loop for impedance parameter adjustment and the inner loop for trajectory tracking. Firstly, a fuzzy adaptive impedance parameter adjuster is proposed to solve the problem caused by the dynamic change of impedance parameters of human body during rehabilitation training. It takes the human-robot interaction force, the position error and the speed error as inputs, and uses the fuzzy inference to adjust the damping and stiffness coefficients in real time to realize the self-adaptation to human body impedance. Secondly, an indirect adaptive fuzzy controller is designed, in which the fuzzy system is reasonably constructed to approximate the unknown nonlinear system, and the set trajectory reflecting the patient's motion intention is tracked stably. The system stability is proved by Lyapunov method. Finally, the simulation results show that the maximum deviations of the hip and knee trajectories under the proposed method are reduced by 53.43% and 66.87% respectively in comparison with the general method of fixed expected impedance parameters, which verifies the feasibility and effectiveness of the proposed method.