To ensure the effectiveness and safety of active rehabilitation training for patients with lower limb disabilities, this study designs a torque-less sensor admittance control method based on an asymmetric integral barrier Lyapunov function for lower limb rehabilitation exoskeleton robots. Firstly, a sliding mode observer is designed based on the generalized momentum method to estimate the human-robot interaction torque in real-time, thereby reducing the reliance on torque sensors by the lower limb exoskeleton. Then, an admittance control framework is constructed for the lower limb exoskeleton, utilizing a saturation function to constrain the reference trajectory generated by the outer loop admittance model within a predetermined safe range, ensuring the compliance and safety of the exoskeleton. Additionally, for the inner loop gait trajectory tracking, an inverse sliding mode control algorithm based on an asymmetric integral Lyapunov function is designed to guarantee the system's output constraint performance and gait trajectory tracking accuracy. Finally, the stability of the control system is proven based on the Lyapunov stability criterion, and the effectiveness and superiority of the proposed control algorithm are verified on a lower limb rehabilitation robot experimental platform.