Stable physical interaction between rehabilitation robots and human users is essential for safe and effective therapy. Traditional position-control approaches often lead to high interaction forces and instability when interacting with uncertain human dynamics. In contrast, the human sensorimotor system maintains stability through impedance modulation, sensory feedback integration, and hierarchical control organization. This paper investigates a sensorimotor-inspired impedance control strategy for rehabilitation robotics with a focus on stability analysis and system performance. A dynamic model of the human–robot interaction was developed, and stability was analyzed using Lyapunov methods and passivity-based control theory. Simulation and experimental results demonstrate that appropriate impedance tuning significantly improves system stability, reduces interaction forces, and enhances trajectory tracking performance. The proposed approach provides a biologically inspired framework for safe and adaptive rehabilitation robot control.