In an effort to alleviate the issue of knee joint fatigue and injury during lower limb ambulation, a novel passive assisted exoskeleton robot with human–machine interaction is investigated to assist the movement of the human knee joint. The design of the exoskeleton configuration takes into consideration the physiological structure and gait function of the knee joint, ensuring that it satisfies the requirements for motion, force, and gait function of the knee joint. To explore the interaction between the wearer and the exoskeleton, a human–machine kinematic model after wearing exoskeleton is established, which is instrumental in analyzing the integration motion of the wearer and exoskeleton. In addition, the dynamic and static models of the knee joint after wearing the exoskeleton are established, utilizing the Newton–Euler method and force polygon method, respectively, to evaluate the effectiveness of the exoskeleton. Moreover, the size parameters and spring stiffness of the exoskeleton are optimized, using both human body kinematic model and mechanic model. Furthermore, the effectiveness of the exoskeleton in proving assistance is evaluated through human body simulation, using OpenSim. The results indicate that the exoskeleton significantly reduces the knee joint torque by 48.42%.