An adaptive haptic interaction architecture for knee rehabilitation robot

In rehabilitation system, the patient has an evolving feature on motor ability. This feature imposes an important constraint on designing of effective human-robot interaction architecture, which not only maintains training effectiveness but also ensures patient safety. In this paper, adaptive haptic interaction architecture is proposed, which utilize passive control for passive phase training and active control for active phase training. Various active control modes, such as active static, active iso-kinetic modes, are implemented using admittance control method and a big force bandwidth can be achieved. These active modes can be realized to strengthen patient's motor capability during active phase training. A prototype is developed and human subject experiments are carried out. The results show that the proposed architecture is able to provide a closed loop rehabilitation solution which includes diagnosis, treatment and feedback.