Research on interactive force acquisition and active-passive mode determination of rehabilitation robot

Lower limb muscle strength is an important indicator of the effectiveness of rehabilitation training for patients with stroke and hemiplegia. When patients are trained with a lower limb rehabilitation robot, real-time human-machine interaction force acquisition can not only grasp the muscle strength of patients' lower limbs in time but also serve as an important reference for active/passive training mode. To accurately collect the human-machine interaction force to identify the patient's movement intention and determine the active/passive mode, a method is proposed to collect the human-machine interaction force based on a resistive strain gauge leg pressure sensor. The method can be applied to a variety of rehabilitation platforms to collect human-machine interaction force quickly and accurately and to identify and classify behaviors through probabilistic neural networks to determine patients' active/passive training patterns. By analyzing the active/passive walking of rehabilitation patients, the motor status of patients in that period is derived. The validity and correctness of the designed sensor and determination method were verified.