Adaptive cooperative control of a soft elbow rehabilitation exoskeleton based on improved joint torque estimation

A current challenge with robot-assisted rehabilitation is to combine the active involvement and voluntary participation of patients into the rehabilitation training process to enhance therapy outcome. This paper presents a soft elbow exoskeleton for the rehabilitation training of disabled patients. An adaptive cooperative control strategy is investigated to promote active participation based on an improved joint torque estimation method and a time-delay sliding mode control scheme. The surface electromyography signals from biceps and triceps are input into a Hill-type musculoskeletal model and a Gaussian radial basis functional network to estimate human elbow joint torque and motion intention. The joint torque estimation experiment, intention-based trajectory following experiment and patient-active free-motion training experiment are carried out by five healthy volunteers and five stroke patients to evaluate the performance of proposed control strategy. The results indicate that the proposed scheme can guarantee the joint torque estimation accuracy and position control accuracy. Moreover, it allow patients to actively dominate the training path based on their motion intention with different training intensity.