Human-robot-interaction Technology And Application Of Rehabilitation Exoskeleton Based On Biological And Physical Perception Feedback

There are a large number of special populations with limb movement disorders in China.Among them,the lower limbs of paraplegic patients lose movement ability and proprioceptive awareness.The aging of the Chinese population will further increase the number of people with lower limb movement disorders.The rehabilitation and nursing for these special group of people will become increasingly prominent.The lower limb exoskeleton robot has been developed into an effective auxiliary rehabilitation equipment that can contribute paraplegics to independently living and working.As a human-machine synergetic intelligent system,the lower limb rehabilitation exoskeleton system not only needs to perceive the wearer’s movement intention to obtain control commands,but also needs to transmit the necessary physical-state of the machine feedback to the wearer to compensate for the wearer’s lack of proprioception in the lower limbs.Traditional exoskeleton requires the wearer to control exoskeleton by operating physical buttons to drive the limbs to complete rehabilitation or auxiliary movement tasks.The traditional human-robot interaction method does not make full of the human body’s active neural signals.In order to improve the neurorehabilitation effects of the exoskeleton system,a human-robot interaction control method based on the identification of active motion intention with human physiological signals,including sEMG and EEG,is becoming the mainstream.However,the accuracy and stability of motion intention recognition is still a difficult problem of research.On the other hand,most existing lower limb extremity exoskeleton systems only have feedforward control from the wearer to the exoskeleton robot,and lack the necessary machine physical state feedback channel from the exoskeleton to the wearer.In this paper,from the perspective of human-robot interaction,the recognition of motion intention with physiological signals is firstly accomplished.Then,the physio-logical state of the human body is perceived and quantified with sEMG and the status of exoskeleton is classified with physical motor angles.Finally,a bidirection human-robot interaction paradigm for the lower limb exoskeleton based on physiological signals and electrotactile feedback is constructedFirstly,aiming at the problem of the low accuracy of the current motion intention recognition with EEG signals,human body’s natural,intuitive and concealed interactive mode-tactile perception is utilized to enhance the EEG characteristics of the user’s motor imagery and improve the practical effect of motion intention recognition with EEG.A novel motor imagery paradigm for brain-machine interface based on the combination of visual and haptic induction is designed and implemented.The implemented motor-imagery brain-machine interface is successfully applied to walking task of lower limb rehabilitation exo skeleton robotSecondly,aiming at the human-robot collaboration task of the lower limb extremity rehabilitation exoskeleton system,a quantification method muscle dynamic contraction fatigue based on sEMG is proposed.The proposed method enables more intuitive and quantifiable expression of muscle fatigue.The effectiveness of the method is verified by fatigue experiment.The quantification of muscle fatigue in exoskeleton experiment would contribute to developing a more ergonomically designed rehabilitation exoskele-ton systemLastly,aiming at the problem that the existing lower limb exoskeleton system lacks necessary feedback to human body,a human-exoskeleton bidirection interaction framework based on sEMG interface and electrotactile feedback is proposed.The exoskeleton with the proposed framework can not only conveys the motion intention of a paraplegic patient to a exoskeleton,but also utilizes the electrotactile to feed back the muscle fatigue information and physical status of exoskeleton to the sensory skin of paraplegic patients,thereby realizing a bidirection human-robot interaction paradigmThe human-exoskeleton bidirection interaction paradigm based on physiological signals and electrotactile feedback establishes a“human-in-loop”states perception and stimulus feedback loop,which improves the effectiveness of online motion intention recognition and enhances the human-machine coupling of rehabilitation exoskeleton As a result,it provides a new prospective for the human-robot interaction research in the field of rehabilitation exo skeleton robots.