Enhanced Human-machine Interface Using Electric Stimulation Feedback

Nowadays electric stimulation is widely used in medical rehabilitation.Electric stimulation with high current can directly activate muscles and helps therapists to rehabilitate physically disabled patients.And electric stimulation with low current can activate tactile receptors and bring users rich artificial sensations,which is able to improve performance of human machine interface(HMI).As a typical HMI in medical rehabilitation,available commercial intelligent prostheses generally lack tactile feedback and cannot be widely recognized for high mind demands on users and little communication with users during operation.To solve the problems,this paper introduces electrotactile to enhance bionic HMI.First of all,this paper analyses the features and drawbacks of available electric stimulators,and then raises the requirements of portable electric stimulators according to intelligent prosthesis demands.Based on these,this paper has finished the software and hardware design of light-weight multichannel stimulator——SJTU-Stimulator,and applied it to four persons verifying its ability of providing abundant electrotactile.Overall,the stimulator could reproduce pressure sensation,vibration sensation,slip sensation and so on.Electric stimulation will disturb surface electromyography(sEMG)and harm the control performance of intelligent prostheses.This paper has done theoretical analysis and detection of electrotactile artifacts,and proposed a kind of online cascading high-order IIR notch filter based on simplified second-order notch filter.This filter tracks the targeted frequency of artifacts utilizing the least square method,achieving great improvement in EMG signal-to-noise ratio.Highly recovered control signal ensures that electrotactile improves the performance of bidirectional HMI while not introducing negative effects.The contact information between prostheses and objects could improve grasp performance.This paper embeds flexible force sensors in finger tips of the prosthesis,takes advantage of electrotactile to provide real-time feedback according to contact condition between prostheses and objects,and finally closes the control loop.Actual grasp experiment has been executed on this platform to verify the reliability,rapidity and stability of this close-loop control method.The result shows that with the electrotactile feedback strategy including slip,pressure and vibration sensations,the intelligent prosthesis could make grasp manipulation more stably and reliably,and operate faster when faced with hard grasp tasks.Not adding more cognitive load,electrotactile is a reliable way to enhance bidirectional HMI.Finally,we cannot ignore the urgent needs of physically disabled patients to use intelligent prostheses,instead of their atrophic muscles.This paper proposes a way to activate the specific muscles based on high-density electric stimulation arrays.The experiment shows that able-bodied people could passively achieve many intricate hand gestures and stroke patients could also passively achieve several finger movement actions due to this method.It could be seen that the long-term experiment would largely improve the activity of their specific muscles and help them possibly control the intelligent prosthesis well.