Most upper limb rehabilitation patients are still hard to feel the accuracy force they have imposed in the end of arm after a systematic upper limb rehabilitation. In order to provide an accurate end-of-arm force for those disabled people, a force display system based on wearable arm gesture sensors and Electromyographic (EMG) sensors is designed and given in this paper. The wearable arm gesture sensors and EMG sensors are specially placed to detect the arm gesture and the EMG signal of the arm, and a force sensor is used to measure the force and verify the force display effect. In order to control the rehabilitation arm move slowly at a constant speed, the kinematic model of the upper arm is analyzed. A special experiment platform is established so as to get the simultaneous data of end-of-arm force and the arm gesture and EMG signal, then the Generalized Regression Neural Network (GRNN) is brought in to catch the relationship between them. A group of horizontal movement experiment and vertical movement experiment are designed specially and verify the feasibility and effectiveness of the system. The result shows that the information fusion based on GRNN for this system could accurately display the end-of-arm force.
Read full abstract