| Popis: |
The upper limb impedance parameters of stroke patients are changing with the rehabilitation process. If the patient's upper limb impedance parameters can be identified, the impedance control parameters of the rehabilitation robot can be adjusted in real time according to the patient's limb mechanical impedance parameters, which will effectively improve the safety of human-machine interaction and enhance the effect of rehabilitation training. This paper introduces the basic principle of the impedance parameter identification method based on the mechanical disturbance method. Based on the principle of the mechanical disturbance method, a three-layer neural network is used to simulate the mechanical impedance parameter identification. The simulation results show that the model calculated in this article is the result of the training. The coefficient of determination of is 0.94428, and the use of neural network can identify mechanical impedance well, which lays the foundation for follow-up research. |
