| With the continuous progress of science and technology, robotic technology has entered our dayly lives. As the most important aspect of robotic technology, the rehabilitation robot has received much more attentions and achieved greater progress. This paper is focus on the perception system.This paper designs and implements a perception system of the rehabilitation robot from three aspects, the rehabilitation theory, the hardware and software, and the experiment. This perception system is consist of encoder, gyroscope, accelerometer, magnetometer, pressure sensor and SEMG(surface electromyograpy) sensor and provide accurate signal to the control system. The main work of this article can be summarized as follows:(1) The study of the symptom of stroke patients and the rehabilitation training method if elaborated, only in this way, can we design the most efficacious structure and place the sensors in the right place to get the imformation.(2) Put forward the design requirement of lower limb rehabilitation robot, the design of mechanical structure、control system、perception system and power system are introduced in detail.(3) Design and implement the software and hardware of the perception system. This system is based on CAN communication, each of the sensor is a CAN node. Use different processing and filtering methods in different signals. Especially, use multisensorys’ information fusion to get the posture of the robot.(4) Research an active training based on SEMG, the SEMG is captured by surface electrode, then amplified and filtered to judge the patients’ walking intention. The SEMG also can be the feedback of the control system. Put forward an effective active training method.The innovation of this thesis is designing every sensor to be a CAN node. It’s very easy to add or delete a node and has no effect to the whole perception system. Bring SEMG in the perception system. It can give us a clearer view to the rehabilitation efficacy. |
PDF Full Text Request