采摘机器人SEMG手势识别研究——基于RNN循环神经网络

为了实现采摘机器人通过手势动作进行远程控制的目标,采用MYO手环采集人体手势动作信号,将信号进行滤波、放大和A/D转换等预处理后,通过无线通信模块发送给PC机;PC机提取右移、左移和采摘等动作的特征值,送入RNN网络中进行训练和识别,并将识别结果以指令的方式发送给采摘机器人,控制采摘机械手进行相应操作。实验结果表明:采摘机器人SEMG手势识别算法识别率较高,结果非常理想,验证了采摘机器人通过手势进行远程控制的可行性。

SEMG Gesture Recognition of Picking Robot Based on RNN Cyclic Neural Network

In order to achieve the goal of remote control of picking robot through gesture action,it uses MYO hand ring to collect gesture action signal of human body,filter,amplify and A/D conversion to preprocess the signal,send it to PC through wireless communication module. PC extracts the characteristic values of right,left and picking actions,sends them to RNN network for training and recognition,and recognizes them. The results were sent to the robot by command,and the robot was controlled to operate. The experimental results show that the SEMG gesture recognition algorithm has a high recognition rate,and the result is very ideal,which verifies the feasibility of remote control of the picking robot through gesture.