摘要
本文提出一种基于Arduino单片机和神经网络的智能拳击手套设计方法。该方法采用Arduino单片机内含的姿态传感器采集三轴加速度数据,首先将采集的标准模型数据输入Curie神经网络芯片进行学习训练,然后将采集的动作数据输入训练好的网络进行判决,最后通过蓝牙端口将判决结果显示在电脑端的Arduino IDE。实验表明,本系统能准确识别直拳、勾拳、摆拳三种基本的拳击动作。
A design of intelligent boxing gloves based on Arduino MCU and neural network is proposed.In this method,the attitude sensor embedded in Arduino MCU is used to collect the triaxial acceleration data.Firstly,the collected standard model data is input into Curie neural network chip for learning and training,then the collected action data is input into the trained network for judgment,and finally the judgment result is displayed in Arduino IDE of the computer end through the blue tooth port.The test results show that the system can accurately identify three basic boxing actions,straight boxing,hook boxing and swing boxing.Compared with the traditional design.
作者
单巍
方振国
王江涛
崔少华
SHANWei;FANG Zhenguo;WANG Jiangtao;CUI Shaohua(School of Physics and Electronic Information,Huaibei Normal University,Huaibei Anhui,235000,China)
出处
《阜阳师范大学学报（自然科学版）》
2020年第3期54-58,共5页
Journal of Fuyang Normal University:Natural Science
基金
安徽省高等学校自然科学研究项目(KJ2018B10)
安徽省大学生创新创业训练计划项目(201910373027)
企业委托课题(RD/PY-01-2019,RD/ZD-01-2019)资助。
