The electromyography signal is a weak signal generated by human muscle groups during movement,it contains control information sources related to movement.An intelligent car system based on electromyographic signals is proposed in this article.The system consists of four modules i.e.,muscle electrical signal acquisition module,wireless transmission module,car control module,and display module.The entire system is divided into two parts of master and slave.The STM32F103ZET6 microprocessor is used as host to collect the multi-channel electromyography signals,and extract the feature values of the collected signals.Then,we divide the feature values into test sets and training sets,and label different gesture signals accordingly.The KNN algorithm is used to analyze the accuracy of the test set for different gesture recognition.The recognition results are sent to the slave car through the wireless transmission module,the car takes corresponding actions after receiving the signal sent by the host.The test results show that the average accuracy of the proposed method at different times can reach over 91.14%,and thereby the system has good robustness.
关键词
STM32F103ZET6微处理器/肌电信号采集/K最近邻算法/手势识别
Key words
STM32F103ZET6 microprocessor/EMG signal acquisition/KNN algorithm/gesture recognition
维普
万方数据