Recognition Algorithm of Pedestrian Motion Patterns Based on Low-cost MEMS Inertial Sensors
A motion pattern recognition algorithm based on the low-cost foot-mounted MEMS inertial sensors is pro-posed for the common motion pattern recognition problems of pedestrians such as walking,going upstairs,and go-ing downstairs.Firstly,according to the installation relationship of the MEMS inertial sensors and the actual data collected,the output features of the accelerometer and gyroscope under different motion pattern are analyzed.Then,the output data of the accelerometer in the x-axis direction and the gyroscope in the y-axis direction are se-lected,and the three common motion patterns of pedestrians such as walking,going upstairs,and going downstairs are identified based on the logical sequence of wave peaks and valleys in the output.Finally,ten sets of data col-lected by four experimenters are analyzed.The results show that the average success rate of walking status recogni-tion can reach 100%,the average success rate of going upstairs recognition reaches more than 97.25%,and the average success rate of going downstairs recognition reaches more than 98.99%,which verifies the effectiveness of the proposed method.
万方数据
维普
