Gait Estimating Based on Accelerometer of the SmartPhone

摘要

With the development of the mobile communication technology, the network technology and the surveying and mapping technology, the location service related to smartphone has become an important part of the modern geographic information industry. Because of the low price, large noise and unstable signal of the smartphone, so it needs for further studying on how to eliminate the smartphone sensors signal instability components, providing users with better location services. The article takes MeiLan NOTE3 smartphone as the research object. Based on the research, it has been proved that accelerometer signals carried by the foot have a strong correlation with the movement state of the pedestrian centroid and it has been analyzed that the phone has better consistency when carried by foot than that when placed in the crotch. The signal after Kalman filter can well estimate the step frequency of pedestrian movement. Combined with the height, weight and pace of pedestrians, a new step size estimation model is proposed. The measured signals of different pedestrians walking at slow speed, normal speed and fast speed are used to verify the reliability of the method. The estimated distance is within about 50m and the estimated error is within 0.25m. The estimation accuracy of the new model is better than the existing ones.