Estimation of Walking Speed Using Accelerometer and Artificial Neural Networks

摘要: This paper proposes a practical method for estimating human walking speed using accelerometer data. A portable device based on was developed to objectively record signals. An Artificial Neural Network estimate the average of walking. We extracted six parameters, namely step number, subject’s height, root mean square and difference between maximum minimum vertical frontal signals as inputs artificial neural network. To validate performance proposed method, we tested data collected from 35 subjects under free-living conditions. The experiments shows that accuracy is 96.96% which better than previous works. Based results these experiments, concluded tri-accelerometer promising tool accurate assessment speed.