Pedestrian Positioning Method with Phone Attitude Estimation

With the rapid development of Internet of Things technology, pedestrian navigation has attracted extensive attentions in recent years. Pedestrian dead reckoning (PDR) is a widely used algorithm for indoor positioning, which usually requires that the moving directions of the mobile phone and the person are aligned. Such limitations prevent the real application of the PDR. To address this issue, this paper proposes a PDR-based method with phone attitude estimation, which relaxes the requirement of the direction alignment issue. The proposed method includes a gradient descent-unscented Kalman filter algorithm to obtain the attitude of the mobile phone, and then adopts a machine learning method to identify several mobile phone usage behaviors. Based on the behavior pattern recognition results, robust step detection and heading estimation methods are designed. The experimental results show that the average positioning error of the proposed method is about 0.55 meters, which has great potentials in future applications.