Synthesis of PDR Algorithm and Experimental Estimation of Accuracy of Step Length Estimation Methods

Today there are a large number of cases in which it is required to position workers indoors with high accuracy. To meet accuracy requirements, combined processing of UWB radio measurements and inertial measurements is often used. Pedestrian positioning allows extracting additional information about various phases of human walking from inertial measurements. The purpose of this work is to experimentally evaluate and compare the positioning errors of various step length estimation methods. Conducted experiment confirmed the workability of algorithm and results were the following: maximal mean positioning error of examined models doesn't exceed 30 cm and minimal mean positioning error is less than 20 cm for total traveled distance equal to 16.2 m. The main contribution of this work can be called: description of proposed PDR algorithm and assessment of the applicability of considered step length estimation methods based on biomechanical models for the problems of high-precision indoor navigation.