Synthesis of the Step Detection and Step Length Estimation Algorithms Based on IMU Measurements

This paper is devoted to the synthesis of algorithm for pedestrian dead reckoning (PDR) using an inertial measurement unit (IMU). In the introduction, a brief review of current solutions in the field of PDR is presented, and the relevance and feasibility of the research discussed in this paper is shown. The second section (Algorithm Synthesis) gives a detailed description of the algorithm under development, which consists of two stages: step detection and calculation of the walked distance based on biomechanical models (Weinberg, Mikov, Kim, and Scarlett) with adaptive estimation of the scaling factor. In the same section, a metric characterizing the accuracy of the proposed algorithm, calculated as the norm of relative errors of estimated distance based on a set of experiments, is introduced. Then, a detailed description of the proposed algorithm is given. The third section (Experiment) describes an experiment in which users walked along a given trajectories of different lengths. The fourth section of this paper presents the results of the experiment and their brief analysis. In this paper, we proposed an algorithm for step detection, and we combined direct and indirect methods for step length estimation, which helped to improve the accuracy of the PDR algorithm.