Indoor Geomagnetic Positioning Based on a Joint Algorithm of Particle Filter and Dynamic Time Warp

摘要

The indoor geomagnetic field is affected by the steel structure and other ferromagnetic materials, resulting in the local anomaly of the magnetic field and the indoor geomagnetic field will be specificity. Benefit from this phenomenon, indoor magnetic positioning technology to be achieved. However, in large buildings, the specificity of geomagnetic field will be weakened, which leads to ambiguous positioning results. In response to this phenomenon, this paper presents a path matching based indoor geomagnetic positioning system, by increasing the number of matching features to solve this problem. The system combines Dynamic Time Warp (DTW) algorithm and Particle Filter (PF) algorithm to track the target in the path matching mode. Finally, Measuring robot loaded with magnetic sensor used to verify the results. The results show that the path matching has sufficient number of geomagnetic features to resolve the ambiguity of positioning results under condition of reduced specificity, and the positioning accuracy is better than 1 meter.