Research on Indoor Constraint Location Method of Mobile Phone Aided by Magnetic Features

• Main Authors: Jiawei, Chen, Wenchao, Zhang, Dongyan, Wei, Xiaofeng, Sun
• Format: Conference Proceeding
• Language: English
• Subjects: complementary filtering; DTW; geomagnetic matching; pedestrian dead reckoning; QSMF
• ISSN: 2471-917X
• DOI: 10.1109/IPIN54987.2022.9918149

As an important passive location method, pedestrian positioning technology plays a large role in indoor scenes and other scenes that cannot be covered by GNSS signals. To enhance the positioning accuracy of mobile phones, geomagnetic matching and other methods are often used to correct the resulting position. However, some geomagnetic matching methods, such as ICCP, are not accurate enough and require a large amount of calculation. In the meantime, they also lack the heading constraint. When the indoor magnetic interference is severe, it will have a great impact on calculating the magnetic heading and lead to the divergence of the final position. For the above problems, this paper proposes a heading and position constrained algorithm based on indoor geomagnetic characteristics. On the one hand, the QSMF (Quasi Static Magnetic Field) algorithm is used to divide the indoor magnetic quasi-static and non-quasi-static range. In the magnetic instability interval, the non-magnetic heading calculated by the combination of accel and gyro is adopted, and the combination of non-magnetic heading and magnetic heading is adopted in the magnetic stability area, so as to keep it able to continuously provide a stable heading in the case of indoor magnetic interference. On the other hand, in order to reduce the impact of errors caused by low-cost sensor recording onDTW, the inertial recursive information is integrated into geomagnetic DTW, and the combination of inertial information and geomagnetic information is used in calculating the grid module value, so as to improve the final matching success rate and enhance the reliability of final position correction. Experiments show that this method can provide a stable heading in the room with serious magnetic interference, and its heading error accuracy is improved by 70% and position error accuracy is improved by 50%.