Indoor positioning method based on LSTM and multi-sensor fusion

Abstract

The invention relates to the technical field of the indoor positioning, and provides an indoor positioning method based on LSTM and multi-sensor fusion. The indoor positioning method comprises the following steps: firstly collecting triaxial acceleration, triaxial magnetic value, triaxial angular speed and location coordinate data of the pedestrian through an accelerometer of a mobile phone, a geomagnetic sensor and a gyroscope, and preprocessing a data set; performing coarse positioning on the pedestrian by utilizing WiFi information, and computing an initial location of the pedestrian; performing step counting on the pedestrian by utilizing the triaxial acceleration data and performing the step length estimation; performing direction estimation on the pedestrian by utilizing the triaxialmagnetic force value data and the triaxial angular speed data so as to form a pedestrian motion model, and then estimating the pedestrian location; and then constructing and training a LSTM model forpedestrian positioning, and predicting the pedestrian location by utilizing the trained LSTM model; and finally performing Kalman filter fusion on the pedestrian location computed in previous two steps, thereby accomplishing the pedestrian positioning. Through the indoor positioning method disclosed by the invention, the indoor positioning precision can be improved, the cost is low, and the deployment is easy.

Description

technical field [0001] The invention relates to the technical field of indoor positioning, in particular to an indoor positioning method based on LSTM and multi-sensor fusion. Background technique [0002] Indoor Positioning System (IPS) refers to a system that uses mobile devices to collect radio waves, magnetic fields, acoustic signals or other sensory information in buildings to locate objects or people. Indoor positioning is a basic service in the mobile Internet era, which can provide consumers with better services and bring economic benefits to enterprises. With the continuous development of indoor positioning technology, indoor positioning technology will play an important role in people's daily life, intelligent scenic spots and other industrial applications, as well as underground public areas such as subways. Although the current indoor positioning technology is flourishing, such as infrared and ultrasonic positioning technology based on unlimited signals, positio...

AI Technical Summary

Problems solved by technology

However, the accuracy of these positioning technologies is about 3 meters. Although they can meet the current positioning needs, most of them need to deploy additional equipment, such as deploying an iBeacon every 6-8m, which is very expensive.

Images