Indoor positioning method of inertial navigation based on bp neural network

Abstract

The invention discloses an inertial navigation indoor positioning method based on BP neural network, which includes a motion data collection and preprocessing stage, an off-line training stage and a real-time positioning stage. Data collection is through the left and right ankle wearable devices to collect the movement data of the personnel's two ankles, the personnel's height and step length; the data preprocessing is to use multi-sensor information fusion processing to obtain 7 neural network input data - left and right legs Attitude angle changes θ1 and θ2, ankle acceleration average, variance variance, left and right foot step duration time1 and time2, and height stature; the offline training phase includes: establishing a neural network step size estimation model; 7 input data and step length data input into the neural network for training; the real-time positioning stage includes: step recognition inside the right ankle wearable device, real-time prediction of step length, and real-time positioning through dead reckoning using step length and direction angle. The invention improves the accuracy and real-time performance of indoor positioning, and enhances the reliability.

Description

technical field [0001] The invention relates to the field of indoor positioning, in particular to a BP neural network-based inertial navigation indoor positioning method. Background technique [0002] Existing indoor positioning technology solutions mainly include Wi-Fi positioning technology, infrared positioning technology, ultrasonic positioning technology, visible light positioning technology, UWB technology, ZigBee technology, radio frequency tag identification technology, and computer vision positioning technology. Because these technologies need to build equipment for auxiliary positioning, the higher the positioning accuracy, the more the number of auxiliary equipment is required, so these technologies are not universally applicable and fast. Wi-Fi positioning technology needs to estimate the location of the receiver by setting up a Wi-Fi network indoors. Visible light positioning technology needs to clarify the position of the visible light source in advance. If th...

AI Technical Summary

Problems solved by technology

The currently used strapdown inertial navigation system uses acceleration data integration to obtain the step length. Since the acceleration measured by MEMS is not high in accuracy, the inertial navigation positioning error will accumulate rapidly.

Images