Autonomously updated indoor positioning method

Abstract

The invention discloses an autonomously updated indoor positioning method. The method comprises the following steps: constructing a feature extraction network; constructing a hidden Markov model; constructing a new landmark detection network; calculating two-dimensional plane coordinates of the new landmark; writing the calculated two-dimensional plane coordinates of the new landmark into a database; and obtaining a new feature extraction network by taking the ROI and the category of the new landmark and the ROI and the category of the unchanged landmark as training data. By combining the hidden Markov model and the motion recovery structure technology, the output of the feature extraction network is judged, the classification error of the feature extraction network can be further reduced,and the misjudgment rate is effectively reduced; meanwhile, a new landmark detection network is constructed; according to the method, the relative positions of the horizontal plane of the three-dimensional point cloud and the new landmark can be determined, then the three-dimensional coordinates of the new landmark are mapped back to the planar map space, indoor map updating is completed, a largeamount of manpower does not need to be consumed for updating and maintaining the indoor planar map, and the later-period maintenance cost of video indoor positioning is greatly reduced.

Description

technical field [0001] The invention relates to an indoor positioning method, in particular to a self-updating indoor positioning method. Background technique [0002] In recent years, with the rapid development of information technology and the popularization of smart phones, people's production and life work is increasingly inseparable from various mobile phone applications. There has always been a huge demand for accurate indoor location positioning. The current mainstream indoor location positioning There are two methods, active positioning and passive positioning. [0003] However, since the signal strength of active positioning will fluctuate with the change of indoor environment and the automatic adjustment of transmission power, it is necessary to re-collect fingerprint information at regular intervals, and the deployment cost of hardware is relatively high, while passive positioning cannot be used for indoor landmarks Continue to provide accurate positioning in the...

AI Technical Summary

Problems solved by technology

[0003] However, because the signal strength of active positioning will fluctuate with the change of indoor environment and the automatic adjustment of transmission power, it is necessary to re-collect fingerprint information at regular intervals, and the deployment cost of hardware is relatively high, while passive positioning cannot be used for indoor landmarks Continue to provide accurate positioning in the case of changes and updates, and when the indoor landmarks change, the misjudgment rate is relatively high. It takes a lot of manpower to update and maintain the indoor planar map and retrain the landmark classification network to reposition.

Images