Indoor visual positioning method based on ceiling lamp

Abstract

The invention relates to an indoor visual positioning method based on a ceiling lamp, which is used for realizing the positioning of a movable object in a target indoor area, a lamp for realizing illumination is arranged at the top of the target indoor area, the movable object is provided with a calculation unit, and the positioning is realized by utilizing a camera which is arranged on the movable object and is used for shooting upwards and utilizing the lamp. The method has the advantages of high positioning precision, high real-time performance and relatively low hardware cost.

Description

technical field [0001] The invention relates to a method for positioning a movable object located in an indoor space. Background technique [0002] In the absence of GNSS signals, it is an engineering problem to achieve high-precision positioning of vehicles indoors. It was rated as one of the top 10 engineering problems by the China Association for Science and Technology. At present, in order to realize the high-precision positioning of vehicles indoors, it is necessary to carry out field end transformation, or to establish a high-precision field end map. At the same time, it is also necessary to combine the vehicle-side sensors and computing units to complete the matching with the high-precision field-side map to achieve the positioning effect. [0003] At present, the following technologies are usually used to achieve high-precision positioning of vehicles indoors: 1) Lidar is used. Although Lidar has high positioning accuracy and can be used indoors and outdoors, it is ...

AI Technical Summary

Problems solved by technology

[0003] At present, the following technologies are usually used to achieve high-precision positioning of vehicles indoors: 1) Lidar is used. Although Lidar has high positioning accuracy and can be used indoors and outdoors, it is expensive and not suitable for popularization

2) Using the camera with the vision facing down to achieve positioning requires multiple cameras to shoot downwards. If the ground indication line is damaged, blocked, or the ground is reflected, the positioning will be inaccurate or even invalid.

3) The millimeter-wave radar is used for positioning. Due to the inaccurate ranging of the millimeter-wave radar, the positioning accuracy is low and the applicable scenarios are limited.

6) The problem of Bluetooth 5.0aoa positioning technology is the same as that of UWB technology, and it also needs to rely on array antennas

7) The accuracy error of the WiFi positioning method is 3 meters, which is not suitable for the automatic tracking function of the underground car garage

Although visual features consume less computing resources than semantics, they are greatly affected by light changes

11) Positioning is realized based on the forward-looking lane line, but this positioning method is mostly used on expressways. Under the condition of road congestion, the forward vision is easily blocked, resulting in inaccurate positioning or even failure