To help supervisors understand the positions of workers in real-time, provide safety guidance for workers and reduce the occurrence of accidents. This study proposes a real-time positioning algorithm based on multi-source information coupling, aiming to solve the problem of workers’ autonomous positioning in signal-blind areas.
The proposed algorithm utilizes the visual SLAM and IMU sensors to perceive the environment, construct three-dimensional images, improve the accuracy of corner point matching, pre-integrate the raw IMU data, and adopt the tightly coupled method to couple the visual and inertial navigation data, thereby establishing a binocular visual SLAM and IMU coupling real-time positioning model.
The real-time positioning technology based on the coupling of visual SLAM and IMU shows good positioning effect and calculation speed in indoor sites, has good adaptability and accuracy in different building construction scenarios, and the positioning error can be controlled within 3%.
The successful construction of the real-time positioning method effectively alleviates the problem of inaccurate positioning caused by signal blind areas in the existing positioning management system, helps protect the lives and safety of construction site workers and improves the management efficiency of construction site supervisors.
