Multisensory System for Solving Problems of Reverse Engineering of Buildings

Multisensory System for Solving Problems of Reverse Engineering of Buildings

연구 분야: Analysis

학회: 2024 International Russian Automation Conference (RusAutoCon)

The paper examines the problem of reverse engineering buildings to obtain digital models. The purpose of the article is to develop and test a multisensory system for automating the construction and refinement of 2D drawings and diagrams of real objects. The work explores sensors for gathering information about object characteristics, and also proposes a developed data collection and processing system using SLAM methods. Information is gathered using a short-range 2D LiDAR; a camera for reading visual markings; an inertial measurement module, the data from which helps reduce rotation angle search time when processing LiDAR information. The developed system collects data from sensors connected to a single-board computer with the ROS software package installed. Subsequent processing of the collected data is done using MATLAB. As a result of the work, data acquisition algorithms and processing methods were identified, and the system was tested in a real environment. The results obtained during the work demonstrate that the application of the developed data collection and processing algorithms, implemented in the presented system, enables the refinement of drawings and the creation of static digital building models.