ANALISIS PENGARUH JARAK VERTIKAL DAN HORIZONTAL TERHADAP HASIL AKURASI REGISTRASI POINT CLOUDS DARI PENGUKURAN TERRESTRIAL LASER SCANNER (TLS) UNTUK DETEKSI PENYIMPANGAN LATERAL OBJEK BANGUNAN BERTINGKAT (STUDI KASUS: APARTEMEN X DI BANDUNG)

• Main Author: Anjelia Selan, Murni
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/65137
• Institution: Institut Teknologi Bandung

Monitoring of high-rise structures is important to see the impact of changes in the structure of the building not only on the structure itself but on the surrounding area. One of the monitoring methods that can be done is to use a Terrestrial Laser Scanner (TLS). The quality of TLS measurement data is influenced by measurement techniques and strategies, one of which is the distance between the scanning tool and the object. In this final project, an analysis is carried out regarding the effect of the vertical and horizontal distances of objects using TLS by looking at the accuracy of the results of the point clouds registration for the difference in the distance between the tool and the object and the difference in the coordinates of the point clouds resulting from data processing with the coordinates of measurements using a Robotic Total station (RTS). This study also analyzes the lateral deviation of the building seen from the coordinates of the point clouds data resulting from data processing in every corner of the building on each floor and makes a 3-dimensional model of the research object using the scan-to-BIM method and parametric modeling. The case study of this research is an apartment in Bandung which consists of 28 floors. The data collection method was carried out by scanning the building exterior using TLS (free scan and traverse scan), RTS using GPS control points, and GPS measurements using the static differential method. Through the stages of data processing, namely post-processing of GPS data, processing of coordinates from RTS, processing of point clouds from TLS measurements in the form of georeferencing, filtering, and registration, as well as 3-dimensional modeling, point clouds is obtained for analysis and 3-dimensional modeling of the research object. Analysis of the effect of the distance between the tool and the object on the registration results using the Iterative Closest Point (ICP) algorithm shows that the Root Mean Square Error (RMSE) value of the registration results is directly proportional to the vertical and horizontal distances, with a maximum RMSE value of 0.06 mm from the registration results and 6.8 cm from validation with RTS at an average vertical distance of 98 m and an average horizontal distance of 55 m, as well as a minimum RMSE of 0.02 mm from the registration results and 2.8 cm from validation with RTS at an average vertical distance and the smallest horizontal (8 m and 36 m). The lateral deviation of the building towards the X and Y axes tends to get bigger as the floor increases with the largest deviation value towards the X axis is 13 cm and the largest deviation towards the Y axis is 23 cm on the 28th floor, with an average deviation direction towards the axis The X is positive and the Y axis is positive. The 3-dimensional model has an accuracy of LoD3, with an RMSE value of 7.6 cm.