Drift analysis and sectional post-processing of indoor simultaneous localization and mapping (SLAM)-based laser scanning data
Aino Keitaanniemi, Petri Rönnholm, Antero Kukko, Matti Vaaja
Abstract
Even though the SLAM laser scanners (LSs) have the required resolution for construction analysis, a drift error has been detected in georeferencing phase of the data processing. The reduction of drift error in SLAM LSs with third-party software has not been studied. This paper demonstrates the drift error behavior of commercial SLAM LSs and their correction with sectional post-processing methods, and shows that the drift error is mainly in a vertical direction, and it is largest in the middle of the trajectory. The drift error was reduced from 10.6 cm to 0.2 cm. Circa 30 s length sections have better results than larger sections with non-rigid transformation algorithm. The method has shown potential to improve the quality and cost-effectiveness of multitemporal construction documentation. In addition, locations of additional control points were analyzed to prevent drift error, with adjustments to the walking path.