Tower Crane–Mounted Multisensor System for Real-Time 3D Spatial Mapping in Modular Integrated Construction

Tower crane operations (TCOs) in modular integrated construction (MiC) face significant safety risks due to limited operator situational awareness when handling heavy, bulky modules in congested sites. Existing approaches, including fixed-site sensors and crane-mounted vision systems, do not provide practical, operator-perspective three-dimensional (3D) maps in real time. This study addresses this gap by proposing and experimentally validating a crane-mounted multisensor 3D spatial mapping framework that integrates LiDAR and an red, green, blue (RGB) camera to generate dense, colorized, real-time 3D maps from the operator’s viewpoint. The system leverages Robust, Real-time, RGB-colored, LiDAR-Inertial-Visual tightly-coupled state Estimation and mapping (R3LIVE), a simultaneous localization and mapping (SLAM)-based framework known for real-time applications, to dynamically reconstruct high-fidelity 3D spatial maps during crane movement. Laboratory experiments were conducted in a controlled indoor environment using a scaled tower crane model under both ideal and low-visibility conditions. System performance was evaluated using decision-relevant metrics, including map completeness of the region of interest and object localization accuracy. Results demonstrated 100% map completeness within the region of interest under both ideal and low-visibility conditions. Additionally, the system maintained subcentimeter localization accuracy under ideal conditions [mean absolute error ( MAE ) = 0.0096 m and root-mean square error ( RMSE ) = 0.0102 m ] with robust centimeter-level precision under challenging lighting ( MAE = 0.015 m and RMSE = 0.019 m ). These findings contribute to the body of knowledge by proposing and validating a crane-mounted multisensor fusion framework that overcomes the limitations of static perspective in fixed-site systems and the lighting vulnerabilities of vision-only approaches. The proposed framework advances existing crane monitoring and mapping techniques by enabling continuous, operator-centric, real-time 3D spatial reconstruction in dynamic construction environments. However, translating these technical capabilities into full-scale operational environments requires further validation in real-world conditions.