SLAMORE: Simultaneous Localization and Mapping with Object Recognition for 3D Radio Environment Reconstruction

To enable accurate yet efficient 3D radio propagation modeling in dynamic environment, we rise to the challenge of effectively extracting and reconstructing the 3D radio propagation environment. We propose a feature and object-based monocular SLAM algorithm called SLAMORE, that can detect, track, localize, and reconstruct the major obstacles for electromagnetic waves. The embedded convolutional object detector recognizes objects as major obstacles, poses adaptive constraints to associate the 3D map points with the objects, and provides reference to estimate the room scale and real world coordinate. We conduct a proof of concept in an office environment and show that the mean absolute percentage errors for room size estimation and object position estimation achieve approximately 2.5-6% and 2-3%, respectively. We also show that SLAMORE is a viable method to enable efficient 3D radio propagation modeling and augmented radio map visualization.