Distributed Object Subtraction 3D Mapping

When using RGB-D SLAM algorithms, a robot often takes into account dynamic and static objects during the 3D mapping step. This paper aims to present a novel approach in the mapping step of Simultaneous Localization and Mapping (SLAM) and map merging algorithms to a multi-robots system. The Object Subtraction Mapping (OS-Map) algorithm subtracts movable objects from the 3D environment and creates a clean 3D map. Therefore, it will be possible to merge the maps in a multi-robot scenario. With the availability of low-cost RGB-D sensors such as the Microsoft Kinect, our approach can be applied to low-cost robots. Both algorithms were applied to a TurtleBot 2 platform which has a RGB-D sensor as its main sensor. The results demonstrated the efficiency on generating and merging clean and accurate maps, using the OS-Map algorithm alongside a RGB-D SLAM approach.