Asynchronous Collaborative Autoscanning with Mode Switching for Multi-Robot Scene Reconstruction

When conducting autonomous scanning for the online reconstruction of unknown indoor environments, robots have to be competent at exploring scene structure and reconstructing objects with high quality. Our key observation is that different tasks demand specialized properties robots: rapid moving speed far vision global exploration slow narrow local object reconstruction, which are referred as two modes: explorer reconstructor, respectively. requiring multiple collaborate efficient fine-grained questions on when generate how assign those should carefully answered. Therefore, we propose a novel asynchronous collaborative autoscanning method mode switching, generates kinds associated modes, i.e., task reconstructor mode, them execute in an manner highly boost efficiency The assignment optimized by solving modified Multi-Depot Multiple Traveling Salesman Problem (MDMTSP). Moreover, further enhance collaboration increase efficiency, task-flow model actives generation process immediately any finish all its no need wait other complete assigned previous iteration. Extensive experiments been conducted show importance each component our superiority over methods