LAUSR

Fast and flexible dense reconstruction has been extensively studied due to its wide application. Considering the efficiency of multi-robot systems, cooperative reconstruction is gaining attention. Classic methods rely on inter-robot loop closures, which cannot work when there is no common area between robots. Ultra-wideband sensors provide distance measurement and can replace loop closures in multi-robot systems. However, ranging-based schemes do not solve the inconsistency of reconstruction, which is caused by odometry drift. We propose a range-aided cooperative localization and consistent reconstruction system. First, the system's front end tightly couples visual odometry and ranging to perform loop-independent cooperative reconstruction and reduce odometry drift. Second, the back end detects overlapping regions in submaps and proposes a novel dense pose graph optimization (PGO) step to further eliminate all trajectory drift and achieve consistent reconstruction. Extensive field experiments demonstrate the effectiveness and high performance of the system.