An Experimental Study of Robust Distributed Multi-Robot Data Association from Arbitrary PosesReportar como inadecuado


An Experimental Study of Robust Distributed Multi-Robot Data Association from Arbitrary Poses


An Experimental Study of Robust Distributed Multi-Robot Data Association from Arbitrary Poses - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

In this work, we experimentally investigate the problem of computingthe relative transformation between multiple vehicles from corresponding interrobotobservations during autonomous operation in a common unknown environment.Building on our prior work, we consider an EM-based methodology whichevaluates sensory observations gathered over vehicle trajectories to establish robustrelative pose transformations between robots. We focus on experimentallyevaluating the performance of the approach as well as its computational complexityand shared data requirements using multiple autonomous vehicles aerialrobots. We describe an observation subsampling technique which utilizes laserscan autocovariance to reduce the total number of observations shared betweenrobots. Employing this technique reduces run time of the algorithm significantly,while only slightly diminishing the accuracies of computed inter-robot transformations.Finally, we provide discussion on data transfer and the feasibility ofimplementing the approach on a mesh network.



Computational Perception and Robotics - Computational Perception and Robotics Publications -



Autor: Nelson, Erik - Indelman, Vadim - Michael, Nathan - Dellaert, Frank - -

Fuente: https://smartech.gatech.edu/







Documentos relacionados