Primitive shape fitting in point clouds for enabling autonomous underwater grasping

TitlePrimitive shape fitting in point clouds for enabling autonomous underwater grasping
Publication TypeConference Paper
Year of Publication2013
AuthorsFornas, D, Peñalver, A, Prats, M, Fernández, JJ, Sales, J
Conference Name5TH MARTECH International Workshop On Marine Technology
Date Published10/2013
Conference LocationGirona (Spain)
ISBN Number978-84-616-5764-3
Keywordsautonomous grasping, point cloud, RANSAC, shape fitting, underwater
Abstract

Autonomous grasping of unknown objects by a
robot is a highly challenging skill that is receiving increasing
attention in the last years, and is still more challenging in
underwater environments, with highly unstructured scenarios,
limited availability of sensors and adverse conditions that affect
the robot perception and control systems. This paper describes
an approach for increase the autonomy of grasping operations
on underwater primitive shaped objects from floating vehicles,
in particular cylinder shaped objects like an amphora. Various
sources of stereo information are used to gather 3D information in
order to obtain a model of the object. Using a RANSAC primitive
shape recognition algorithm the model parameters are obtained
and a set of feasible grasps are computed. Then the user specifies
the best one or it is specified using analytical constraints. This
approach is tested with the UWSim simulator.