Abstract:This paper proposes a visual system to generate multiple and selective 6D grasps using a Convolutional Neural Network (CNN) called GraspNet and a modified version of a CNN entitled Single Shot Multibox Detector (SSD). GraspNet is a robotic grasping technique, based on Variational Autoencoders, to generate grasps from the point cloud of the object and the end effector. It is capable of generating several grasps for a single object, enabling the exploitation of viable kinematic solutions to avoid collisions with… Show more
Set email alert for when this publication receives citations?
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.