Faculty mentor: Tucker Hermans
Deep learning has enabled remarkable improvements in robotic grasp synthesis for previously unseen objects from partial object views. However, existing approaches lack the ability to explicitly reason about the full 3D geometry of the object when selecting a grasp, relying on indirect geometric reasoning derived when learning grasp success networks. In this work, we utilize learned reconstructions to explicitly model geometry in a constrained optimization grasp synthesis algorithm.
Click below to hear me present my poster!
Click on the social media icons above to share this page.
Questions or comments? Contact me at: email@example.com