Javascript must be enabled for the correct page display

Learning to Detect Grasp Affordances of 3D Objects using Deep Convolutional Neural Networks

Li, Yikun (2019) Learning to Detect Grasp Affordances of 3D Objects using Deep Convolutional Neural Networks. Master's Thesis / Essay, Artificial Intelligence.

[img]
Preview
Text
mAI_2019_YikunLi.pdf

Download (21MB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (113kB)

Abstract

Grasp affordances detection is one of the challenging tasks in robotics because it must predict the grasp configuration for the object of interest in real-time to enable the robot to interact with the environment. In this research, we present a new deep learning approach to detect object affordances for a given 3D object. The method trains a Convolutional Neural Network (CNN) to learn a set of grasping features from RGB-D images. We name our approach Res-U-Network since the architecture of the network is designed based on U-Network structure and residual network-styled blocks. It devises to be robust and efficient to compute and use. A set of experiments has been performed to assess the performance of the proposed approach regarding grasp success rate on simulated robotic scenarios. Experiments validate the promising performance of the proposed architecture on a subset of ShapeNetCore dataset and simulated robot scenarios.

Item Type: Thesis (Master's Thesis / Essay)
Supervisor:
Supervisor nameSupervisor E mail
Schomaker, L.R.B..L.R.B.Schomaker@rug.nl
Kasaei, H.Hamidreza.Kasaei@rug.nl
Degree programme: Artificial Intelligence
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 31 Aug 2019
Last Modified: 20 Nov 2019 10:27
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/20874

Actions (login required)

View Item View Item