Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter

Da Wit Kim, Hyun Jun Jo, Jae-Bok Song

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Along with the development of deep learning, efforts are being made to grasping with the robot using only the camera. Above all, a lot of research is being done for grasping in an environment where various objects are mixed. To perform grasping in complex environments, it is necessary to train the grasping algorithm with vast amounts of data to ensure its robustness. However, collecting grasping data takes a lot of time and effort. In this paper, we proposed the depth tile that simply describes a complex situation by processing a depth image. Through this, the grasping algorithm can use a light artificial neural network, and training data can be generated automatically without grasping in real-world or simulation to minimize learning data collection costs. Artificial neural network trained through the depth tile can perform grasping with high success rate by estimating the grasping angle, which is less likely to interfere with obstacles. In this paper, the proposed grasping method, through experiments to empty randomly placed objects, is proved to be robust in complex environments.

Original languageEnglish
Title of host publication2019 16th International Conference on Ubiquitous Robots, UR 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages113-117
Number of pages5
ISBN (Electronic)9781728132327
DOIs
Publication statusPublished - 2019 Jun 1
Event16th International Conference on Ubiquitous Robots, UR 2019 - Jeju, Korea, Republic of
Duration: 2019 Jun 242019 Jun 27

Publication series

Name2019 16th International Conference on Ubiquitous Robots, UR 2019

Conference

Conference16th International Conference on Ubiquitous Robots, UR 2019
CountryKorea, Republic of
CityJeju
Period19/6/2419/6/27

Fingerprint

Grasping
Tile
Neural Networks
Filter
Neural networks
Cameras
Robots
Processing
Costs
Experiments
Artificial Neural Network
Robot
Camera
Likely
Deep learning
Robustness
Minimise
Angle

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Human-Computer Interaction
  • Mechanical Engineering
  • Control and Optimization

Cite this

Kim, D. W., Jo, H. J., & Song, J-B. (2019). Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter. In 2019 16th International Conference on Ubiquitous Robots, UR 2019 (pp. 113-117). [8768779] (2019 16th International Conference on Ubiquitous Robots, UR 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/URAI.2019.8768779

Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter. / Kim, Da Wit; Jo, Hyun Jun; Song, Jae-Bok.

2019 16th International Conference on Ubiquitous Robots, UR 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 113-117 8768779 (2019 16th International Conference on Ubiquitous Robots, UR 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kim, DW, Jo, HJ & Song, J-B 2019, Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter. in 2019 16th International Conference on Ubiquitous Robots, UR 2019., 8768779, 2019 16th International Conference on Ubiquitous Robots, UR 2019, Institute of Electrical and Electronics Engineers Inc., pp. 113-117, 16th International Conference on Ubiquitous Robots, UR 2019, Jeju, Korea, Republic of, 19/6/24. https://doi.org/10.1109/URAI.2019.8768779
Kim DW, Jo HJ, Song J-B. Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter. In 2019 16th International Conference on Ubiquitous Robots, UR 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 113-117. 8768779. (2019 16th International Conference on Ubiquitous Robots, UR 2019). https://doi.org/10.1109/URAI.2019.8768779
Kim, Da Wit ; Jo, Hyun Jun ; Song, Jae-Bok. / Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter. 2019 16th International Conference on Ubiquitous Robots, UR 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 113-117 (2019 16th International Conference on Ubiquitous Robots, UR 2019).
@inproceedings{7bd7b1a3dbd448b593db92ca0ef14111,
title = "Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter",
abstract = "Along with the development of deep learning, efforts are being made to grasping with the robot using only the camera. Above all, a lot of research is being done for grasping in an environment where various objects are mixed. To perform grasping in complex environments, it is necessary to train the grasping algorithm with vast amounts of data to ensure its robustness. However, collecting grasping data takes a lot of time and effort. In this paper, we proposed the depth tile that simply describes a complex situation by processing a depth image. Through this, the grasping algorithm can use a light artificial neural network, and training data can be generated automatically without grasping in real-world or simulation to minimize learning data collection costs. Artificial neural network trained through the depth tile can perform grasping with high success rate by estimating the grasping angle, which is less likely to interfere with obstacles. In this paper, the proposed grasping method, through experiments to empty randomly placed objects, is proved to be robust in complex environments.",
author = "Kim, {Da Wit} and Jo, {Hyun Jun} and Jae-Bok Song",
year = "2019",
month = "6",
day = "1",
doi = "10.1109/URAI.2019.8768779",
language = "English",
series = "2019 16th International Conference on Ubiquitous Robots, UR 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "113--117",
booktitle = "2019 16th International Conference on Ubiquitous Robots, UR 2019",

}

TY - GEN

T1 - Grasping Method in a Complex Environment using Convolutional Neural Network Based on Modified Average Filter

AU - Kim, Da Wit

AU - Jo, Hyun Jun

AU - Song, Jae-Bok

PY - 2019/6/1

Y1 - 2019/6/1

N2 - Along with the development of deep learning, efforts are being made to grasping with the robot using only the camera. Above all, a lot of research is being done for grasping in an environment where various objects are mixed. To perform grasping in complex environments, it is necessary to train the grasping algorithm with vast amounts of data to ensure its robustness. However, collecting grasping data takes a lot of time and effort. In this paper, we proposed the depth tile that simply describes a complex situation by processing a depth image. Through this, the grasping algorithm can use a light artificial neural network, and training data can be generated automatically without grasping in real-world or simulation to minimize learning data collection costs. Artificial neural network trained through the depth tile can perform grasping with high success rate by estimating the grasping angle, which is less likely to interfere with obstacles. In this paper, the proposed grasping method, through experiments to empty randomly placed objects, is proved to be robust in complex environments.

AB - Along with the development of deep learning, efforts are being made to grasping with the robot using only the camera. Above all, a lot of research is being done for grasping in an environment where various objects are mixed. To perform grasping in complex environments, it is necessary to train the grasping algorithm with vast amounts of data to ensure its robustness. However, collecting grasping data takes a lot of time and effort. In this paper, we proposed the depth tile that simply describes a complex situation by processing a depth image. Through this, the grasping algorithm can use a light artificial neural network, and training data can be generated automatically without grasping in real-world or simulation to minimize learning data collection costs. Artificial neural network trained through the depth tile can perform grasping with high success rate by estimating the grasping angle, which is less likely to interfere with obstacles. In this paper, the proposed grasping method, through experiments to empty randomly placed objects, is proved to be robust in complex environments.

UR - http://www.scopus.com/inward/record.url?scp=85070551071&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070551071&partnerID=8YFLogxK

U2 - 10.1109/URAI.2019.8768779

DO - 10.1109/URAI.2019.8768779

M3 - Conference contribution

AN - SCOPUS:85070551071

T3 - 2019 16th International Conference on Ubiquitous Robots, UR 2019

SP - 113

EP - 117

BT - 2019 16th International Conference on Ubiquitous Robots, UR 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -