Vision-based robotic system for polyhedral object grasping using kinect sensor

Pablo Gonzalez, Ming-Yang Cheng, Wei Liang Kuo

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

Abstract

Since the introduction of the Microsoft Kinect sensor, real-Time 3D perception of objects in a close-range scene has become one of the major trends in the current research on robotics and computer vision domain. As part of a long-Term goal to develop a robust object recognition system for industrial applications, this paper focuses on the research topic of 3D planes recognition and separation for polyhedral object grasping using industrial manipulators and Kinect as a 3D vision system. Since the data acquired by the vision sensor such as Kinect are invariably noisy, a pipeline of algorithms has been implemented to reduce the noise and clustering the data points properly. With the feature vectors computed for each data point that represent the object of interest given, a split and merge algorithm has been developed to cluster the data exploiting the flexibility and robustness of the fuzzy logic inference system and then fitting each segment to a planar model using RANSAC. This work mainly uses the eye-To-hand Kinect-based vision system to retrieve the 3D position and normal vectors of the planar faces of the object and then give instructions to a robotic arm for grasping. Experimental results indicate that the proposed pipeline of algorithms written in C# for plane segmentation can successfully segment the polyhedral object, and the Kinect-based robotic vision system developed in this work can achieve real-Time automatic polyhedral object grasping with high precision.

Original languageEnglish
Title of host publication2016 International Automatic Control Conference, CACS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages71-76
Number of pages6
ISBN (Electronic)9781509041084
DOIs
Publication statusPublished - 2017 Jul 10
Event2016 International Automatic Control Conference, CACS 2016 - Taichung, Taiwan
Duration: 2016 Nov 92016 Nov 11

Publication series

Name2016 International Automatic Control Conference, CACS 2016

Other

Other2016 International Automatic Control Conference, CACS 2016
CountryTaiwan
CityTaichung
Period16-11-0916-11-11

Fingerprint

Grasping
Robotics
Sensor
Computer vision
Sensors
Vision System
Pipelines
Industrial manipulators
Robotic arms
Object recognition
Fuzzy logic
Industrial applications
Position vector
RANSAC
Real-time
Normal vector
Object Recognition
Manipulator
Industrial Application
Feature Vector

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Control and Systems Engineering
  • Control and Optimization

Cite this

Gonzalez, P., Cheng, M-Y., & Kuo, W. L. (2017). Vision-based robotic system for polyhedral object grasping using kinect sensor. In 2016 International Automatic Control Conference, CACS 2016 (pp. 71-76). [7973886] (2016 International Automatic Control Conference, CACS 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CACS.2016.7973886
Gonzalez, Pablo ; Cheng, Ming-Yang ; Kuo, Wei Liang. / Vision-based robotic system for polyhedral object grasping using kinect sensor. 2016 International Automatic Control Conference, CACS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 71-76 (2016 International Automatic Control Conference, CACS 2016).
@inproceedings{b53a3d4fe96b47ffa62b7a0691141ccc,
title = "Vision-based robotic system for polyhedral object grasping using kinect sensor",
abstract = "Since the introduction of the Microsoft Kinect sensor, real-Time 3D perception of objects in a close-range scene has become one of the major trends in the current research on robotics and computer vision domain. As part of a long-Term goal to develop a robust object recognition system for industrial applications, this paper focuses on the research topic of 3D planes recognition and separation for polyhedral object grasping using industrial manipulators and Kinect as a 3D vision system. Since the data acquired by the vision sensor such as Kinect are invariably noisy, a pipeline of algorithms has been implemented to reduce the noise and clustering the data points properly. With the feature vectors computed for each data point that represent the object of interest given, a split and merge algorithm has been developed to cluster the data exploiting the flexibility and robustness of the fuzzy logic inference system and then fitting each segment to a planar model using RANSAC. This work mainly uses the eye-To-hand Kinect-based vision system to retrieve the 3D position and normal vectors of the planar faces of the object and then give instructions to a robotic arm for grasping. Experimental results indicate that the proposed pipeline of algorithms written in C# for plane segmentation can successfully segment the polyhedral object, and the Kinect-based robotic vision system developed in this work can achieve real-Time automatic polyhedral object grasping with high precision.",
author = "Pablo Gonzalez and Ming-Yang Cheng and Kuo, {Wei Liang}",
year = "2017",
month = "7",
day = "10",
doi = "10.1109/CACS.2016.7973886",
language = "English",
series = "2016 International Automatic Control Conference, CACS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "71--76",
booktitle = "2016 International Automatic Control Conference, CACS 2016",
address = "United States",

}

Gonzalez, P, Cheng, M-Y & Kuo, WL 2017, Vision-based robotic system for polyhedral object grasping using kinect sensor. in 2016 International Automatic Control Conference, CACS 2016., 7973886, 2016 International Automatic Control Conference, CACS 2016, Institute of Electrical and Electronics Engineers Inc., pp. 71-76, 2016 International Automatic Control Conference, CACS 2016, Taichung, Taiwan, 16-11-09. https://doi.org/10.1109/CACS.2016.7973886

Vision-based robotic system for polyhedral object grasping using kinect sensor. / Gonzalez, Pablo; Cheng, Ming-Yang; Kuo, Wei Liang.

2016 International Automatic Control Conference, CACS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. p. 71-76 7973886 (2016 International Automatic Control Conference, CACS 2016).

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

TY - GEN

T1 - Vision-based robotic system for polyhedral object grasping using kinect sensor

AU - Gonzalez, Pablo

AU - Cheng, Ming-Yang

AU - Kuo, Wei Liang

PY - 2017/7/10

Y1 - 2017/7/10

N2 - Since the introduction of the Microsoft Kinect sensor, real-Time 3D perception of objects in a close-range scene has become one of the major trends in the current research on robotics and computer vision domain. As part of a long-Term goal to develop a robust object recognition system for industrial applications, this paper focuses on the research topic of 3D planes recognition and separation for polyhedral object grasping using industrial manipulators and Kinect as a 3D vision system. Since the data acquired by the vision sensor such as Kinect are invariably noisy, a pipeline of algorithms has been implemented to reduce the noise and clustering the data points properly. With the feature vectors computed for each data point that represent the object of interest given, a split and merge algorithm has been developed to cluster the data exploiting the flexibility and robustness of the fuzzy logic inference system and then fitting each segment to a planar model using RANSAC. This work mainly uses the eye-To-hand Kinect-based vision system to retrieve the 3D position and normal vectors of the planar faces of the object and then give instructions to a robotic arm for grasping. Experimental results indicate that the proposed pipeline of algorithms written in C# for plane segmentation can successfully segment the polyhedral object, and the Kinect-based robotic vision system developed in this work can achieve real-Time automatic polyhedral object grasping with high precision.

AB - Since the introduction of the Microsoft Kinect sensor, real-Time 3D perception of objects in a close-range scene has become one of the major trends in the current research on robotics and computer vision domain. As part of a long-Term goal to develop a robust object recognition system for industrial applications, this paper focuses on the research topic of 3D planes recognition and separation for polyhedral object grasping using industrial manipulators and Kinect as a 3D vision system. Since the data acquired by the vision sensor such as Kinect are invariably noisy, a pipeline of algorithms has been implemented to reduce the noise and clustering the data points properly. With the feature vectors computed for each data point that represent the object of interest given, a split and merge algorithm has been developed to cluster the data exploiting the flexibility and robustness of the fuzzy logic inference system and then fitting each segment to a planar model using RANSAC. This work mainly uses the eye-To-hand Kinect-based vision system to retrieve the 3D position and normal vectors of the planar faces of the object and then give instructions to a robotic arm for grasping. Experimental results indicate that the proposed pipeline of algorithms written in C# for plane segmentation can successfully segment the polyhedral object, and the Kinect-based robotic vision system developed in this work can achieve real-Time automatic polyhedral object grasping with high precision.

UR - http://www.scopus.com/inward/record.url?scp=85027549452&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027549452&partnerID=8YFLogxK

U2 - 10.1109/CACS.2016.7973886

DO - 10.1109/CACS.2016.7973886

M3 - Conference contribution

T3 - 2016 International Automatic Control Conference, CACS 2016

SP - 71

EP - 76

BT - 2016 International Automatic Control Conference, CACS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Gonzalez P, Cheng M-Y, Kuo WL. Vision-based robotic system for polyhedral object grasping using kinect sensor. In 2016 International Automatic Control Conference, CACS 2016. Institute of Electrical and Electronics Engineers Inc. 2017. p. 71-76. 7973886. (2016 International Automatic Control Conference, CACS 2016). https://doi.org/10.1109/CACS.2016.7973886