The Pose Adjustment System of Robotic Arm Adopts Binocular Vision and Machine Iearning

DOI

10.2991/cnci-19.2019.54How to use a DOI?

Keywords

Binocular vision, visual measurement, machine learning, neural network, robotic arm, intelligent plastering robot.

Abstract

In order to solve the problem of rapid positioning of doors, windows and frames existing in the plastering process of the intelligent plastering robot, a pose adjustment system of robotic arm based on binocular vision and machine learning was proposed. The system tested a variety of feature point detection algorithms, description algorithms and matching algorithms, and proposed a combination scheme with good detection speed, feature point number and match accuracy: SURF algorithm and FLANN matching algorithm, and added mismatching filtering and density control strategy. The innovative method of using machine learning neural network to learn the reconstructed 3D point cloud improves the system's fault tolerance ability and response speed, and can quickly convert the complexly point cloud information into the attitude adjustment information required by the robotic arm. The experimental results show that the system to a single location within 2 s time, including robotic arm action time of 0.3 s to 0.5 s, after 1 to 4 times of circulation adjustment, can achieve the displacement error of 1 mm and 1 ° rotation error. The system can meet the requirements of accuracy, time and stability in the actual working environment.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yabing Ren
AU  - Hongyang Yu
PY  - 2019/05
DA  - 2019/05
TI  - The Pose Adjustment System of Robotic Arm Adopts Binocular Vision and Machine Iearning
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 391
EP  - 401
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.54
DO  - 10.2991/cnci-19.2019.54
ID  - Ren2019/05
ER  -