Intelligent Path Recognition against Image Noises for Vision Guidance of Automated Guided Vehicles in a Complex Workspace

Applying computer vision to mobile robot navigation has been studied more than two decades. The most challenging problems for a vision-based AGV running in a complex workspace involve the non-uniform illumination, sight-line occlusion or stripe damage, which inevitably result in incomplete or deform...

Full description

Saved in:
Bibliographic Details
Published in:Applied sciences 2019-10, Vol.9 (19), p.4108
Main Authors: Wu, Xing, Sun, Chao, Zou, Ting, Xiao, Haining, Wang, Longjun, Zhai, Jingjing
Format: Article
Language:English
Subjects:
Accuracy
Approximation
Approximation method
Automated guided vehicles
Automatic vehicle identification systems
Automation
Back propagation networks
Cameras
Computer vision
curve approximation
Digital signal processors
Factories
Illumination
Image acquisition
Image processing
Low cost
Methods
model estimation
Neural networks
Occlusion
Optimization
Parameter estimation
path classification
Pixels
Principal components analysis
Robots
Running in
Stream flow
vision guidance
Visual perception
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Applying computer vision to mobile robot navigation has been studied more than two decades. The most challenging problems for a vision-based AGV running in a complex workspace involve the non-uniform illumination, sight-line occlusion or stripe damage, which inevitably result in incomplete or deformed path images as well as many fake artifacts. Neither the fixed threshold methods nor the iterative optimal threshold methods can obtain a suitable threshold for the path images acquired on all conditions. It is still an open question to estimate the model parameters of guide paths accurately by distinguishing the actual path pixels from the under- or over- segmentation error points. Hence, an intelligent path recognition approach based on KPCA–BPNN and IPSO–BTGWP is proposed here, in order to resist the interferences from the complex workspace. Firstly, curvilinear paths were recognized from their straight counterparts by means of a path classifier based on KPCA–BPNN. Secondly, an approximation method based on BTGWP was developed for replacing the curve with a series of piecewise lines (a polyline path). Thirdly, a robust path estimation method based on IPSO was proposed to figure out the path parameters from a set of path pixels surrounded by noise points. Experimental results showed that our approach can effectively improve the accuracy and reliability of a low-cost vision-guidance system for AGVs in a complex workspace.
ISSN:2076-3417
2076-3417
DOI:10.3390/app9194108