Enabling Flexibility in Manufacturing by Integrating Shopfloor and Process Perception for Mobile Robot Workers

Angelos Christos Bavelos,Sotiris Aivaliotis,Christos Gkournelos,Niki Kousi,Konstantinos Lotsaris,George Michalos,Sotiris Makris

doi:10.3390/app11093985

Angelos Christos Bavelos, Sotiris Aivaliotis + Show 5 more

Open Access

https://doi.org/10.3390/app11093985

Copy DOI

Journal: Applied Sciences	Publication Date: Apr 28, 2021
Citations: 18	License type: CC BY 4.0

Affiliation: University of Patras

Abstract

Robotic flexibility in industry is becoming more and more relevant nowadays, especially with the rise of the Industry 4.0 concept. This paper presents a smart execution control framework for enabling the autonomous operation of flexible mobile robot workers. These robot resources are able to autonomously navigate the shopfloor, undertaking multiple operations while acting as assistants to human operators. To enable this autonomous behavior, the proposed framework integrates robot perception functions for the real-time shopfloor and process understanding while orchestrating the process execution. A Digital World Model is deployed synthesizing the sensor data coming from multiple 2D and 3D sensors from the shopfloor. This model is consumed for the perception functions enabling the real-time shopfloor and process perception by the robot workers. This smart control system has been applied and validated in a case study from the automotive sector.

Highlights

Reconfigurable manufacturing systems are necessary for the development of flexible production systems [1,2]
The methods for shopfloor and process perception introduced in this paper produced
The methods for shopfloor and process perception introduced in this paper produced the results

Summary

Introduction

Reconfigurable manufacturing systems are necessary for the development of flexible production systems [1,2] For this reconfigurability to be possible, industrial robots must receive real-time sensor feedback, an important part of which is visual feedback. This is necessary for a variety of applications, such as navigation, part identification and handling, human collaboration, etc. Scene-related vision systems regard the awareness of the robot for the environment around it. Scene-related vision belongs to environment perception, which is the cognitive ability of Academic Editor: Alessandro Gasparetto. Scene-related vision belongs to environment perception, which is the cognitive ability of the robot robot to to understand understand its its environment environment in in order order to todetect detectobstacles, obstacles, areas areasof ofinterest, interest,its its the own place within a robotic cell, etc

Methods

Results

Conclusion