�ݺ�ߣ

Towards the Deployment of Cloud Robotics at
Factory Shop Floors: a Prototype for Smart
Material Handling
Date: MAY, 2018
Contact information
Tampere University of Technology,
FAST Laboratory,
P.O. Box 600,
FIN-33101 Tampere,
Finland
Email: fast@tut.fi
www.tut.fi/fast
Conference: 1st International
Conference on Industrial Cyber-Physical
Systems (ICPS-2018). Saint-Petersburg,
RUSSIA – MAY 15-18, 2018
Title of the paper: Towards the
Deployment of Cloud Robotics at
Factory Shop Floors: a Prototype for
Smart Material Handling
Authors: Ali Hussnain, Borja Ramis
Ferrer, Jose L. Martinez Lastra.
If you would like to receive a reprint of
the original paper, please contact us

Towards the Deployment of Cloud
Robotics at Factory Shop Floors: a
Prototype for Smart Material
Handling
Ali Hussnain, Borja Ramis Ferrer, Jose L. Martinez Lastra
ali.Hussnain@student.tut.fi, {borja.ramisferrer, jose.lastra}@tut.fi
Tampere University of Technology, Tampere, Finland
1st International Conference on Industrial Cyber-Physical Systems (ICPS-
2018). Saint-Petersburg, RUSSIA – MAY 15-18, 2018

Outline
Factory Shop Floors: a Prototype for Smart Material
Handling
3
• Introduction
• Objectives
• Cloud Computing
• Cloud Robotics
• Problem Statement
• The approach
• Technologies
• Study Case
• Conclusions and further work

Handling
4
Introduction
• Industry 4.0 and Cyber-Physical Systems (CPS)
• Cloud Computing from Cyber World can be integrated to
Physical World for enhancing the capabilities of systems
on industrial shop floor.

Handling
5
Objectives
• The ultimate objective of this research work is to
implement Cloud Robotics in order to develop a basic
system for smart material handling.
• As a requirement, cloud based image processing tools
and embedded hardware for material handling must be
integrated

Background: Cloud Computing (1/2)
Handling
6
Source: Eurostat – http://ec.europa.eu/eurostat/statistics-explained/
index.php/Cloud_computing_-_statistics_on_the_use_by_enterprises

Background: Cloud Computing (2/2)
• Provide remote resources which can be
called upon through proper communication
channels.
• Storage, Complex Algorithms, …
• Applications
– Image processing, voice recognition and data
handling
Handling
7

Background: Cloud Robotics (1/3)
• Mentioned by J.J. Kuffner in 2010
• Robots are restricted to perform in real
environments
• Limited amount of processing, storage, and
programming capabilities
Handling
8

• Platform of networked robots can utilize
parallel processing powers and share their
knowledge with a network of robots.
• Resources are not physically situated with
the robot but can be requested
Handling
9

• Minimum onboard hardware needed for
processing
• Longer battery life
• Knowledge sharing… “Different bodies
common minds”
• Same cloud resource can be used by multiple
robots.
Handling
10

Problem Statement (1/3)
Handling
11
? ? ? ? ?
Context

• Context identification crucial for material
handling and flow.
• Context maybe its name, color ,weight etc.
anything describe the item.
• Efficient material handling is important for
overall productivity of modern industries
Handling
12

• QR Codes
• Barcode
• RFID
• Manual handling
• Need to be designed and printed additionally.
• Items in ware houses are needed to be
sorted before placing at the right position.
Handling
13
Current
scenarios

Approach (1/3)
• Acquire picture of item through camera
• Use cloud image processing tools for image
analysis
• Acquire Label (text) out of the picture
• Use context (label) for making decision
regarding material handling.
Handling
14

Approach (2/3)
• Components
Handling
15

Approach (3/3)
• Mapping agent responsible for mapping the
context of items in area
• Supervisory agent responsible for
communication between devices and storing
the information of items on pallet
• Lifter agent responsible for bringing desired
product to user
• Web Interface permits users to request
mapping and retrieve items
Handling
16

Handling
17
Technologies
• NodeJS and ExpressJS for communication
developments in agents
• Python for picture acquisition and interaction with Google
Cloud Vision API
• Google Cloud Vison API for cloud image processing

Study Case (1/6)
• Two LEGO robots with Raspberry Pi
Controller
– One Robot with Pi cam is operated by
mapping agent
– Supervisory agent is a computer but also can
be embedded hardware
Handling
18

Study Case (1/6)
• Two LEGO robots with Raspberry Pi
Controller
Handling
19

Study Case (2/6)
• Web interface for communicating and
instruction to system component.
Handling
20

Study Case (3/6)
• Google Cloud Vision API for image
processing
• Works on Representational State Transfer
(REST) architectural style
• Secure, Credential for each user
• Efficient processing
Handling
21

Study Case (4/6)
Handling
22
PA PB
PC PD
PE
PJ PI
PH PG
PF
Mapping
Robot
Lifter
Robot
Scenario
1
2

Study Case (5/6)
• User select map area option
• Mapping agent in robot map the all pallets in
the area
• Acquire image and process it in cloud for
labels, store with supervisory agent
• User input product required in interface
• Lifter agent in robot pic the product for user
Handling
23

Study Case (6/6)
Handling
24

Handling
25
Conclusions and further work
• This work presents an approach for smart material
handling using the concept of Cloud Robotics
• Minimum onboard resources required, cloud image
processing
• Google Cloud Vison API can be used by multiple robots
in field
• Further, the approach will be implemented in an
industrial scenario for enhancing the flexibility of
manufacturing systems. This will be in the scope of using
cloud image processing tools for context awareness in
manufacturing industry

Acknowledge
The research leading to these results has received funding from the European
Union’s Horizon 2020 research and innovation program under grant agreement
n 636909, correspondent to the project shortly entitled C2NET, Cloud
Collaborative Manufacturing Networks .

Handling
27
THANK YOU!
Any questions?
http://www.youtube.com/user/fastlaboratory
https://www.facebook.com/fast.laboratory
http://www.slideshare.net/fastlaboratory

�ݺ�ߣ

Towards the Deployment of Cloud Robotics at Factory Shop Floors: a Prototype for Smart Material Handling

More Related Content

Towards the Deployment of Cloud Robotics at Factory Shop Floors: a Prototype for Smart Material Handling