EVALUATION OF EXISTING ROBOT TECHNOLOGIES FOR DEEP LEVEL MINING APPLICATIONS

thunderclingAI and Robotics

Nov 13, 2013 (4 years ago)

151 views

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

i


EVALUATION OF
EXISTING
ROBOT
TECHNOLOGIES FOR DEEP LEVEL
MINING APPLICATIONS


S
TUDENT
:

S.E.

L
OUW

15411273

S
TUDY
L
EADER
:

M
R
S
TEPHEN
M
ATOPE

8

D
ECEMBER
2011















Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

ii


Acknowledgements


Many thanks go out to everyone that assisted in making this study a success.
Without their
contribution this study would not have been possible.

We are
grateful

to the mining companies, specifically AngloGold
Ashanti, which

helped us to gain a
better understanding of
the mining environment. We thank VCI technology for their assistance in
helping to design a mechanized mining concept.

We give special thanks to Je
remy Green, from CSIR, for his contribution of literature and advice
given during the course of the study.

We would like to acknowledge and thank Dr. Noel du Toit, personnel at California Institute of
Technology, which gave valuable input regarding the con
straints of deep level mining.

Additionally, we thank the ISEM Conference

2011, held at Spier, South Africa
(Western

Cape
Province)

for providing us the opportunity to be part of it and share our study outcomes.


Fi
nally, Dr. G.T. Oosthuizen, Sonja Louw, a
nd Amir Shaalane

and the Stellenbosch University
Department of Industrial Engineering for giving the opportunity to explore and study this project.












Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

iii


Declaration


I Sonja Etherisia Louw,
student number 15411273,
hereby declare t
hat I am aware of
Stellenbosch
University’s

plagiarism rules and regulations. I understand that if I am found to be guilty of
plagiarism that appropriate action can be taken against me.

I declare that I did this
thesis is my

on my
own original work

and that I have not prev
iously in its
entirety or in part submitted it at any university for a degree


If it is found that I was in any way not truth
ful

I accept the consequences and actions that are
appropriate in this regard.



……………………………………………





………………………………………….

Signature







Date













Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

iv


ECSA Exit Level Outcomes

Table 1 show the

ECSA Outcomes assessed in this module
. They include:

O
UTCOME
1:


P
ROBLEM
S
OLVING

Demonstrate competence to identify,
assess, formulate and solve convergent
and divergent engineering problems
creatively and innovatively


Deep level mining risks were identified as a challenge economically as well as the use of
humans in such environments. Information about the environment associated with
deep level mining were collected and analysed. Possible ap
proaches were generated
and formulated for the solution of such environment. Also, possible robotic solutions
were evaluated and the near optimal one were chosen and presented in this report


O
UTCOME
5:


E
NGINEERING
M
ETHODS
,

S
KILLS AND
T
OOLS
,

INCLUDING
I
N
FORMATION
T
ECHNOLOGY

Demonstrate competence to use
appropriate engineering methods, skills
and tools, including those based on
information technology


Design thinking approach was used to find an innovative solution for the problem. Also,
decision tables w
ere used as a tool to compare the robotic technology against each
other. Life cycle approach was used as a tool to assess the constraints of deep level
environment and therefore evaluate robotic
technology. Also, Microsoft Project was
used to plan this stu
dy, and Inventor CAD was used to design a very basic concept.

O
UTCOME
6:

P
ROFESSIONAL AND
T
ECHNICAL
C
OMMUNICATION

Demonstrate competence to
communicate effectively, both orally
and in writing, with engineering
audie
nces and the community at large


A project proposal, progress report and final report
were

submitted during the course
of the project duration. These were uploaded on the EDEN website where study leaders
could view the documents and make relevant recommendations and advice on the
documents. Also, taking part at the 2011 ISEM conference, an ISEM

paper on the
project was submitted and a presentation was done at the conference about the
project. Numerous experts in the field of mining were contacted for more information.
Finally, a presentation will be done on the project for the general public and

where two
external examiners will be present.

O
UTCOME
9:

I
NDEPENDENT
L
EARNING
A
BILITY

Demonstrate competence to engage in
independent learning through

well
developed learning skills


Adequate research was done on deep level mining environment. The sources of
research were evaluated and applied in this study. The use of design thinking approach
is a new way of assessing a problem and was used to conduct this study.

O
UTCOME
10:

E
NGINE
ERING
P
ROFESSIONALISM

Demonstrate critical awareness of the
need to act professionally and ethically
and to exercise judgement and take
responsibility within own limits of
competence

Dealing with experts throughout the course of the study required continuo
us
professionalism. Also, as a result of the variety of robotic technology available own
judgement was needed to decide which robot technologies are more applicable for
evaluation for this study.
It may be that there exist better applications, but what is
presented here, creates a step change for future studies. Also included is a declaration
on work done and acceptance for actions

Table
1
: ECSA Exit Level Outcomes


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

v



ABSTRACT


Mining starts with the extraction of
underground

resources, but quickly progresses to more complex
situations. As the mining depth increase the technical challenges and diffic
ulty to retrieve resources
rise
s
. The future deep level mining environment is considered too immense a risk for human labour.
The
refore, robot technology is considered as an alternative. This imposes the need to develop and
improve current mining technology and equipment. This study evaluates robot technologies for
deep level mining applications. Firstly, the constraints of robots a
ssociated in deep
-
level mining
environments are identified. Thereafter, various existing robot technologies are analysed to
categorize functional attributes of each robot. These were assessed with regard to the constraints,
establishing a basis for selecti
on of feasible robot technology. Recommendations are made on how
to improve the existing robot technology to compensate for specific conditions. It is concluded that
it is vital to develop improved technology on existing robots technologies in order to min
e at deep
levels. In collaboration with technology
-

and mining companies a mechanized mining concept was
developed from these evaluations.














Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

vi


OPSOMMING


Mynbou begin met die onttrekking van ondergrondse hulpbronne, maar vorder vinnig tot meer
kom
plekse situasies. As die mynbou
-
diepte verhoog
, so
doende styg die

tegniese uitdagings en
problem om die hulpbronne

op te haal. Die toekoms
tige

d
iep vlak mynbou
-
omgewing word beskou
as n veels te ernorme risiko
vir menslike arbeid
. A
s


n gevolg, word robot
tegnologie beskou as
'n
alternatief. Dit plaas die behoefte om te ontwikkel en te verbeter

op

huidige mynbou
-
tegnologie en
toerusting. Hierdie studie

evalueer robot tegnologie vir
'n diep vlak mynbou
-
gebruike
. Eerstens is die
beperkings van die robotte wat

in diep vlak mynbou omgewings geïdentifiseer. Daarna is verskeie
bestaande robot tegnologie ontleed
om
funksionele eienskappe van elke robot te kategoriseer.
Hierdie is beoordeel met betrekking tot di
e beperkinge, tot stigting van
'n basis vir die keuring

van
haalbaar robot tegnologie. Aanbevelings word gemaak oor hoe om die bestaande robot tegnologie
te verbeter om te vergoed vir spesifieke toestande.
Dit is die gevolgtrekking gekom dat dit
noodsaaklik is om verbeterde tegnologie op bestaande robotte tegn
ologie in om te myn diep vlakke
te ontwikkel. In samewerking met t
egnologie
-
en mynmaatskappye is
'n

gemeganiseerde mynbou
-
konsep

ontwikkel uit hierdie evaluerings.













Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

vii


Table of Contents

L
IST
OF
F
IGURES

................................
................................
................................
................................
....

ix

L
IST OF
T
ABLES

................................
................................
................................
................................
......

x

G
LOSSARY

................................
................................
................................
................................
............

xi

Chapter 1 : INTRODUCTION

................................
................................
................................
....................

1

1.1

Background

................................
................................
................................
.............................

1

1.2

PURPOSE OF STUDY

................................
................................
................................
................

3

1.3

Rese
arch Approach

................................
................................
................................
.................

3

Chapter 2 : Robot Technologies

................................
................................
................................
..............

6

2.1

Introduction

................................
................................
................................
............................

6

2.2

Rock Breaking

................................
................................
................................
..........................

6

2.3

Monitori
ng of rock fall hazard risk

................................
................................
..........................

6

2.4

Types of robots and their applications

................................
................................
...................

7

2.4.1

iRobot 210 Negotiator

................................
................................
................................
....

8

2.4.2.

iRobot 510 PackBot

................................
................................
................................
.......

10

2.4.3

iRobot 710 Warrior

................................
................................
................................
.......

12

2.4.4

The Big Dog

................................
................................
................................
...................

13

2.4.5

The LS3
-
Legged Squad Support System

................................
................................
........

15

2.4.6

RiSE
-

The Amazing Climbing Robot

................................
................................
...............

16

2.4.7

RHex

................................
................................
................................
..............................

17

2.4.8

The Remotec ANDROS F6A

................................
................................
...........................

18

2.4.9

Samsung Tango Vacuum Cleaner Robots

................................
................................
.....

19

2.4.10

Petman (The Protection Ensemble Test Mannequin)

................................
...................

20

2.4.11

da Vinci Surgical System

................................
................................
................................

21

2.4.12

ASIMO, the walking robot

................................
................................
.............................

23

2.4.13

Rolling Robot

................................
................................
................................
.................

24

2.4.14

Lucas, ASORO Service Robot

................................
................................
.........................

24

2.4.15

Vstone Robovie M
-
Version3

................................
................................
.........................

25

2.4.16

Kyosho
-
Manoi
-
AT01

................................
................................
................................
......

26

2.4.
17

Water Strider
................................
................................
................................
.................

27

2.4.18

The Groundhog

................................
................................
................................
.............

27

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

viii


2.4.19

K10 Black and K10 Red Moon robots

................................
................................
............

28

2.4.20

M2 Robot

................................
................................
................................
......................

29

2.5

Conclusi
on

................................
................................
................................
.............................

30

Chapter 3 : Life Cycle of Robots

................................
................................
................................
............

31

3.1

Introduction

................................
................................
................................
..........................

31

3.2

Challenges
-

Life Cycle

................................
................................
................................
............

31

3.
2.1

Installation and Setup

................................
................................
................................
...

32

3.2.2

Optimization and Usage

................................
................................
................................

32

3.2.3

Maintenance

................................
................................
................................
.................

33

3.2.4

Disposal and Redesign

................................
................................
................................
..

33

3.3

Conclusion

................................
................................
................................
.............................

34

Chapter 4 : EVALUATION OF DIFFERENT ROBOT TECHNOLOGIES

................................
........................

35

4.1

Introduction

................................
................................
................................
..........................

35

4.2

Evaluation Process

................................
................................
................................
................

35

4.3

Conclusion

................................
................................
................................
.............................

42

Chapter 5 : PO
SSIBLE SOLUTIONS AND RECOMMENDATIONS

................................
.............................

43

5.1

Introduction

................................
................................
................................
..........................

43

5.2

Recommendations and Solutions

................................
................................
.........................

43

5.3

Conclusion

................................
................................
................................
.............................

45

Chapter 6 : CONCLUSION

................................
................................
................................
......................

46

References

................................
................................
................................
................................
........

49

Appendices

................................
................................
................................
................................
........

54

Appendix A
-

List of Robots

................................
................................
................................
................

55

Appendix B


Questionnaire

................................
................................
................................
.............

57

Appendix C
-

Updated Project Plan

................................
................................
................................
....

60





Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

ix


L
IST OF
F
IGURES


Figure 1: Graph showing leading nations supplying gold

................................
................................
.......

2

Figure 2: The research approach to design robots for deep level mining applications

..........................

4

Figure 3: iRobot 210 Negotiator (Ground Robots
-
210 Negotiator, 2010)

................................
..............

8

Figu
re 4: iRobot 510 PackBot (Melanson, 2008)

................................
................................
..................

10

Figure 5: iRobot 710 Warrior (Ground Robots
-

710 Warrior, 2010)

................................
.....................

12

Figure 6: The Big Dog (Dynamics, BigDog
-

The most advanced rough terrain robot on earth, 2009)

.

14

Figure 7: The LS3 Robot Concept (Spectrum, 2010)

................................
................................
.............

15

Figure 8: The RiSE robot (Christensen, 2010)

................................
................................
.......................

16

Figure 9: The RHex robot (RHex: Robotic hexapod, 2011)

................................
................................
...

17

Figure 10: The Remotec ANDROS F6A (Army
-
tech, 2011)

................................
................................
....

18

Figure 11: Samsung Vacuum Cleaner (Hearn, 2010)

................................
................................
............

20

Figure 12: The Petman (Dynamics, Endoskeleton Prototype, 2011)

................................
....................

21

Figure 13: The da Vinci Surgical System (Surgical, 2010)

................................
................................
......

22

Figure 14: ASIMO (Honda's Humanoid R
obots and ASIMO, 2011)

................................
.......................

23

Figure 15: Toyota's Rolling Robot (Toyota, 2003)

................................
................................
.................

24

Figure 16: Lucas, ASORO service robot (ASORO, 2010)

................................
................................
........

25

Figure 17: Vstone Robovie M
-
version 3 (Laboratory, 2011)

................................
................................
.

25

Figure 18: Kyosho Manoi
-
AT01 (Kyosho Manoi
-

AT01
-

Humanoid robot, 2007)

................................
.

26

Figure 19: Water Strider M
icro
-
Robot (Water Strider, 2010)

................................
...............................

27

Figure 20: Groundhog robot

(Mellon C. , 2004)

................................
................................
...................

28

Figure 21: K10 robot

(NASA, 2011)

................................
................................
................................
.......

29

Figure 22: M2 robot

(MIT, 2010)

................................
................................
................................
..........

30

Figure 23: Life Cycle of robot technologies for deep level mining applications

................................
...

31

Figure 24: Possible solutions for future rob
ot platform

................................
................................
.......

44

Figure 25: Study Template

................................
................................
................................
....................

47

Figure 26: List of robots

................................
................................
................................
........................

56










Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

x


L
IST OF
T
ABLES


Table 1: ECSA Exit Level Outcomes

................................
................................
................................
........

iv

Table 2: Capabilities and Limitations
of the iRobot 210 Negotiator

................................
.......................

9

Table 3: Capabilities and Limitations of the iRobot 510 Packbot

................................
.........................

11

Table 4: Capabilities and Limitations of the iRobot 710 Warrior

................................
..........................

13

Table 5: Capabilities and Limitations of the Big Dog

................................
................................
............

14

Table 6: Capabilities and Limitations of

the LS3 robot

................................
................................
.........

15

Table 7: Capabilities and Limitations of the RisE

................................
................................
..................

16

Table 8: Capabilities and Limitations of the RHex

................................
................................
................

17

Table 9: Capabilities and Limitations of the Remotec

................................
................................
..........

19

Table 10: Capabilities and Limitations of the Samsun
g Vacuum Cleaner Robot

................................
..

20

Table 11: Evaluation of robot technologies according to the installation and setup constraints

........

36

Table 12: Evaluation of robot technologies according to the optimization and usage constraints

.....

38

Table 13: Evaluation of robot technologies according to maintenance constraints

............................

40

Table 14: Evaluation of robot technologies according to disposal and redesign

................................
.

41


















Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

xi


G
LOSSARY


Parlor trick:


a simple magic trick which is generally easy to execute

CSIR:


Council for Scientific and Industrial Research

is one of the leading scientific and technology
research, development and implementation organisations in Africa. It undertakes di
rected research
and development for socio
-
economic growth.

COMRO:


Chamber of Mines Research Organisation
of South Africa is a prominent industry
employers' organisation which exists to serve its members and promote their interests in the South
African min
ing industry

Geophysical:

The physics of the earth and its environment, including the physics of fields such as
meteorology, oceanography, and seismology

AziSA:


Architecture

for underground measurement and control networks

WiFi:

a local area network that
uses high frequency radio signals to transmit and receive data over
distances of a few hundred feet; uses
Ethernet

protocol

Autonomous:

Not controlled by others or by outside forces; independent

Lateral motion:

movement of, at, from, on, or toward the
side; sideways

VCI
:

Volatile Corrosion
Inhi
bitor

is a class of chemical compounds which emit a rust
-
inhibiting
vapour

in the air to inhibit corrosion on a metal surface


Chapter 1

:
INTRODUCTION

1.1

Background


Thomas Edison created the electric light bulb and then wrapped an entire industry around it. The
light bulb is most often thought of as his signature invention, but Edison understood that the bulb
was little more than a parlor trick without a system of ele
ctric power generation and transmission to
make it truly useful. So he created that, too. Edison’s approach was an early example of what is now
called “design thinking”


a methodology that imbues the full spectrum of innovation activities.
Innovation is p
owered by a thorough understanding, through direct observation, of what people
desire and need in their lives and what they like or dislike

(Brown, 2008)
.

Technology teams and
groups

(Green, 2009)

are currently designing an innovative robot platform that can mine reefs that
are too narrow for economic exploitation by miners or by current mechanized systems. These
groups together with mining companies helped evaluate and give suggestions for a robot
platform.
This innovative solution can convert in South Africa 22 000 tons of extra gold in currently un
-
mineable narrow reefs, from resources to reserves

(Green, 2009)
.
Currently 40 000 tons of gold is
removed from Witwatersra
nd and mining is extracting 350 tons per year. Thus, the narrow deposit
mining method could create a gold reserve comparable to the Witwatersrand itself. The Chamber of
Mines Research Organization (COMRO) attempted to introduce mechanization, but the techn
ology
at the time limited the success of the outcome. The stopping review

(CSIR, 1988)
] is a
comprehensive evaluation of this work, with parts analysed

(Pogu, 2006)

even more in depth.
Technology ha
s shown tremendous advances in the last 20 years compared to science available
almost half a century ago

(Kao, 2009)
.

Therefore, there are developed technologies that can open
new possibilities if applied in mining. The latest

geophysical tools nowadays allow us to track the
ore
-
body

(Vogt D. V., 2005)
.


South Africa has been the leading supplier globally. The following graph, in Figure 1, shows major
gold producing nations and the amount of gold th
ey have been producing over the past years.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

2



Figure
1
: Graph showing leading nations supplying gold


The graph clearly shows that the amount of gold supplied by the countries, with the exception of
China, is decreasing every year
. This is because the mines are using up reserves that can be easily
obtained through human labour. The ore trapped in narrow reefs and at deeper levels cannot be
extracted and therefore, the mine suffers production losses and eventually has to cease becau
se of
the technology available limits the ability to mine deeper. This does not only affect the mine itself
but the entire country’s economic growth as well as the international gold market.

Thus, it is important to adopt a new generation technology to cre
ate step change increases in
productivity and remove people from the high risk working environments. The technology step
change is also required within the life of the mine timeframe enabling businesses to mine deeper,
safely and profitably. Therefore, it
is no doubt that in order to develop further develop economically
new innovative technologies should be explored to allow for further exploitation of resources and to
help meet this step change challenge to fast forward the development of technology soluti
ons for
the mining industry. .



0
50
100
150
200
250
300
350
400
2002
2003
2004
2005
2006
2007
2008
2009
Gold 000's tons

Year

Gold Supplied Annually

South Africa
China
Australia
Russia
Peru
USA
Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

3



1.2

PURPOSE

OF STUDY


The purpose is of this project was to develop a mining concept for deep, narrow reef level mining
operations. Various robot technologies were evaluated and considered. A design thinking approach
was used to create a
concept.

The evaluation is done to explo
re existing robot technologies to be
used in deep level
mining applications that are

too big a risk for human labour.



1.3

Research Approach


The constraints in this study were derived from the three laws of robotics

(Isaac Asimov's "Three
Laws of Robotics", 2001)

and then incorporated with the life cycle phases of robot technology
discussed later. The three laws include:



Law 1: A robot may not injure a human



Law 2: A robot must obey the orders given to it by a h
uman being as long as it does not
conflict with the 1
st

law



Law 3: A robot must protect its own existence as long as it does not conflict with the 1
st

and
2
nd

laws

Considering these laws as well as all the limitations imposed by the constraints, the concep
tual
design of the ideal robot must be such to
incorporate all these factors.
The method of design
thinking is described which consists of four phases, namely:



Empathize



Explore



Evaluate



Design


As a design engineer it is important to understand the people and environment for which the
innovative robot technology is designed.
Therefore, in order to evaluate and design for deep level
mining companies the engineer must build empathy for
who they are

and
what is important to them
.
As illustrated in
Figure 2
the first phase of this study was to understand the mining environment
better. This was done by contacting specialists in the field and conducting a thorough literature
study. Thereby, it was possi
ble to identify possible robot applications for the mining environment.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

4



In the second phase
, the exploration phase,

various robot technologies were explored for these
applications.

These ranged from humanoids robots, water robots, and military robots to s
pace
robots. No category was established on what type of robots had to be assessed.
The important
aspect was what each could do to contribute in developing the mine robotic concept.


Figure
2
: The research approach to design robot
s for deep level mining applications


Thereafter,

in the evaluation phase
,

various robot technologies could be evaluated
from
questionnaires

by comparing different
functions and the effect of the demands from the mining
environment.
The basic layout of the questionnaire can be found in Appendix A. Some factors, such
as the cost and finance, came forward in the questionnaires, but those are outside the scope of this
study, so it was not considered as
determining

factor.

The

strength an
d weaknesses were also
identified.

Finally, a robot concept could be developed for a specific mining application by combining the
strengths and innovative technologies.
Knowing that key growth imperatives succeed best when
specialized teams share skills,
experience and insight across the silos

(Cash)
, collaboration with
specialists, technology
-

and mining companies was a necessity.

The challenges of this study are that literature on this topic is rather limited. Also, researching

the
constraints of such deep level environments are mainly done on predictions, because of the
uncertainty of what can really be expected at such levels. Also, designing an appropriate robotic
technology from existing technology can become tedious because

of the variety of robots in
Empathize


Mining
Environment


Applications

Explore


Robot
technologies

Evaluate


Compare


Identify
strength and
weaknesses

Design


Combine
strengths


Innovative
technologies

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

5


existence. It is important to understand that this study is a preliminary one and it does not take into
account the finance factor with respect to the concept developed. Only once a thorough
understanding and foundation has bee
n done, this can be regarded as a factor.

Unfortunately, due to the lack of available time and large distances of the relevant mine sites from
the University of Stellenbosch, mines could not be visited. This could have contributed enormously
to increasing
the depth of the study. Also, pictures of the uncertain environment (>5km) discussed
could not be obtained to exactly view what is being dealt with. This is due to mines not having the
appropriate technologies to gain the necessary data. Also, an important

factor that is beyond the
scope of this study is the cost for the actual development and implementation of such technologies.
For now, it is assumed that the funding is limitless just to start with theoretical development of such
technology.















Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

6


Chapter 2

: Robot Technologies


2.1

Introduction


Numerous robot applications exist in mining. However, studying all of them in depth is out of the
scope of this study. The two major applications identified are:



Rock Breaking



Monitoring rock
-
fall hazard risk

This
chapter briefly discusses the need and purpose for using robot technology for these
applications.




2.2

Rock Breaking


Significant work has been done on evaluating the potential of existing rock breaking
technologies for a machine targeting narrow ore bod
ies

(Harper G. , 2008)
. Work to date within
the CSIR has been focused on large machines to remove the narrow seams. The project has
shifted the
mind
-
set

and focus to a robot technology of comparable size to the deposit to be
mi
ned. The use of electric rock breaking has been investigated with specific reference to South
African ore bodies and fits this requirement well. It also has a significant spin off potential in the
form of an electric discharge rock drill (EDD), which will
be the initial developmental focus of the
technology

(Harper J. , 2008)
. Electric rock breaking requires very little thrust force and
therefore is well suited to an autonomous robotic platform. The challenges will be the power

supply and control.

2.3

Monitoring of rock fall hazard risk


An open standard architecture called AziSA for communication of sensor data, and a reference
implementation using that standard

(Vogt D. B., 2008)

has been develope
d.

It is an architecture for
measurement and control networks that can be used to collect, store and facilitate the analysis of
data from challenging underground environments. The architecture was created because the
Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

7


existing identified protocols could not

offer an organized and open architecture for low power, low
-
cost, wireless systems

(Stewart, 2008)
. Innovative robot technologies can therefore make use of this
architecture. The current communication standard of choice is Wi
Fi with its open architecture, high
bandwidth and freely available hardware

(WiFi, 1997)
.

Within the development of AziSA a sonic beacon has been

designed to be used for underground
localization of the sensors

(Ferreria, 2008)
. This sensor will be upgraded to enable robot
technologies to localize (in 3 dimensions) in the stope environment. The sonic beacon is mounted on
the end of a roof bolt and transmits both a 40 kHz ultrasonic signal and a 2.4
GHz radio (EMS) signal
simultaneously. The receiver, mounted on the robot platform, calculates the difference in time of
flight for the two signals. This is used to compute a distance that the platform is from the
transmitting beacon. Triangulation of the
signals from multiple beacons with known position can
allow robot technologies to determine is position accurately
(Green, 2009)
.

Various sensors can be integrated with robot technologies to help monitor the rock fall hazard r
isk.
The range of sensor functions can include micro
-
seismic, acoustic, closure and differential
movement, infrared, support loading, and seismic velocity. Similar technologies than the GOM ATOS
Camera, found in the rapid product development laboratory at
Stellenbosch university, can be also
be fixed on the platform to capture and convert mine layout in the CAD drawings. Different types of
sensors are available for light, motion, temperature and pressure. Infrared sensors have much better
coverage of an are
a. Infrared technologies can be used to monitor gas leak detection, water leak
detection, and pipe and cable detection
(Sewerin Web site. , 2011)
. Sensing of radio activity emitted
can allow for the early detection of the amount of gold mined
(Brink, 2008)


2.4

Types of robots and their applications


After identifying the possible robot applications, different robot had

to be explored. Again, these
robot types range from military, water, and cleaner to space robots. From the list of fifty robots in
Appendix B, only 20 robot technologies are described in detail.


Numerous robotic technologies for deep mining applications
were researched. The initial group of
fifty robots was prioritized with the help of mining specialists and mining companies to twenty robot
technologies. Appendix B shows a list that with all initial robot technologies. It gives a brief
Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

8


description of each

robot as well as the source of reference. The robots selected for this study are
marked with a bullet .The attributes and limitations of each robot were studied and then compared
with the mining constraints. A brief description of the capabilities and the

limitations of each robot
follow with regard to the constraints.


2.4.1

iRobot 210 Negotiator


The
iRobot 210 Negotiator

was initially designed for military purposes. It is used to analyse
potentially dangerous situations from a safe distance, especially those in which is considered too big
a risk for humans to be
in. The robot does surveillance,
reco
nnaissance, bomb identif
ication, and
search and rescue as well as detects hazardous material
. Figure 3

shows the
iRobot 210 Negotiator
.


Figure
3
: iRobot 210 Negotiator

(Ground Robots
-
210 Negotiator, 2010)


Table 2 shows the capabilities and limitations of the robot. The
iRobot

210

Negotiator’s

dimensions
would allow it to fit the shaft. The robot is easy to use and goes into areas where people won’t have
to risk their safety. It has

flipper mobility allowing

it to climb stairs and move over obstacles
. The
video camera ensures
better vision and

the

remote monitoring system allows video and sound heard
to be monitored on a PC.
Integrated joystick is easy to use and significantly reduces on
-
robot training
time
.

The robot has a f
ixed day and night camera system to keep watch behind the robot
-
6 infrared
LED’s providing night illumination

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

9



Capabilities

Limitations

Length of 63.5cm

Width of 53.3cm

H
eight of 19.3cm

W
eight of 15.4kg

G
as monitor detection sensors

Two
-
way audio communication

Water

resistant

I
nternal colour video camera with a 360° pan
and 180° tilt

Video and sound monitoring can only be done if
command station is 150m away


Batteries only last 2
-
3hrs when in operation

Table
2
: Capabilities and Limitations of the iRobot 210 Negotiator

Car charger
-

power inverter
of
300W
is integrated
which keeps
iRobot 210
Negotiator’s

batteries
charged using the car’s batteries.

This robot does not carry any load and is only used to detect
hazardous material.
Fortunately the robot is expandable for multiple accessories.
The robot also h
as
g
auge display of batteries

and v
ariable speed ranges
.

The
iRobot 210
Negotiator

also provide numerous benefits during tactical operations for the SWAT
team

and the during crisis negotiations. Some benefits are:



Substantial operational time savings



Minimized risk during dangerous building/structure breaching tasks



Minimized risk during building/structure search



Reduced team size and/or operations duration



Red
uced liability by gathering better intelligence



Increased response options



Full communications, when and where needed



Ability to deliver small packages or payloads



Better situational awareness



Traceable documentation in real time



Ability to act as an
advance

or first scout



Ability to go up and down stairs



Improved ability to assess

tactical situations


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

10


2.4.2.

iRobot 510 PackB
ot


The
iRobot 510 PackBot

shown in
Figure 4

is used for searches,

reconnaissance as well as bomb
disposal. It has a variety of payloads, sensors and manipulators, and quickly adapts as requirements
changes.

This robot identifies and neutralizes roadside bombs, car bombs and other IED’s by
examining potentially dangerous areas.



Figure
4
: iRobot 510 PackBot

(Melanson, 2008)


The capabilities and limitations of this robot are listed in Table 3. It is mobile over rock and move
through snow, mud and other tough terrain.
The
iRobot 510

Packbot
withstands being thrown out a
window, tumbling down stairs and submersion in water.








Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

11


Capabilities

Limitations


Length = 88.9cm;

Width = 52.1cm;

Height = 17.8m;

Weight = 10.8kg.

Climbs: 60°;

Moves < 9.3 km/hr.

Operator: See environment &

Knows position;
on
-
screen view from multiple high resolution
cameras, 3
-
D graphics showing robot’s
o物敮瑡瑩Wn⸠

L楦瑩ig 捡c慣楴y: 2.27
-
6⸸.g ⡳(慬a
m慮楰u污瑯r⤠ ☠ 4⸵4
-
13⸶.歧k ⠳
-
汩nk
m慮楰u污瑯r⤠

呷o⁷ay⁡畤楯⁣ommun楣i瑩Wn

䍡C敲愠 睩瑨w 污獥l
-
牡ng攠 晩fT楮g ☠
T慹an楧U琯low
-
汩gUW⁶楳ion⁣慰慢楬楴i



Video & sound monitoring limited

Battery: 4 hrs of continuous runtime

Robot arm can only extend to 2m.

The gripper can only hold things the size of
about 10cm.


Table
3
: Capabilities and Limitations of the iRobot 510 Packbot


It is functional in all weather conditions and the operating control unit uses a graphical user interface
that has several highlights which make it easy to use. Additionally
, it has a pre
-
set function to enable
fast positioning of robot
.
The robot also does route clearance and ordinate lift system that digs
around, moves and carries objects. Improvements on arm extension and holding capacity can still be
done.

Manipulators al
low the robot to pick up a variety of objects and allows for precise targeting
and placement of tools
. It is e
quipped with eight payloads that have high

quality electronics
including E
thernet, USB, power and two video channels
.

It also protects bomb squad
technicians and ground troops from danger and defeat a variety of
ordnance and SWAT threats. It is equipped

with powerful cameras and

the vision and targeting head
en
sures that the operator can see
and overcome an
ything in the robot’s path. The
included
Quick
Clamp

Accessory Mount allows for
fast installation an
d removal of a variety of tools
at the head
.

The
iRobot 510

Packbot

also has a variety of software packages that can be used to increase its
ability to perform tasks. For deep level mining this dif
ferent software packages can be developed for
different purposes and to increase in functionality.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

12


2.4.3

iRobot 710 Warrior


The
iRobot 710 Warrior

is designed for operation in dangerous environments and detection of
chemicals. More than 3500 of these rob
ots shown in Figure
5

have been delivered to military and
defence stations.

It has been described as a
‘multi
-
mission, modular robot with superior power


(iRobot, 2010)
.


Figure
5
: iRobot 710 Warrior

(Ground Robots
-

710 Warrior, 2010)


The
iRobot 710
Warrior

easily moves over tough terrain

and
is
suitable for indoor as well as outdoor
use. It has a wireless range of only 800 meters and a limited battery life as illustrate
d in Table 4.







Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

13


Capabilities

Limitations


Climbs stairs & slopes of 45° inclines

Wireless connections

Robot’s 2
-
汩n欠U敡vy
-
lift manipulator’s arm has
慮⁥硴敮獩潮eo映192⸲⁣m

L楦琠io慤猠s100 歧
-
瑵牲整s

佢獴慣汥⁡ o楤慮捥c獥湳or猠

䙩敬e
-
Ⱐ 獵獴慩nm敮e
-

慮T T数e琠 m慩aW敮慮ce
捡c慢楬楴i




Wireless range < 800m

The time in operation limited due to battery
life.




Table
4
: Capabilities and Limitations of the iRobot 710 Warrior


The
iRobot 710
Warrior

allows getting real
-
time intelligence and completing situational awareness,
and can be quickly deployed from standard vehicles. It also moves heavy payloads. It facilitates easy
configuration and training and it maintains high mobility on tough terrain in

urban environments
while being functional in all weather conditions.


The 2
-
link heavy lift manipulator has the following features:



Arm Extension:
192.2 cm



Arm Weight:

54 kg



Lift Capacity:
100 kg (
maximum lift
)



Turret camera



Gripper cameras



Quick disc
onnect from chassis



Quick disconnect gripper



4 payload ports providing power, Ethernet and video


2.4.4

The Big

Dog


This robot shown in Figure
6

was specifically designed to go anywhe
re a human or animal can go
(Dynamics, BigDog
-

The most advanced rough terrain robot on earth, 2009)
. This robot can walk, run
and climb in rough terrains while carrying heavy loads.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

14



Figure
6
: The Big Dog

(Dynamics, BigDog
-

The most advanced rough terrain robot on earth, 2009)


The
Big Dog

can move through a variety of terrains, which include rubble, mud, snow and shallow
water. Various sensors allow this robot to control a variety of movements as listed in Table 5.

Capabilities

Limitations


Multi
-
jointed legs: Absorbs shock & recycle
Ener
gy: Step
-
by
-
step movement

On
-
board computer: Controls variety of
sensors, locomotion & balance the legs

Locomotion sensors: Control joint positioning,
joint force, ground contact, ground contact &
Additional sensors: measures robots internal
state

Move:

6.44km/h & travels 20.6km without
stopping or refuelling.

Carrying capacity=154.2kg; Length=91.44cm;
Tall=76.2cm

Temperature range of hydraulic seals in legs is
between
-
70°C and 260°C



Maximum range=12 miles if it doesn’t carry
m慸amum⁰慹 o慤

䍡Cno琠 p楣i up 汯慤s or T楧 睩瑨w慢獥湣攠of
捡ce牡猠

No⁶楳i慬⁣慰慢楬楴i

啳攠g慳o汩n攠fo爠晵敬



Table
5
: Capabilities and Limitations of the Big Dog

The
Big Dog

is not operational at night. Only follows person using laser range under limited
conditions and doesn’t respond to verbal commands. The gasoline may cause ventilation problems
in deep level mining.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

15


2.4.5

The LS3
-
Legged Squad Support System


This is an upg
raded and improved version on the
Big Dog

previously described. It is designed to be
able to go everywhere soldiers and marines can go.

Figure 7

shows the
LS
3 robot.



Figure
7
: The LS3 Robot Concept

(Spectrum, 2010)

The
LS3
carries maximum load while at maximum speed.
In contrast with the
Big Dog
the
LS3

is
operational at in day and night and, responds to verbal commands.

Table 6 gives the capabilities and
limitations of the
LS3
.


Capabilities

L
imitations


T
ravels 32.2km without refuelling
lasting

24hrs

U
ses computer vision, sensors o
r GPS to follow
leader/operator


It has ability to walk, trot, run and wade
through water Its stereo vision system and
laser range finder system allows it to sense
environment.




GPS signals are poor in certain environments
e.g.: forests, jungles and cities with tall buildings.
C
annot pick up loads or dig.

The responds to verbal
commands is

limited.

Runs

on a gasoline power engine.


Table
6
: Capabilities and Limitations of the LS3 robot

This robot is still in development stage and has limited mobility at
night.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

16


2.4.6

RiS
E
-

The Amazing Climbing Robot


The design goal of the
RiSE

shown in Figure
8

was to create a bio
-
inspired robot that can walk on
land and climb surfaces.



Figure
8
: The RiSE robot

(Christen
sen, 2010)


This robot can be used for surveillance and monitoring operations in deep mining. Its size and ability
allows it to be useful for entering surfaces difficult to reach as listed in Table 7.


Capabilities

Limitations


Climb vertical terrain

Changing posture: conform to surface

Communicates: Operator commands



Needs charging station (Recharge itself)

Slow movement


Table
7
: Capabilities and Limitations of the RisE


Dimensions are suited
for purpose used in deep level mining. The robot’s speed needs to be
improved for faster responds and feedback.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

17


2.4.7

RHex


This is an autonomous hexapod robot shown in Figure
9
, with compliant legs and only one actuator
per leg. It is currently the only
robot that can perform such a wide variety of activities as a single
autonomous robot.



Figure
9
: The RHex robot

(RHex: Robotic hexapod, 2011)

The

Rhex’s

use of legs instead of wheels or tracks allows for a variety of behaviours. Passive
compliance in the legs overcomes limitations of under
-
actuation and helps simplify mechanical
design, yielding robustness and sprawled posture creates passive stabilizatio
n of lateral motion. The
capabilities and limitations of this robot technology are listed in Table 8.


Capabilities

Limitations


Climbs: stairs & slopes up to 45 degrees

Run for 45 minutes (covering < 4.8km)

Moves over rough terrain

Autonomously follow

a line on the ground
without any operator control

Performs simultaneous localization & mapping
by using artificial landmarks scattered over
natural terrain



Remote control <150m (distance)


Table
8
: Capabilities and Limitations of the RHex


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

18


To recover nominal body orientation the robot can flip itself over. The flexibility of its legged design
leaves significant room for additional behaviours. Its high mobility over natural terrain opens up new
poss
ibilities for specific application domains for which components to achieve autonomy are still in
their infancy. Commercialization of the
RHex

platform requires significant platform development as
well as further behavioural research to improve its performa
nce


2.4.8

The Remotec ANDROS F6A


This robot is viewed as the most versatile heavy duty robot available. One of the best heavy duty
robots on which responders worldwide rely to help assure a safe, successful outcome for their most
challenging missions. Th
e
Remotec

ANDROS F6A

is shown in
Figure 10
.



Figure
10
: The Remotec ANDROS F6A

(Army
-
tech, 2011)


The capabilities and limitations of this robot technology are listed in Table
9
. The
Remotec ANDROS

F6A

can move in wet and dry conditions because it has a sealed weather resistant enclosure. Also,
for precision the manipulator’s speed can be varied.




Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

19


Capabilities

Limitations


Height=143.5cm; Width=73.7cm; Length=
132.1cm; Weight=220kg

Joystick: Easy navigation, video
-

& audio
output, heavy duty, portable & water
resistant.

2
-
way audio system (weatherproof speaker &
microphone).

Colour camera: Low
-
light switching

Extra
-
low
-
light colour pan/tilt/zoom (full 360°
continuous
-
pan 180° tilt
).

Gripper & continuous rotate 61cm camera
extender.

Multiple
-
mission tool & sensors.

Automatic arm positioning.



Manipulator arm: Only extend up to 61cm

Wireless & radio control only functional to
distances < 1.9km.



Table
9
: Capabilities and Limitations of the Remotec


This robot has potential for a variety of applications, but is currently only used in the military.


2.4.9

Samsung Tango Vacuum Cleaner Robots


This robotic system
, as shown in Figure 11
,

moves through a ro
om sucking dust from any surface it
comes in contact with. It runs on a battery.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

20


Figure
11
: Samsung Vacuum Cleaner

(Hearn, 2010)

This robot has low noise levels without any high frequency elements usu
ally generated by the motor
and the brush. Table 10 shows the capabilities of this robot.


Capabilities

Limitations


Has
13 sensors allowing interaction with
environment


Some are crash sensors which detects furniture
and feet 2cm away

Uses remote control

for operator ease


H
as a daily scheduling option


Remote control has a limited range in how far it
can work


Operation time is only 2hrs


Table
10
: Capabilities and Limitations of the Samsung Vacuum Cleaner Robot


The various sen
sors integrated in this robot allow it to clean with precision and find its way back to
its docking station.


2.4.10

Petman (The Protection Ensemble Test Mannequin)


This robot shown in Figure
12

is an anthropomorphic robot designed to test the chemical pr
otection
clothing in the military. Walking robots based on passive
-
dynamic principles can have human
-
like
efficiency and actuation requirements. However, movements are mostly in sagittal plane and in
straight line, being extremely difficult to turn, go bac
k, seat. The motion is mostly symmetrical. A
sequence of tests and performance measures must be done to decide on the feasibility of their use
in the mining discipline. Therefore, time studies and continuous improvements must be done for
such implementatio
n.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

21



Figure
12
: The Petman

(Dynamics, Endoskeleton Prototype, 2011)


Petman

can balance itself and move freely about the environment. Simulate body physiology by
controlling temperature, humidity and sweating. This humanoid has structure of average human and
can do simple activities such as walk and crawl. However, this robot is
still in the development stage.


2.4.11

da Vinci Surgical System


The high
-
quality intensity of vision, precision, handiness and control of this system, surgeons will be
able to perform the most delicate and very complex operations that require small incis
ions.
The
da
Vinci

Surgical System

is powered by state
-
of
-
the
-
art robotic technology that allows the surgeon’s
hand movements to be scaled, filtered and translated into precise movements of the
EndoWrist

instruments working inside the patient’s body

(Surgical, 2010)
.
Figure 13

shows the different
components that make up this system.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

22



Figure
13
: The da Vinci Surgical System

(Surgical, 2010)


The system consists of the followi
ng components:



Surgeon Console

that provides a high definition 3D image of the inside of the patient’s body
while the surgeon sits at a console and the surgeon’s hand

,wrist and finger movements are
translated into precise real time movements



Patient
-
side
chart

is where the patient is positioned during the surgery



Robotic arms

which carries out the surgeon’s commands



Vision system
equipped with high
-
definition, 3D endoscope and image processing
equipment that provides true
-
to
-
life images



EndoWrist instrume
nts
that have seven degrees of freedom with each one having a specific
surgical mission.









Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

23


2.4.12

ASIMO, the walking r
obot


This robot can walk and is typically used to assist the elderly people. Walks on two legs and can use
its hands
f
or a variety
of tasks. Figure 14

shows a picture of
ASIMO
. It is 130cm in height and weighs
5
4
kg.

At this height,
ASIMO

can effectively use switches, tables, and other household products, as
well as look at a human sitting down at eye level.

ASIMO

is composed of

both o
ld and new
technologies and c
oordination allows
ASIMO

to turn while walking or running without pausing.


Figure
14
: ASIMO

(Honda's Humanoid Robots and ASIMO, 2011)


Honda
developed

ASIMO

for household use and therefore

the ability to recognize specific people
and interact with them needed to be addressed. Developers implemented software that allows
ASIMO

to recognize moving objects, postures, gestures, and other environmental stimuli, as
well as
distinguish the source of sounds and distinguish specific faces.
ASIMO

can also provide users with
information from the Internet via its wireless connectivity

(Honda's Humanoid Robots and ASIMO,
2011)
.





Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

24


2.4.13

Rolling Robot


This is the ‘rolling type’, as shown in Figure 16,

of the original walking model of the humanoid
robots, which means it moves quicker. It is 100cm tall and weighs 35kg.
It c
an maintain balance on
one axle.


Figure
15
: Toyota's Rolling Robot

(Toyota, 2003)


The rolling model zooms along quickly without taking up much space.
This robot also uses its hands
to do a variety of tasks. It can be used in manufacturing and mobility


2.4.14

Lucas, ASORO Service Robot


The robot shown in Figure 16

is a platform robot that can be used to do tasks like information
stands/kiosks, butler, tele
-
presence, patrol and tele
-
operation. This robot can perform breathing
expression and is easy to configure

and operate.

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

25



Figure
16
: Lucas, ASORO service robot

(ASORO, 2010)


Lucas

is 1.5m high and weights 60kg.

It has 8 microphones at the head for 3D sound localisation and
4 microphones on chest for speec
h enhancement.

The robot is made of soft fabric skin and has an
acrylic outer structure and aluminium for its internal structure.


2.4.15

Vstone Robovie M
-
Version3


Robot designed to play soccer. Thus, main design purpose was for entertainment.

Figure 17

shows a
picture of this robot.


Figure
17
: Vstone Robovie M
-
version 3

(Laboratory, 2011)


This robot can throw objects, punch, somersault, handstands, run, and side step and play soccer. It
can detect when it fell and can automatically lift itself to a standing position. Operations
are
controlled through remote control and PC and responds time
is relatively fast.

However,
t
his robot
Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

26


does not have digital camera eyes and cannot recognise words or talk. Walking time on single charge
is only 25min which is very poor.


2.4.16

Kyosho
-
Ma
noi
-
AT01


Figure 1
8

shows the
Kyosho Manoi

AT01
. This robot exec
utes a variety of smooth and natural
commands. It responds to instantaneously to commands given. However, its main design purpose is
for experiments.


Figure
18
: Kyosho Manoi
-
AT01

(Kyosho Manoi
-

AT01
-

Human
oid robot, 2007)

Kyosho

Manoi AT01

c
an per
form a wide range of movements such as w
alk, run, step climbing, back
-

an
d front somersaults and dancing.
Additional sensors can be integrated

for different purposes. The
s
peed of moving between two positions
can be set with a click

on the PC or remote control.

To map

the
movements, draw lines connecting boxes using a

mouse. E
ach motion of
a motor is
represented by a box.







Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

27


2.4.17

Water Strider


The

Water S
trider

robot in Figure 19

is a miniature robot. I
ts development goal was to develop a
micro
-
robot that takes advantage of the surface tension of water to stay and manoeuvre on water
with power efficiency and agility

(Water Strider, 2010)
.


Figure
19
: Water Strider Micro
-
Robot

(Water Strider, 2010)


M
icro
-
actuators and DC Motors

are used

to simulate
the
Water S
trider

movements

and to
understand the physics of water striders to model their characteristics of floating on th
e
surface of
water. The various materials and leg shapes are studied to
improve the ability of the robot to float

on water and manoeuvre.
Water S
trider

robots
are

small and relatively efficient. Because it is on the
surface of water and light, the robot wi
ll be highly agile and can reach inaccessible areas for many
different applications.


2.4.18

The Groundhog


The
Groundhog

is a remotely controlled mobile robot

that has a set of integrated algorithms for
estimating global correspondence and aligning robot
paths

t
o build

consistent maps of large mines.

This algorithm enables
the

recover
y of

consistent maps several hundreds of meters in diameter,
without odometric information
. Figure 20

show
s the
G
roundhog

robot.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

28



Figure
20
:
Groundhog robot

(Mellon C. , 2004)


This robot can go into environments too toxic for people to enter in order to provide mapping of the
environment. The
Groundhog

also has limited autonomous ability and has laser range sensors,

a
night vision camera, gas detectors, sink
-
age sensors, and a gyroscope.


2.4.19

K10 Black and K10 Red Moon robots


These robots are being tested for future moon landings in Canada. Scientists chose the polar region
because of the extreme environmental co
nditions, lack of infrastructure and resources, and geologic
features. They carry 3 D Laser scanners and ground penetrating radar. Loaded up with
GPS, stereo
cameras, l
aser scanners and sun trackers,

the
K10 Black

and
K10 Red

can laser map terrain over
lar
ge
distances

and fire radar into the ground and detect features up to
5m down
.
Figure 21

shows a
typical picture of the K10 series

(NASA, 2011)
.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

29



Figure
21
: K10 robot

(NASA, 2011)


K10 Black

and
K10 Red

are using a mix of information previously obtained by aerial and satellite
imaging and data that the robot survey team is gathering. The robots weights 80kg, length of 1.1m,
width of 0.9m, height of 1.3m, and speed

of 0.9m/s.


2.4.20

M2 Robot


The
M2 robot
, shown in Figure 22
,

is a 3D

bipedal robot, still being developed in MIT leg laboratory.
It has algorithms for walking and motion description as well as control techniques. The robot has 12
degrees of freedom and is also equipped with force control actuation techniques and automatic

learning techniques.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

30



Figure
22
: M2 robot

(MIT, 2010)


The design goals of this robot include:




Walk fast (1.0 m/s
)



Walk efficiently



Be reliable



Have a large margin of stability and be robust to small disturbances



Be confident looking



Become
-

a robot which can be reliably used to perform experiments without breaking


2.5

Conclusion


The different robot technologies were assessed with respect to

their strengths and weaknesses.
These help identify their capabilities and limitations for deep level mining applications. This
information were used and incorporated with the constraints of deep level mining to evaluate each
robot technology.

Chapter 3
g
ives information about the life cycle of robots.


Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

31


Chapter 3

:

Life Cycle of Robots

3.1

Introduction


This chapter describes the typical life cycle of robot technologies in order to assess the constraints
associated with deep level mining. The four stages in the life
cycle include:



Installation and Setup



Optimization and Usage



Maintenance



Disposal and Redesign

Each stage of the life cycle has constraints that limit the functionality of the robot technology.
Therefore, it is important to know and understand the constrai
nts to cater for the needs during that
stage.


3.2

Challenges
-

Life Cycle


Various constraints were evaluated on the basis of the environment as
sociated with deep level
mining, deeper than 5 km
. Although this type of environment is relatively new and
under
studied, the analysis done is comparable to what may be expected. The typical life cycle
for robot technologies in deep level mining applications are illustrated in

Figure 23
. Each
constraint is assessed under the different phases of this life cycle.


Figure
23
: Life Cycle of robot technologies for deep level mining applications


Installation
and Setup

Optimization
and Usage

Maintenance

Disposal and
redesign

Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

32


Many

of the installation and setup constraints are identified during this study. It was also important
to consider the deep level mining factors that l
imit the usage and maintenance of the technology.
The disposal and redesign phase received proper attention before the eva
luation phase of this
research.


3.2.1

Installation and Setup


Robots need to have a geotechnical stability to be operationally funct
ional in deep mining
environments. This applies that robots must be able to move with ease in such environment
described as being high siliceous rock environment. As a result of the high uncertainty of deep
mining environment it is vital that robots be wel
l equipped with slip resistant finishes to avoid fall
and possible damage. With increase of depth, there is an increase in temperature and humidity.
Typically the temperature for 5km deep mining is 70°C and pressure level is 920 times normal
atmospheric pr
essure. Therefore, the robot technology must be corrosion resistant, electrically
insulated and in additional to ventilation through the shaft, have a self
-
cooling mechanism where
necessary.
Ventilation is an important aspect within such deep level mining
environment, especially
since many robot technologies use gasoline as a fuel.
The size (dimensions) of the robot must ensure
that the robots will be able to fit into most mining shafts. Currently the shaft allows widths of not
than 3m. The robots must also

be able to be positioned (installed) so that they perform operations
continuously, with minimal interference.


3.2.2

Optimization and Usage


In the deep level mining environment rock falls and backfill effects occur during the mining
operation. Therefore, the robot must be robust and should not have any fragile components. Robots
must be able to communicate and be equipped with sensors to be us
ed for various applications. The
sensors can help management to monitor
rock fall

hazard risk and fragmentation. Ultimately, the
control unit and specialist team can use this feedback (data) in a 3D
-
virtual room on ground level to
make strategic and operat
ional decisions. Robot technology must also have geological stability
especially with slip resistance finishes to operate in the high siliceous environment. The technology
will also consume a considerable quantity of power, especially if electric discharg
e is the chosen rock
breaking method. Un
-
tethered autonomous operation is
favoured

for logistical reasons in cable
Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

33


management, but the ability of the machine to store its own energy may be problematic.
Developments in wireless electricity by wiTricity

(Wireless Electricity Delivered Over Distance. , 200
-
2011)

may also hold the answer. However, it is also hoped that the progress made in battery
technology will supply a power pack suitable for the robot technology. Currently, many

robot
technologies are powered by petrol

(Petrol Engines. , 2011)

engines, which unlock more
maintenance challenges as discussed below.

Robots must have grippers that will be able to load and
unload ore bodies and waste produc
ts.


3.2.3

Maintenance


It will be difficult to inspect and recover (maintain) robots mining at such deep levels. Therefore,
they must have a long operation life capability and be flexible to protect its own existence. Robots
can be developed from self
healing materials like shape memory alloys and shape memory polymers.
These materials allow recovery when deformation is caused by temperature and stress applied.
Epoxy
-
based anti
-
corrosion coatings can be used for healing the damage by autonomic means, or

mechanically by sealing with corrosion products. Protecting metal substrates with coatings through
electrochemical and/or inhibition mechanisms can also be consider
ed

(Self
-
healing Materials


Fundamentals, Design Strategies, and A
pplications., 2009)

. The robot joints may require proper
lubrication of gear oil or grease. Screw terminals can loosen from vibration and therefore tightening
maintenance of wires should be performed. The concept of self
-
configurable robots can be ap
plied

(Scribd Web Site. [Online], 2011)
.


3.2.4

Disposal and Redesign


The robot technology platform must allow for new innovative improvements and upgrading as
needs change without losing other essential operational qualities.

The materials of the robot
technology should also be recyclable.



Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

34



3.3

Conclusion


The life cycle helped to identify the constraints for each of the life cycle stages of the robot
technology. This approach assisted in narrowing down the functionalities
that the robot technology
concept must have.

The evaluation process is discussed next

in Chapter 4
.

















Final Year Project 498

2011


Stellenbosch University: Department of Industrial
Engineering

35


Chapter 4

:

EVALUATION

OF DIFFERENT ROBOT TECHNOLOGIES


4.1

Introduction


This chapter describes the capabilities and performance of the robot tec
hnologies described in
Chapter 2

in context of the typical life cycle of robot technologies.

Each life cycle stage has different
requirements and some may have the same. How the robot technologies perform in each aspect
is

ranked from very bad to very good
.



4.2

Evaluation Process


As outlined in section two, the different robot technologies will be evaluated through a four step lift
cycle. The four steps outlined are: installation and setup, optimization and usage, maintenance and
disposal. The first sec
tion of the life cycle of a robot that was identified was the installation and
setup. Installation and setup is important to deep level mining activities as this environment has a
high level of uncertainty and robots must be able to adapt to these constrai
nts.

Only
15 of the 20
described robots were analysed in the decision matrices because the
y

prove to be more applicable
to the study.

The ratings are as follows:



Very Good

=

5



Good


=

4



Average

=

3



Bad


=

2



Very Bad

=

1








Final Year Project 498