SWAT Reconnaissance Vehicle Final Project Specification

thunderclingΤεχνίτη Νοημοσύνη και Ρομποτική

13 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

189 εμφανίσεις

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10

Page
1

of
78







SWAT Reconnaissance Vehicle

Final Project Specification





Group 4



Senior Design

ECE 441



Written

by


Tobin Lindstrom
, ECE

Chris McR
eynolds
, ECE

Cosmin Cazan
, ECE

Michael Waldstein
, CS

Evan Miles
, ME

Zachary Southworth
, ME

Matthew Nikkel
, ME





ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
2

of
78


Oregon State University


Revision History

Revision
Number

Date

Description

Contributor

1

15 OCT 07

Initial Document Creation

Tobin Lindstrom

2

27 OCT 07

General Revisions

Tobin Lindstrom

3

27 OCT 07

Block Diagram

Chris McR
eynolds

4

27 OCT 07

Introduction

ME Group

5

27 OCT 07

Customer Requirements

ME Group

6

27 OCT 07

Implementation

Approaches

Cosmin Cazan

7

31 OCT 07

Interface Definition

Chris McReynolds

8

31 OCT 07

ME Design Considerations

ME Group

9

28 NOV 07

General Revisions

Tobin
Lindstrom

10

30 NOV 07

Revised Block Diagram

Chris McReynolds

11

02 DEC 07

Testing Plan

Cosmin Cazan

12

03 DEC 07

Added Sections 5,7 and 8

EE Group

13

04 DEC 07

Final Revisions Made

EE Group














































ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
3

of
78



DISCLAIMER


This report was prepared by students as part of a college course requirement. While considerable effort
has been put into the project, it is not the work of a licensed engineer and has not undergone the
extensive verification that is common in

the profession. The

information, data, conclusions

and content
of this report should not be relied on or utilized without thorough, independent testing and verification.
University faculty members may have been associated with this project as advisors,
sponsors, or course
instructors, but as such they are not responsible for the accuracy of results or conclusions.



Please note that the ME members of the design team have their own version of the
Final Project
Specification

in which they will be turning in separately. The ME version of the paper excludes any
ECE and CS content for ease of gr
ading by the reviewing parties.

Due to the creation of two different
Final Specification Papers this paper covers in detail the electr
ical aspects of the project and only
briefly discusses the mechanical por
tions. For detailed mechanical specification please review the ME
Final Project Specification.

Any mechanical drawings, the bill of materials for mechanical components
and mechanica
l specifications are included in the ME Final Project Specification.


























ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
4

of
78


Table of Contents


2.
I
ntroduction
………………………………………………………………………………………………………
7


2.1
Customer Requirements……............................
..........................
....
..............
..
.
...
...........
.............
..........7


2.2
Competitive Analysis
…………………………………………………………………
………..
…….10


2.2.1
Product Space Analy
si
s....................................
..........
..............
............
.....
.............
..............10


2.3 Feature Set……………………………………………………
…...
………
.


………...
………….15


2.3.1 Minimum Requirements………
…..

….
……………………
…...

……….
……………16


2.3.2
Target Feature Set...
.
.
..
..........................................................
..............
.......
.............
............
..17


3. Architectural Overview………………………………………………………
………………
………...
………17



3.1 Implementation Approaches…………………………………………
………..........
..............
.............18




3.1.1 Microcontrollers…………………………………………………………
……….
……...
...18






3.1.2 Video/Audio
Transmission………………………………………………
……….
………
..20




3.1.3 Wireless Communication………………………………………………
……….
…………
21




4. Top Level Description…………………………………………………………
………
……
………..
………..
2
8



4.1 Top Level Block Diagram………………………………………………
………………
……….
.….31




4.1.1 Top Level
Interface Definition……………………………………
………
………
………32





4.1.2 E
nvironment………………………………………………………...
.
........
............
............33


5 Functional Unit Descriptions…………………………………………………………....
...............
...............
.....34




5.1 Servo Controller Block
Diagram……………………………………………
……………
………..
...34




5.1.1 Servo Controller

Interface Definition…………………………

………
………
………..35




5.1.2 Servo Controller

Operation……………………………………

……………
……….
….35


5.2 Battery Charger

Block

Diagram……………………………………………
…………
……….
…….36




5.2.1 Battery

Charger

Interface Definition…………………………
……………
………
...
……
36




5.2.2 Battery Charger

Operation……………………………………
……
……
……….
………..36



5.3 Battery System Block Diagram……………………………………………
………
……….
………..37




5.3.1 Battery System

Interface Definition…………………………

………
……….
…………37


ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
5

of
78




5.3.2 Battery System

Operation……………………………………
……
………
……….
……...37


5.4 Power Regulator Block Diagram……………………………………………
……
………..
………..
.
38




5.4.1 Power Regulator

Interface Definition…………………………
………
……….
……..
….
.
.
38




5.4.2 Power Regulator

Operation…………………………………
………
……….
……...

….
.38



5.5 Central Processor Block Diagram…………………………………………
……
………..
…………..39




5.5.1 Central Processor

Interface Definition…………………………
………
……….
………

40




5.5.2 Central Processor

Operation…………………………………
…………
………
…...

….
40


5.6 Drive Camera Block Diagram…………………
………………………
…………
………..
……
……
41




5.6.1 Drive Camera

Interface Definition…………………………

………
……….
…….
……
41




5.6.2 Drive Camera

Operation……………………………………
……
…………
……….
….

41



5.7 Drive Motor Controller Block Diagram………………………………
……………..
.............
.
……..
42




5.7.1 Drive Motor

Controller

Interface Definition……………………
………
………
…..

.

42




5.7.2 Drive Motor Controller

Operation………………………………
………
……….

……..
42


5.8 Flipper Motor Controller

Block Diagram……………………………………
…………
………..
…..
43




5.8.1 Flipper Motor Controller

Interface Definition……………………
………
……….

……
43




5.8.2 Flipper Motor Controller

Operation……………………………
………..

……….
……..
43


5.9

Controller Interface

Block Diagram………………………………………………

……….
……...44




5.9
.1 Controller Interface

Interface Definition…………………………

………
……….
..
…..
44




5.9
.2 Controller Interface

Operation……………………………………
………
……….
..
……..
45



5.10

Main Camera

Block Diagram…………………………………………………
……
………..
….
….
45




5.10
.1 Main Camera

Interface Definition…………………………
…………
………
……….

45




5.10
.2 Main Camera

Operation……………………………………
………
………

……
……
46


5.11

Ethernet Bridge

Block Diagram…………………………………………………
……...
.............

46




5.11
.1 Ethernet Bridge

Interface Definition…………………………
………
……..

……….
.
46




5.11
.2 Ethernet Bridge

Operation……………………………………
………
……
………
..
…..
47



5.12

Lighting

Block Diagram…………………………………………………………

……….
…..

47


ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
6

of
78




5.12
.1 Lighting

Interface Definition…………………………
……………
………
………
…….
47




5.12
.2 Lighting

Operation……………………………………
……………
……
……….
...
…….
48


5.13

X&Y Servos

Block Diagram……………………………………………………
………..

………49




5.13
.1 X&Y
Servos

Interface Definition…………………………
……
……
……….

………..
49




5.13
.2 X&Y Servos

Operation……………………………………
……
……
……….
.
…………
49


5.14 Arm Servos

Block Diagram…………………………………………………
…...
………
………...
.
50




5.14.1 Arm Servos Interface Definition…………………………
…………
….….

……….
….
50




5.14.2 Arm Servos Operation
……………………………………………………
……….
……...50


5.15 Drive/Flipper Motors

Block Diagram……………………………………………

………..

…..
51




5.15.1 Drive/Flipper Motors Interface Definition…………………………

……….


……
51




5.15.2 Drive/Flipper Motors Operation……………………………………
………
………
……51


7. Testing………………………………………………………………………………………………………….52



7.1 Functional Test……………………………………………………………………………………….52



7.2 System Test…………………………………………………………………………………………..53



7.3 Man
ufacturing Test…………………………………………………………………………………..53



7.4 Safety
Test……………………………………………………………………………………………54



7.5 Reliability Test……………………………………………………………………………………….54


8. Bill of Materials……………………………………………………………………………………………….55


Appendix A: References
……………………………………………………………
………………
…………….57


Appendix B: Glossary
………………………………………………………………
………………
…………….60


Appendix C
: Competitive Analysis Table
………………………………………
.....................
.........................
.....
61


Appendix D:
Schematics..
…………………………………………………………
……
…………….
………….
68


Appendix E: Psuedo Code…………………………………………………………
…………….

…………….
73






ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10

Page
7

of
78


2.
Introduction


Most urban Police Departments have formed special teams

called Special Weapons and Ta
ctics
(SWAT) teams

to handle pa
rticularly dangerous situations
. These teams primarily
handle two types of
situations:

dynamic and static. While dynamic situations can be very dangerous, the element of
surprise is u
sually in the SWAT team’s favor.


This is because the SWAT

team is operating without the
knowledge o
f the problematic individual(s)
. F
or example, a surprise a
rrest
of a
arsonist

is a dynamic
situation. Static situations are typically some type of public threat, and the SWAT team must instead
respond and remove the danger to reestablish order.


In both of these situations
,
the

main concern of the SWAT
team
o
per
ator is to maintain the safety of
the
entire unit.
Dynamic situations typically involve less risk due to choice of time and location, combined
with the element of surprise; h
owever, static situations are usually unanticipated and at uncontrolled
locations, thus involving
greater

danger and risk. In order to minimize the risk, remotely
-
controlled
robots have been utilized by SWAT teams to
clear rooms. Clearing a room gives

th
e SWAT o
perator an
understanding of where threats may be and what areas are known to be sa
fe. Ordinarily, the

SWAT
team
o
perator
is
located at a forward, mobile post to
allow better control over the movements of the
SWAT team, therefore
the data return
ed

from the robot must not limit
the SWAT team’s
movement.


Available c
ommercial

robotic

systems that are available for
this purpose are very expensive

and costly
to maintain, which
motivated

the
Salem Police Department (PD) to ask

the OSU
Senior Design Group

to design and build a prototype robot as a senior project. The current senior design project

will create a

proof
-
of
-
concept

model
that will meet the robot’s essential requirements

as given by the Salem PD
.
The results of this year’s senior project will
be developed further in future years. For 2007
-
2008, the
student team allocated to this project includes three
Mechanical Engineering (
ME
)

students, three

Electrical and Computer Engineering

(
ECE
)

students

and one

Computer Science

(
CS
)

student.


Upon compl
etion of this year’s project, students will have developed a strong interdisciplinary
understanding of product developmen
t and project management. S
tudents will have carried out the
design, manufacturing

and testing of a prototype robot capable of wireles
s operation, including
navigating obstacles, in a variety of environments. In addition, the Salem PD will receive a proto
type
robot to evaluate, allowing them to refine

their assessments of need for following senior projects, at
very low cost. The comple
ted robot may or may not be used in crucial situations where it could
remove significant risk from police officers

however future
students’

efforts should
satiate the robotic
needs of the Salem PD
.



2.1


Customer Requirements & Product Background

Statements

of Need


The following Statements of Need were the design goals directly identified by the Salem Police
Department SWAT Team:


Must have
:

Ability to navigate stairs or debris inside building

Broadcast audio and video back to a base station

Deliver and carry items (possibly
SWAT’s

existing wireless video devices)

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
8

of
78


Would be very helpful
:

Ability to right itself if knocked over

Ability to open interior doors or closed containers

Wireless operation with extended battery life

Low
-
light or no
-
light

video


Nice to have
:

Multiple cameras or camera views

Articulating camera

Water resistant or waterproof

Two
-
way audio capability

External lighting

Customer Requirements


Since this project is to take place as a proof
-
of
-
concept for future development, an
d due to the fact that
the design team is composed of students, the team and its advisors suggested a refined scope of work.
Along this approach, at the first meeting with the customer, the team sought to understand the design
priorities for the prototype
. Each

item in the

Statement of Need was discussed to understand the
reasoning for each need, and
to
determine what design areas should take top priority. Table 1 below
shows the results of the discussion.


Table
1
: Results from Sta
tement of Need Discussion

Statement of Need

Current
Status

Original
Status

Ability to navigate stairs or debris inside building

In Scope

In Scope

Broadcast audio/video back to base station

In Scope

In Scope

Deliver/carry items (possibly SWAT existing
wireless video device)

Secondary

In Scope

Ability to right itself if knocked over

Drop

Secondary

Ability to open interior doors or closed containers

Secondary

Secondary

Wireless operation with extended battery life

In Scope

Secondary

Low light/no ligh
t video

In Scope

Secondary

Multiple camera/camera views

Drop

Tertiary

Articulating Camera

In Scope

Tertiary

Water resistance/waterproof

In Scope

Tertiary

One
-
way audio capability

In Scope

Tertiary

External lighting

Secondary

Tertiary

Other:
Intuitive User Interface

In Scope

Not Covered

Other: Sustainable Design

In Scope

Not Covered


As is visible in Table 1, several Statements of Need were moved to a Secondary Priority status.
C
arrying exi
sting wireless video devices is

less preferable to
incorporating an effective system into the
robot’
s design.

Also,
few other items would need to be carried

on the robot
. The robot’s abi
lity to right
itself was deemed

unnecessary provided that the robot design be stable, which is accounted for under
the ro
bot’s ability to navigate stairs and debris.
The

ability to open doors and closed containers is
outside the scope of this project from a design standpoint, but the design team is to research the
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
9

of
78


possibility of implementing an off
-
the
-
shelf arm. Having an

especially mobile and functional camera
is

preferable to multiple cameras with different views. Similarly, infrared camera function is preferable to
external lighting, but the SWAT Team identified that circumstances occur where external lighting
would be

useful. Finally, although Salem PD had initially asked for two
-
way audio, one
-
way audio
from the robot to
an operator

was deemed acceptable
.


In addition, Table 1 shows that two additional Statements of Need were discussed and included in the
list. An in
tu
itive user interface is

necessary for easy and quick robot manipul
ation. Finally, the robot
needs

to be designed sustainably such
that
Salem PD can

maintain the robot and replace parts as
necessary following use.


Another

result of the meeting was a list

of customer requirements that came from the discussion of the
Statements of Need. As each item was discussed, the project team was able to understand the
customer’s precise needs for each item. The resulting list is
Table 2
.

It should be noted that this

is a
qualitative list and that quantitative values for
the items described is included

in section
2.3.


Table
2
: Identified Qualitative Customer Requirements

Customer Requirements

Statement of Need

Rotating camera operation for a
full field of view

Articulating camera

Camera Raising Platform

Articulating camera

Reasonable
resolution camera

Broadcast audio/video

Encrypted wireless broadcast

Broadcast audio/video

Sufficient camera no
-
light range for normal operation

Low light/no
light video

Able to climb stair

Navigate inside of building

Able to navigate around debris

Navigate inside of building

Able to be carried by one person

Other: Forward operations post

Operates in chemical
-
agent environment

Other: Frequent exposure
likely

Safe operation in natural
-
gas environment

Other: Frequent exposure likely

Rechargeable battery

Sustainable Design

Design to accommodate replace
able batteries

Sustainable Design

Select parts that can be easily replaced

Sustainable Design

1
-
way a
udio capability

One
-
way audio capability

Wireless
-
receiving user interface

User interface

Intuitive controls

User interface

Handheld user interface

User interface

Able to operate in overhead precipitation

Water resistance/waterproof

Sufficient runtime

to clear a small building

Wireless Operation

Sufficient wireless range to clear a small building

Wireless Operation


Many of these Customer Requirements are derived directly from the Statement

of Need discussion;
however i
t should be

note
d

that seve
ral
Customer Requirements in Table 2

do not correspond to a
previously
-
specified Statement of Need. These are due to the anticipated standard environmental
conditions of the robot that were not specified by Salem PD.
For example, t
he robot will be operated
fr
om a fo
rward command post by the SWAT team o
perator, so it needs to be
portable and easily moved
.
Also, operations where the robot will be utilized
can potentially

have high concentrations of tear gas
and other chemical agents. The robot also may encount
er natural gas and safety precautions
. These
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
10

of
78


environmental factors must be accounted for in the design even though they were not explicitly
discussed in the Statement of Need.



2.2
Competitive Analysis


See Appendix C

for detailed competitive analysis
.




2.2.1
Product Space Analysis


While there are a few companies such as Remotec, Foster
-
Miller and iRobot
that
have a strong hold
over much of the market, there is room for additional products. The current products can still be
improved upon to meet the d
etailed specifications law enforcement an
d military organization’s need.


Pros and Cons

of Currently Available Systems

-
Chaos High Mobility Platform


Advantages:

A key feature of this robot is it autonomous capacity. While not needed for this
particula
r project, it an interesting attribute that could be quite useful under certain circumstances. The
four articulating tracked drive system enables it to self right itself as well as climb over substantial
obstacles.


Disadvantages:

The Chaos Platform weig
hs 120 pounds with the manipulator adding
an

additional 35 pounds [1
]. This exceeds our desired maximum weight of 100 pounds.


-
Recon Scout


Advantages:

This highly durable ultra portable robot is meant to be tossed
through a window
.
With the simple h
and held control unit and a weight of 1.2 pounds it is a device that could be stuffed in
a cargo pocket and not be a tactical hindrance.
The Recon Scout also has
black and white video

which
allows the SWAT team to better view the situation.


Disadvantages:

With an indoor range of only 100 feet the Recon Scout won't be able to travel
through an entire house with
the operator located outside [2
]. While black and white video is still a
means of getting more information about a building, color v
ideo would provide an extra level of added
detail.


-
DR
-
PROTO


Advantages:

An ingenious quality of the DR
-
PROTO is its ability to maneuver inverted [3].
This eliminates the need to self right itself. With the high durability and rear flip
-
handle, the
DR
-
PROTO

can be thrown where it needs to go and still maintain camera view because the camera is
mounted in the middle of the device, not on the top like most of the competition
.


Disadvantages:

The primary drawback to the DR
-
PROTO is its fixed camera. T
his means
requires the operator to
continually maneuver
the robot around to get the desired view.


-
Packbot


Advantages:

iRobot's Packbot is rapidly deployable weighin
g only 53 pounds and less than
eight

inches high [4
]. With multiple controller inter
face options, the Packbot has intuitive control that
provides the operator with the real time information needed.

This was a robot that the Salem Police
Department cited as an example of something they were looking for.

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
11

of
78



Disadvantages:

A

sticker price of
$85,000

greatly exceeds the Salem PD’s budget
.


-
Lector HD


Advantages:

The Lector HD has the dimensions, weight, range and capabilities quite similar to
the product being designed

in this project

with a weight of 40 pounds, range of 1200 ft and being
remote control capable
. Lector HD

provides a good reference
by containing
no superfluous qualities.
Being able to tow up to 300 pounds is an incredible feat for a robot of its dimensions
however not
relevant to the Salem PD’s needs. T
he IR video would be ideal

for houses with low lighting [5
].


Disadvantages:

With a price tag of $27,600 it is a good example of what a dec
ent budget can
buy. With a $3,5
00 budget, an attempt
will be
made to try

to cr
eate a similar devi
ce, however

won't be
able to purchase the same quality of components.


-
Lector


Advantages:

Being the little sibling to the Lector HD, the Lector has the same transmitter
system and video package as the HD only in a smaller packa
ge.
The smaller, 4
-
wheel drive variant
weighs only
eight
poun
ds and is
small enough to be
portable [6
].


Disadvantages:

The wheeled design would make it nearly im
possible to climb stairs and
maneuver through any
decent quantity of debris.
This model
has a re
tail price of $6,000.


-
TALON


Advantages:

The TALON

has a variety o
f different features and options. It has three IR
illuminated cameras as well as an
auto focus color zoom camera [7
]. With two isolated firing circuits it
can handle a multitude of opt
ional sensors and even have mounted armament such as a shotgun.


Disadvantages:

The TALON
serves as a good example for this project. Unfortunately, it has

a
base price of $60,000.


-
Mini
-
ANDROS II


Advantages:

The fold down 18 inch camera extender on the

Mini
-
ANDROS II allows for a
controlled 360° view. The two meter arm has four degrees of freedom enabling deft manipulation o
f
the surrounding environment [8
].


Disadvantages:

At 225 pounds, with the control unit weighing 38 pounds, this is not a light
robot. It would take time to unload the robot and setup the control
.



-
West Covina SWAT Bulldog


Advantages:

The scissor lift to elevate the camera is a very creative idea which provides
exceptional camera elevation in a relatively small package. Also,

an arm with an extended lifting load
of 15 pounds is useful in
many situations [9
].


Disadvantages:

This is another wheeled design which would not be able to handle stairs
.


-
MATILDA II


Advantages:

The Matilda II has a very
long
run time of six to
10
hours [10
]. The robot has two
means of control: RF and fiber optic. With multiple arm attachments available it can perform many
different mission types.


Disadvantages:

Should the fiber optic control be used, the operator has to deal with manual
pick up

of the cable. Lacking a cord reel, MATILDA II leaves its cable behind as it travels. This
would be a hassle if it traveled through multiple rooms in a dwelling.



ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
12

of
78


Pros and Cons

of Currently Available Components

The following section is a brief summary o
f the most important researched components for this project.
Given the high number of components necessary for this system, it is impractical to discuss each
particular item. Thus, a complete discussion about this subject is provided in section 3.1. In add
ition, a
detailed tabular review of these components is provided in Appendix C, tables 29
-
40.



Microcontrollers

Qwerk Board

Advantages:

Versatile system with many features such as a servo controller, motor controller,

FPGA, and regulated power supply

Disadvantages:

High price; built
-
in servo controller cannot throughput sufficient current; built
-

in motor controller cannot throughput sufficient current


TS
-
7200 SBC


Advantages:

Lower price than Qwerk board;



Disadvantages:

Requires motor controller

and servo controller


Soekris Net4521


Advantages: Group member has extensive experience with board


Disadvantages: Limited digital outputs; no analog outputs


Motor Controllers

Sabertooth Motor Controller


Advantages:

Built
-
in thermal and overcurrent
protection


Disadvantages:

High thermal waste at higher current draw


Servos

HS
-
422


Advantages:

Extremely common servo that is easily attainable if replacement needed


Disadvantages:

M
oderate durability


HSR
-
5995TG


Advantages: Titanium gears providing
much greater strength than typical nylon gears


Disadvantages: Expensive at a price of $115

Motors

IG52 24VDC Gear Motor

Advantages:

Gear
ed to approximately desired rpm;

Has steel gears
which are stronger than
plastic


Disadvantages:

Noisier than stepp
er motor or servo


Ampflow E
-
150 DC Motor


Advantages:
High torque per dollar ratio of 430 oz
-
in torque for $80


Disadvantages:
High waste

with o
nly 74% efficiency.





ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
13

of
78


Cameras

Trendnet
TV
-
IP312W


Advantages: 20ft. night vision range; two
-
way audio


Disadvantages: High power consumption at 2.5A; expensive at $215


Logitech Quickcam


Advantages: USB 2.0 compatible; lower power draw than TV
-
IP312W


Disadvantages: No night vision




ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10

Page
14

of
78


ME Product Analysis



There are many relevant designs

from which our design team seeks to draw insight for this project.
Excellent designs have been developed focusing on several key mechanical aspects of the project,
including mobility over varied terrain types, camera mobility and articulation, and access
ory
manipulation. However, such designs have far greater development resources at their disposal, and
carry a high price tag. This application will work under a limited budget, and must seek a happy
medium of these three critical mechanical design aspect
s.

Mobility: NASA’s “Urbie” robot

This autonomous all
-
terrain robot was developed specifically for stair
-
climbing and general mobility.
The autonomous operation is a significant plus for all
-
terrain mobility, because the device would only
require basic operator directions, allowing operat
ion of the vehicle’s cameras while the robot moves
about. The robot’s wide platform a low center of gravity aid in allowing the robot to maneuver easily,
and we seek to include these aspects in our own robotic design. The multiple, articulating treads ar
e
particularly effective at ascending stairs, but add a great degree of complexity to the system, and would
shorten the vehicle’s operation [11].

Camera: BULLDOG Tactical Robot by The Machine Lab

This robot was initially developed by a high school group,
which was unable to complete their project
and turned design over to the machine lab. The robot was developed specifically for West Covina’s
SWAT Team. The robot has an excellent camera system, incorporating, pan, tilt, zoom, and infrared
operations. In

addition, the camera is mounted to a scissor
-
lift platform that can extend 30 inches
vertically, greatly enhancing the camera’s functions.
The robot incorporates a
manipulable hand which
can open doors and carry objects. Although the camera has excellen
t functions, the device is not as
mobile as may be necessary for some SWAT applications, the robot is heavier than desired by Salem
Police Department, and the robot was produced only once [9].

Accessory: H.D.E. Manufacturing MURV
-
100

This robot is a fully
-
configurable device that can incorporate various drive systems and accessories.
The standard configuration has a wheeled drive system and arm accessory. Due to its configurability,
the robot is ideal for many types of situations, and even its standard ha
nd accessory is extremely
manipulable. The hand includes pincer
-
style grippers that can be interchanged for other contact points,
a rotating wrist, video camera, and multiple articulating points. This arm is great for any sort of object
manipulation, doo
r opening, and even robot self
-
righting. However, the robot does not look ideal for
stair climbing in any of the standard drive configurations. Also, the system pamphlet does not list a
price, and the device looks quite expensive [12].

Mobility and Camera
: Superdroid Robots Tread
-
Driven Robot Chassis Kit

Superdroid Robots is a company that specializes in designing robotics kits that can be customized for
specific applications. From one of their kits they have developed a low
-
profile tread
-
driven robot wit
h
an articulating camera that would be ideal for outdoors reconnaissance purposes. The camera has great
pan, tilt, zoom, and low
-
light features, and with the tread and low profile, most obstacles would pose
little difficulty. A bonus is that the robot is f
ully
-
submersible, indicating that exposure to other
environmental ingredients would probably not pose a problem, either. However, the device lacks any
sort of manipulable arm, which poses a problem for many SWAT applications involving closed doors.
Also,
the kit is more expensive than the project’s budget would allow [13].

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
15

of
78


Overall Best: iRobot Packbot 510 EOD

Although this robot was designed for bomb disposal purposes, its design matches perfectly with SWAT
robot activities as well. It incorporates an art
iculating tread set, a low profile, and an extending camera
arm that also incorporates a manipulable gripper, and has been well
-
ruggedized to withstand significant
shocks. A total of four cameras ensure that a SWAT team would have a good sense of the surr
oundings,
and the robot is fully capable of either placing or displacing ordinances since it was designed for bomb
disposal operations. The major design disadvantage is that the robot is not particularly operable from a
forward post, requiring more signif
icant electronics support than a simple handheld device. And of
course the unit is far outside of the planned budget [4].


Tri
-
Star

Wheel Locomotion

Tri
-
star locomotion
"uses sets of wheels arranged in triangles." It uses at least at 12 wheels, at all time
s
it has 2 wheels in contact with the surface. This design
is optimal

for
uneven terrain because the high
number of wheels enables it to maintain contact points with the surfaces it is traversing
.
Unfortunately
this

design does not work well with the pro
ject requirements because it has

many

moving parts and is
thus unnecessarily complicated [14].


2.3 Feature Set

Table
3
: Identified Quantitative Customer Requirements

Customer Requirements

Defined Value

Rotating camera operation
for a full field of view

360°

Camera Raising Platform

Single jointed camera arm

Reasonable
resolution camera

640x480

Encrypted wireless broadcast

802.11g Wifi

Sufficient camera no
-
light range for normal operation

5ft. no light video

Able to climb sta
ir

8
” high stairs with a
45° pitch

Able to navigate around debris

Debris size of
15”x33”x8”

Able to be carried by one person

100lbs max

Operates in chemical
-
agent environment

Gasketed enclosures

Safe operation in natural
-
gas environment

Gasketed

enclosures

Rechargeable battery

Capable of 1 hour mission time

Design to accommodate replace
able batteries

Batteries can be removed from robot

Select parts that can be easily replaced

Servos, motors, controller PCBs

1
-
way audio capability

20Hz
-
16kHz

Wireless
-
receiving user interface

802.11g Wifi

Intuitive controls

<10 hours training needed

Handheld user interface

SWAT Laptop

Able to operate in overhead precipitation

Water resistance

Sufficient runtime to clear a small building

1 hour mission time

Sufficient wireless range to clear a small building

300ft. range


Table 3 shows the quantified values of customer requirements for this project. These requirements are
examined and discussed in further detail in the section below.


ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
16

of
78


2.3.1 Minimum
Requirements




Ability to climb standard pitch stairs and moderate quantity of household debris

Be
able to navigate
residential housing

including
going up stairs and maneuvering over
common household debris.

These requirements equate to a maximum climbing
angle (at least
45 degrees
) and a maximum stair height (8

inches) to evaluate the stair
-
climbing ability. For
obstacles in general,
we will consider debris width (
15

inches)
, debris length (33 inches) and
debris height (8

inches), assuming a gradual heigh
t increment rather than the abrupt jump in
height associated to a stair.




Robot and operator interface weigh less than 100lbs.

By having the robot be manageable by the operator alone it does not increase the duties of other
officers at the site. This ma
y mean use of a case with wheels or sometime of harness
arrangement that enables the robot and interface to be managed just by the operator.




1 hour mission

runtime

This robot should be able to be
run

by the
SWAT team operator

for an hour. While mission
s
that this robot may be used for could last multiple hours, it needs to be effective for a minimum
of one hour of typical mission usage. This does not mean that the robot could be driven in
circles for one hour. Rather it means that
the SWAT team operat
or will ask the robot to perform
stop and go movement with some stationary camera panning is expected. Precise duty cycles
of the main components of the robot will be figured out upon further communication with the
customer.




300ft. wireless operating
range

Most situations that this robot will be utilized in will occur at single family dwellings and
apartment complexes. The
SWAT team
operator should be able to be at one corner of the
premises and have the robot maneuver to the other end without any los
s of operation or control.




Replaceable and rechargeable battery pack

T
he battery pack will be designed in such a manner that it is easy to swap out a depleted battery
with a new one. This leads to a desired design in which batteries can be recharged wh
ile in the
robot as well as when removed from the robot.





Operator interface receives video and audio from robot

The camera must

have specified a minimum 640 x 480 resolution, and a 90 degree viewing
angle
;
but still

maintain
a portable
size without comp
romising performance.
The intent of this
robot is not to acquire evidence to be used in court, but to increase the safety of the officers that
are put in danger.

The robot will have
one
-
way audio
capability, which

require
s

a sound
frequency response rang
e of
20Hz to 16kHz.





Intuitive Interface

The operator interface for this product will be easy enough to operate that after initial training
,

the operator will be able to consistently manipulate that robot in the desired manner. This ease
of use should be intuitive enough such that any refresher training needed is minimal and not to
exceed one hour per month.
It should take less than 10 hours
for a new user to initially learn the
interface and controls.

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
17

of
78


2.3.2
Target Feature Set




360° viewing capability

T
he video will be affixed to the robot i
n such a manner that it has 360 degrees

panning range of
motion separate from the robot's maneuvering ca
pabilities.




IR directed lighting

T
he robot must be capable of supplying video in a low light environment. IR lights will enable
t
he color video camera to capture

viewable vid
eo in a low light situation.
A no
-
light camera
range of 15 feet will

be suffic
ient.




Camera Arm

A camera platform of adjustable height would enable operators to see over small obstacles. A
raising height of 14 inches will fulfill this capacity.




Default

Camera Positioning

There may be situations in which it would be beneficial for the camera to reset to a base position
with the press of a single button.
The base position would have the camera arm positioned in the
best location that allows for optimal driving conditions.




Cost Efficient

The current budget

is any amount less than or equal to $3500.



3
.

Architectural Overview



The primary purpose of the
SWAT Reconnaissance Vehicle is to

be the eyes and ears for the SWAT
entry team. The entry team currently has a device that performs a similar capacity, but has many
limitations. The Eyeball R1 lacks the range and mobility that the Salem SWAT team desires.
This
robot will be an improvem
ent over the Eyeball R1
’s

deficiencies.


The greatest deficiency of the Eyeball R1 is its lack of mobility. With this in mind
,

the
new
robot will

be geared towards mobility.

A
tracked drive system will be the robot’
s means of locomotion. This
tracked drive will be capable of overcoming small obstacles and climbing building code standard stairs.
The robot will move at a pace that is conducive to intelligence acquisition. This means that the drive
system will propel

the robot in a way such that the
SWAT team
operator can easily maneuver and view

its surroundings
.


T
here is a wide range of
options for control of a robotic reconnaissance vehicle. While tethered
options exist, it is prudent for this design to be a wire
less variant. Wireless capability removes the
hindrance of a connection to the operator interface. An additional aspect is the reduced strain of the
robot’s drive system as it reaches the limits of its range. A tethered version has a large length of cab
le
following it
on the ground which
could become as hassle as the robot is maneuvered through multiple
rooms. A
wireless
operation

would not have this issue
. The most common occurrence in which this
robot will be used is in situations at single family dwe
l
lings and apartment complexes. A
range of 300
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
18

of
78


feet

should
be sufficient for these

circumstances. While a larger range would be preferable,
the
increased power consumption would affect the robot’s runtime.


The

robot must be able to function long enough
for the SWAT entry team t
o gain sufficient
information about

the area of
concern. A mission runtime of one

hour will be acceptable to cover
typical missions.

While
the standard run
time will be one

hour, there will be situations which require
less motor u
sage and some that require more.
The standard operation for the robot equates to moving
for a period of time, scanning with its video camera for a while, and then proceeding forward.
Since the
power draw will vary depending on the duration that the motors

are used, it will be beneficial to have a

low battery indicator
. Additionally, it will be imperative that the battery system is replaceable and
rechargeable. When there are missions that exceed the defined standard, the SWAT team needs to be
able to sw
ap out a depleted battery with a charged battery so that the robot can perform during the
needed duration of the mission. There also needs to be a convenient method of recharging

depleted
batteries both at the station and in the field.




A visual

reference

into a hostile situation can provide invaluable information on how to best handle the
circumstances. This is why video must be
of high enough quality that the operator can determine
distinct objects.

In order to ensure this, a minimum resoluti
on of 6
4
0 x 480 has been
required. To
increase the camera’s effectiveness it will be capable of panning and tilting as well as be mounted on a
moveable arm.


An appendage
, such as a movable arm,

that extends the camera

above its normal position on the rob
ot will
allow it to see over small obstacles and increase its overall viewing ability. When used in conjunction with
the camera’s pan and tilt functions, the camera should be ab
le to view the robot itself. This may prove to be
a desirable driving style f
or some operators as well
as

a tool for troubleshooting an issue if the robot
becomes stuck or malfunctions.



3.1 Implementation Approaches


ECE Considerations


The SWAT Reconnaissance Vehicle project is a complex and broad project. Design approaches can
stem from many different feature sets that the project requires. Thus, to simplify and objectify the
implementation design of this project, it is best to start with the core of the vehicle: the main processing
and control unit. After exploring dozens of ex
isting products, three main branches remain in the final
stages of deciding on an implementation approach. Ultimately, only one of these will remain as the top
and final choice for this project.

Note: For detailed information and clarification about produc
ts discussed in the following paragraphs,
please refer to the tables in Appendix C.



3.1.1
Microcontrollers


Soekris net4521 Communication Computer

One implementation approach for this project is centered on the Soekris net4521, an embedded
computer board

designed and optimized as a wireless communication appliance [15]. In addition to
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
19

of
78


meeting most interface requirements that are expected for the reconnaissance vehicle, this board boasts
long life and low power consumption which are essential characteristi
cs for this project. Furthermore,
one of the team members has extensive experience and skill working with this embedded computer,
thus decreasing the learning curve for the group.


However, the net4521 board lacks certain features which would reduce the ne
ed for additional
components. The most significant disadvantages of this board are its lack of USB support and its few
digital I/O ports. Having no USB support poses implications for interfacing video and audio, requiring
additional hardware, thus offsetti
ng the lower price of this product. Having fewer I/O ports, this
product is not very suitable for this application. Unforeseeable design changes at later stages could
require many additional ports, which would make this product nearly unusable. Lastly, in
comparing
this product with other chosen solutions, it does not include support for seamless integration with
motor controllers.


The net4521 solution is a cost effective approach for the reconnaissance vehicle; however when
compared with other products, i
t doesn’t meet our
needs
. Its design for quick wireless integration is
ideal, but its lack of effortless integration with motor controllers keeps this board from being the best
implementation approach.



The Qwerk Board

Another

choice for the implementatio
n approach of this project is using the Qwerk Board, an excellent
integrated solution for all
-
inclusive control [
16
]. This board, produced by CharmedLabs, boasts
many
advantages above

other choices for control boards/ microcontrollers. Still standing at a
reasonable price
of $350, the Qwerk board includes features such as integrated WiFi support, Webcam Input, USB 2.0,
digital and ana
log I/O ports, and many others. With such a superb feature set, this board is a tempting
solution.


Initially, the

Qwerk

Boar
d
was considered

because of its integrated

2A motor controllers and 16 RC
-
servo controllers. Designed for robotic purposes, this board incorporates the major interfaces such a
project would need, in a single robust enclosure.
This seemed ideal for
the

system. However, these
integrated controllers were designed for small applications, not for a combat
-
type robot. Thus, after
further research, d
esign, and consultation with the

Mechanical Engineering group, it was

determined
that the
motors and servos nee
ded for the

project would exceed the driving capabilities that the Qwerk
Board could provide. As a result, the features that initially made this board so attractive were rendered
unusable. Nonetheless, the Qwerk Board could still be used with external moto
r controllers and servo
controllers. Its many remaining features such as WiFi support and USB 2.0 still significantly improve
the feasibility of this approach when compared to the aforementioned Soekris net4521.


In this implementation, the Qwerk Board wo
uld act

as the brai
n of the entire vehicle. It would

control
the motors, the servos, video/audio, an
d

wirelessly relay the information to the control station. After
researching many products, the few best picks in each category have been compiled in Append
ix D.
The reconnaissance vehicle
would

feature the Kinamax WCM
-
NV13 webcam as its eyes and ears [
17
].
The webcam would

connect via USB 2.0 to the Qwerk Board, which will transmit wirelessly to the
control station via the WaveRV II wireless adapter [
18
]. H
aving a range of up

to 1 mile, the wireless
adapter would

have no trouble transmitting the data via the 802.11g protocol to the control st
ation. This
control station would
consist of a laptop provided by the Salem Police Department. Operating from a
safe d
istance, or in the
front field, a SWAT officer would

easily control the vehicle via the software
interface built by the SWAT Reconnaissance Vehicle Group. Taking commands from a few in
tuitive
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
20

of
78


buttons, the laptop would

transmit controls to the vehicle via t
he same protocol, 802.11g. The Qwerk
Board will be programmed to interpret the commands in the correct fashion and relay them to the
integrated motor and servo controllers. These components will then send signals to the
DeWalt 24VDC
Hammer Drill motors whi
ch will drive the vehicle in the desired direction [
19
].


Ultimately, the Qwerk Board would be a perfect solution for a system of a smaller scale and with less
demanding requirements. However, some of its features are not ideal for the project at hand, thu
s not
justifying a top spot in our pick for a microcontroller unit.



TS 7200
-
16

Linux Embedded Computer

Ultimately, the top

implementation approach is founded on a Linux embedded co
mputer, the TS 7200
-
16
, from EmbeddedArm [
20
]. This
Single Board Computer
(
SBC
)

boasts many features suited for the
reconnaissance vehicle. Havin
g easily upgradable memory on flash

cards, less than 2 second startup
time, seamless audio/video integration and USB 2.0 ports, this product is a big improvement over the
net4521 board
discussed in the previous section. Having USB support is a significant advantage.
Webcams for audio and video transmission commonly use USB connectivity. Also, wireless
transmission can be achieved in mere seconds, using an inexpensive USB Wireless adapter
. T
his board
includes 20

digital I/O ports, allowing for easy integration with sensors and additional control features.
Additionally, it comes with an 8
-
channel DAC, enabling seamless integration with motor and servo
controllers.
In terms of integration wi
th peripheral devices
,

the TS 7
2
00 SBC is an excellent product.


In this implementation, the TS 7200 SBC would be the

central processing unit for the

system. The
vehicle will feature the Logitech STX

961410 as

a drive camera, connected via USB 2.0
interface
directly to the TS
-
7200 [
21
]. Additionally, a main camera suspended on a pan/tilt arm will provide
video, audio, and night vision, as necessary. This camera will be the Treadnet
TV
-
IP312W

and will
be
connected via Ethernet directly to the

Ethernet bridge

[
22
]. A SWAT officer operating from a safe
distance will control the vehicle with a laptop, via a software interface which will be built by the CS
member of the SWAT Reconnaissance Vehicle group. Thus, the commands will be transmitted
wirel
essly, via the 802.11g protocol to a Linksys WRT
-
54G Ethernet bridge installed on the vehicle
[
23
]. The commands will then be processed by the TS 7200 SBC and transmitted to the various systems
on the vehicle via the digital and analog outputs. Controls co
ncerning the movement of the vehicle will
be sent to the 2x10
A

and 2x25
A

motor controllers [
24
]. Controls related to camera movement will be
relayed to the SSC
-
32 servo controller which controls the pan/tilt system of the camera [
25
]. Each of
these control
lers will then send the appropriate voltage levels and control signals to the respective
motors and servos.

In this manner, the TS 7200 is an ideal board for controlling the SWAT Reconnaissance Vehicle.
Interconnecting all the systems of the vehicle and pr
oviding the processing power and necessary
interfaces to receive and transmit signals via the Ethernet bridge, it is able to fulfill all the necessary
requirements for the project hand.


3.1.2
Video
/Audio Transmission

In addition to the microcontroller,
the transmission of video/audio is also an important element of this
project. Thus, careful considerations were made in choosing the design and products of the video/audio
system.


WCM
-
NV13

Camera

This USB camera, made by Kinamax is an excellent, affordabl
e solution for transmitting video
[17]
.
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
21

of
78


Costing only $20.00, this webcam supports USB 2.0 and has an integrated microphone for a 2
-
in
-
1
audio/video system. Featuring a resolution of 640x480 pixels, this camera meets the resolution
requirements for this proj
ect.


However, this camera does not have support for night vision. Using this camera would require either
external lighting (which is not preferred by the Salem PD SWAT team), or a secondary camera with
night vision. Also, the camera does not have support
ed Linux drivers. Thus, using this camera would
also require writing a Linux driver. Although this camera is very affordable, it does not seem to provide
enough support to be usable



SE 200 Spy Camera

Another approach could be using the SE 200 camera by S
py Associates. This camera is very costly, at
$200.00, but it is also very powerful. With a night
-
vision range of 30 feet, this camera has no problem
transmitting usable video in full darkness. Designed for use as an outdoor security camera, it is also
wat
erproof.


On the other hand, this camera is designed for input into a television or monitor that supports RCA.
One way around this could be to use a Hauppauge RCA to USB converter. The cost of this product
would be $50.00. More importantly, Linux compatibi
lity of this device is limited (as it is with most
devices of this type), and would take significant student time to achieve a smooth, working system.
Although this setup is a viable solution, its need for additional components and student time keeps it
fr
om being a top choice.


TV
-
IP312W and QuickCam STX

The top approach for implementing a video/audio system is with the TV
-
IP312W network camera and
the QuickCam STX webcam. The TV
-
IP312W network camera costs $215.00 dollars, while the
webcam costs $36.99. A
lthough somewhat costly, these two cameras together provide a system that is
well suited for the needs of this project. The network camera will be used as a main camera and will be
placed on a pan/tilt system (built by the ME group), while the webcam will
be used as a drive cam,
placed low on the vehicle to make obstacles can easily be avoided.


Both cameras have a resolution of 640x480, thus providing great video for both main surveillance and
driving the vehicle. In addition, the network camera features two
-
way audio capability adding an extra
dimension, should it ever be needed. The Ethernet in
terface of the network camera and the USB 2.0
connectivity of the webcam also make these cameras convenient to use, given the TS 7200 SBC used
for this project. Having two cameras also allows for unexpected malfunctions or damage to one of the
cameras, whi
le continuing to have video from the remaining camera. This video/audio system is the
chosen implementation approach for this project.


3.1.3 Wireless Communication

Another important component of the SWAT Reconnaissance Vehicle is the transmission means be
tween
the controller of the vehicle, and the vehicle itself. In accordance with the requirements of this project,
only wireless transmissions were considered for this project.


WUSB300N

On way to implement a wireless transmission system is by using the WUS
B300N USB wireless
adapter, made by Linksys. This product features USB 2.0 for fast data rate transfers and the new
802.11n standard. Being a small USB device, power consumption is minimal, taking away the need for
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
22

of
78


valuable energy. This product would work
by being connected to microcontroller board via USB, and
transmitting data between the microcontroller and a wireless router (such as the Linksys WRT
-
54G),
which would be positioned on the Salem PD SWAT bus


the vehicle used by the Salem SWAT team to
driv
e to the mission site. This would act as a bridge between the controller and the vehicle.


However, a downside to this implementation is that the transmission would be dependent of the
position of SWAT bus. Should the vehicle need to be moved out of the r
ange of the wireless signal,
connection would be lost. There may be many circumstances as to why the SWAT bus would need to
be moved during the course of a mission. Therefore, this implementation would not be the most
efficient.



WaveRV II

Another possible solution for the wireless communication system is the WaveRV II wireless adapter.
Featuring USB 2.0 this product is power efficient as well. Main advantages of this product are its
increased range which is claimed to be up to 1 mile

[18]
.
Having a configuration similar to the one
described in the WUSBN200N section, this adapter would minimize the concern of moving the SWAT
bus. This way, even is the bus would be moved the distance of 1 mile, the transmission of the signal
should remain acti
ve.


Nonetheless, this configuration still does not fully eliminate the initial predicament; it only allows for
some movement of the SWAT bus around the mission site, but not the freedom to leave the scene, if
necessary. In addition, the cost of the WaveRV

II is $150.00, and its range questionable when
obstructed by buildings and vehicles. Although it is a lucrative way to increase the range, this
implementation could still constrain the SWAT team in certain situations.


WRT
-
54G

The top implementation appro
ach for a wireless communication system is by using the Linksys WRT
-
54G wireless Ethernet bridge by itself. Placing it on the reconnaissance vehicle directly, it would
eliminate the need for an intermediary router/bridge to be placed on the SWAT bus. Thus,

the
transmission of data will be directly from the controller to the wireless bridge, and into the
microcontroller via Ethernet connection. The cost of the WRT
-
54G is only $49.99 and by connecting it
directly to the microcontroller board, it would elimina
te the need for any additional wireless adapters.
Also the delay of the signal will be minimized by eliminating an intermediary device, as proposed for
the aforementioned approaches.


This approach is not perfect. Although achieving independence of the SWA
T bus, this increases the
power need of our system. The WRT
-
54G Ethernet bridge draws 1A of current, thus significantly
increasing the power consumption of the wireless system. All things considered, this is the most
effective approach of the three. Althou
gh it requires more power, it allows our system to become self
-
sufficient, consisting of only the vehicle itself and the controller.


The components discussed above are the primary components for this product and require the most
thought due to the complex
ity of the system.
Additional components are required to complete the
product, but there are a wide variety of acceptable components allowing for many different options.
These options have been examined and the results may be viewed in Appendix
D
.



ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
23

of
78


ME C
onsiderations


Drive System

Tank Tread

The tank tread is a simple design that has been implemented in recreational, military, and toy vehicles.
The setup for each tread involves a set of wheels that are free to rotate, at least one drive motor, a drive
train to transfer power from the motor(s) to one wheel, and a tensioned track system wrapping all the
wheels. When the motor is powered, it drives the one wheel via the drive train. The track then
functions to transfer power to all the other wheels, creat
ing a long surface area with which the vehicle
contacts the ground. Most tread designs incorporate two track systems


one on either side of the
vehicle. Track designs have been well
-
developed for many applications.

See Figure
1 below, showing
a basic t
readed design where the non
-
driven wheels each have a suspension system.



Figure
1
: Single Tread System [
26
]



Two options were considered with a basic tank tread design


a total of 2 or 4 drive motors (1 or 2 per
side). Both tread designs have quite good abilities of navigating obstacles, climbing stairs, and general
maneuvering. They would be very simple to de
sign and fairly simple to build, with a 2
-
motor design
slightly favorable. Unfortunately, in order for a tread design of this type to climb stairs, a long robot is
required. In addition, if the motors break or if a tread falls off, the robot’s function i
s lost entirely, so
neither design is particularly robust.


A 2
-
motor system has very low power consumption and cost, and would not require intricate control
system design. Implementing 4 motors, on the other hand, costs more, draws more power, and requ
ires
some special controls. Neither system would require terribly many fragile parts, so although the
designs are not robust it risks little damage to the system.

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
24

of
78


Multiple Pivoted Treads

One level of complexity up from the simple tank tread design is a mu
ltiple tread design. This design
incorporates all the same components as the tank tread, but rather than a single tread per side, there are
two shorter treads. Although this adds some benefits to terrain negotiation, it vastly increases the
complex
ity of

the system and requires four

or more drive motors. Multiple tread designs have not been
implemented

into many commercial products due to the complexity of the system.


For the multiple treads design, two systems

are considered
. The first incorporates two

parallel sets of
treads as
mentioned

above. We also incorporated a feature allowing the treads to pivot about their
center, allowing the treads to adapt to terrain. We called this the Sheila design. The second system
incorporates one set of parallel tr
eads similar to the tank tread describ
ed above, but also incorporates

a
pivot feature. However, a third tread, mounted to the center and rear portion of the robot, provides the
main drive power and incentive to surmount stairs. Due to the three treads, w
e named this design the
tri tank.



Figure
2
: ATV with a Sheila style Multiple Treads Design [
27

]


We evaluated both multiple treads designs against the same criteria as the Tank Tread designs. Both
multiple treads incorporate f
eatures that allow improved ability to climb stairs. The Tri Tank, however,
is anticipated to have difficulty with variable terrain, and would require some ingenuity to develop an
excellent general drive function. Sheila excels in both these categories.

Neither would be difficult to
design or to build. The tri
-
tank depends heavily on its rear tread, so it is not particularly robust. Sheila,
on the other hand, should

still

function having lost any one

motor or tread, and could possible r
etain
limited mo
bility without two

motors or treads, so this design is very robust.


Both multiple treads designs allow for a smaller size of robot that Tank Tread designs. The cost and
power consumption is comparab
le to a tank tread design with four

motors. Control sys
tem design
would not be too difficult for either system. Due to slightly increased part counts and complexity

the
possibility for damage to the system arises.
The Sheila design wins bonus points for its excellent
ability to navigate random obstacles, whi
le tri
-
tank’s specialized stair
-
climbing design gives it a bonus
factor as well.

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
25

of
78


Articulating Treads

The modern state
-
of
-
the
-
art design in small r
obots is the articulating tread

design. This design begins
with a tank tread design directly attached to the
robot frame.

Connected to the
motor drive are

flipper
attachments

that increase mobility.

However, a pair of servomotors allows the second set of treads to
rotate about the shaft’s axis, therefore rotating relative to the first set of treads. This artic
ulation proves
particularly handy for surmounting obstacles and climbing stairs, because the angle of the articulating
treads can be adjusted to adapt to the terrain.

See Figure
3 below, should the iRobot 510.



Figure
3
: iRobot

510 demonstrating articulating treads design [4]


The Articulating Treads design was evaluated in the same manner as the other treaded drive systems.
This design has an excellent ability to climb stairs and to navigate obstacles, and has a great general
drive function. The system is very difficult to design and build from a mechanical standpoint due to
the many, intricate parts, and high tolerances required to prevent interference. The design is fairly
robust because a lost tread or failed articulator wo
uld not prevent the robot from functioning.


The design allows for very small robot size due to the drive function versatility. Terrible marks are
received for the cost and powe
r consumption due to requiring four or six

specialty motors. Similarly,
the Articulating Treads design would need some very special control system designs. A significant
bonus to the design is the state
-
of
-
the
-
art status, and the overall best
-
in
-
class maneuverability is very
important.


Camera A
rticulation

Lifting Arm

One option for the articulating camera arm is a single member that can rotate around its base and also
allows articulation of a platform mounted to its end. This design requires three articulation motors and
several structural piec
es. The basic arm’
s function is shown in Figure
4. Attachment of the arm could
be inside or outside the frame, allowing for some interesting adaptation to the robot’s body and frame.
ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
26

of
78


The arm would be used to raise the camera to in order to give it the l
argest viewing range possible.
Also the arm could lower the camera into a drive view position to optimize the control of the robot.



Figure
4
: Superdroid Robots kit with lifting camera arm [13]


Space
-
efficiency is a major ad
vantage of using this lifting arm. The arm is stored close to the body of
the vehicle, ensuring a low center of gravity, and the number of unnecessary parts is minimal. Few
moving parts are used in this design which makes it easy to design and build. In

addition, the design is
stable due to its few parts, and that is helpful in preventing damage to the system. The relative
simplicity of this lifting arm makes it very adaptable to future designs.


Future projects could easily expand into a jointed arm
or the implementation of a hand as well.
Relative to a jointed arm, the control system is anticipated to be simple, and routing wiring should not
be difficult either. The single member lifting arm has a decent range of motion which allows for several
view
ing positions. The range of motion is limited both by the length of the arm and by the amount the
base can articulate. A bonus to this design is that the arm can extend beyond the body of the vehicle. If
designed properly, a sort of self
-
righting mechan
ism can be incorporated.


Jointed Lifting Arm

A second design to be considered is the Jointed Lifting Arm, which is similar to the Lifting Arm but
incorporates additional points of articulation. In order to increase the range of a mounte
d camera, the
desi
gn in Figure 5

uses three members, each nearly the length of the robot’s body. These three rotating
arm segments increase the distance the arm can reach away from the robot, as well as the number of
views for which a camera can position. A rotatable came
ra platform attached the end of the jointed
lifting arm gives this design eight degrees of freedom.


Figure
5
: iRobot 510 EOD model includes an articulating arm [6]

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
27

of
78


This design is nearly as space
-
efficient as the simple Lifting Ar
m, but the increased arm reach is
particularly useful. The operator can lift the robot off an obstacle and self
-
right itself using the jointed
arm. Both a camera and door opening capabilities could be incorporated into the design.
Unfortunately this des
ign is significantly harder to design, build, and operate compared to the simple
Lifting Arm. There are many fragile parts, and the components would cost significant amounts.


Due to the numerous joints, many motors are required to control this arm, whi
ch has several negative
consequences. Control systems become very difficult, energy usage increases significantly, and camera
stability and wiring could become problematic. Many moving parts means that there are many
potential break
-
points. However, the

many configurable viewing positions is still a significant bonus,
as it the easy adaptability to implementing a manipulable hand.

Scissor
-
Lift Platform

A scissor
-
l
ift platform, shown in Figure
6, is the third alternative that we considered. This platfor
m
mechanism utilizes many crossing members linked at ends and centers. When the ends of two crossing
members are forced towards each other, the kinematics of the linkage
extends

its top upward. Few
motors are required and the design is very simple.



Figure
6
: Scissor
-
lift platform from West Covina SWAT Robot [9]


Unfortunately, a scissor
-
lift is not very space
-
efficient due to the many members required for it to work.
It is, however, the easiest to design and build, since many similar systems have been implemented in
industry for general applications. Due to the m
any parts and reliance upon every part for structure, the
Scissor
-
Lift is not very robust. Higher material costs are required due to the many identical members,
and the lift mechanism is not very efficient from an energy perspective.


A control system f
or the scissor
-
lift would be extremely easy, but the platform’s stability depends on
the many members being identical. Only one viewing position is possible, and the design is not
adaptable to future articulation needs. Wiring can be tricky due to the ma
ny bends that occur.
Furthermore, the many moving parts are worrisome weak links.






ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
28

of
78


4
.

Top Level Description


Figure 7 is a visual representation of how the world interacts with the product as well as major
components of the product.

Table 4 des
cribes

these interactions in de
tail.

This design is based upon

a
wireless communication method which uses 802.11 protocols. The preliminary design of the controller
is a computing device with wireless access such as a laptop.




Figure
7
: Top Level Description


ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
29

of
78


Table
4
: Top level interface description


ME
Rational
e

for Selecting Design

Multiple (four) treads combine
d with the ability to pivot and a simple lifting arm with an articulating
camera system are the selected designs for the SWAT robot. These designs meet all the customer
requirements while keeping the project within the current budget. The sponsor request
ed for this
device to bridge the gap between the advanced functions of the Salem Bomb Squad Robots and what
the Salem SWAT Team needs out of a low cost RC
-
Style robot. The combination of the chosen designs
fulfills this request. The lifting arm has enoug
h range to be positioned in such a way as to give a good
drive view angle for the vehicle. Only primary customer requirements were met with these designs as
a result of the limited budget and time available for the production of this device.


The (Sheila) tread design is a multiple tread system that has the ability to pivot about the center of each
tread;

it scored the highest at a rating of 4.2 and was the design choice chosen for the drive system.
Articulating treads scored the second highes
t with a 3.96. Five categories were weighted at or above
10% for the drive system matrix: ability to climb stairs, ability to navigate obstacles, general drive
function, cost/power consumption, and a bonus factor. Climbing stairs is one of the customer
r
equirements, this ability greatly increases the mobility of this robot and it was expressed that it would
be primarily used in apartment complexes with stair cases. Navigating obstacles is of high importance
as reconnaissance and use of the device will be

severely limited if it cannot traverse the messy
environment it will be exposed to. Without the ability to drive the robot would be quite useless, as
such it was weighted the highest. The cost and power consumption of the robot realistically limit the
o
btainable parts for each design. Lastly a bonus factor was considered for each design, this score was
found based
extra benefits associated with each design.

Interface

Description

Human Input Control

The user interfaces with a written computer program, Java,
which will utilize the keyboard buttons such as the arrow
keys of a laptop computer. It will control the
robot via
wireless communication using 802.11g protocol on the
2.4ghz range.

Environment

Overhead sprinkling precipitation; obstacles that are no
larger than the robot itself 15x33x8 inches (WxLxH) ; stairs
maximum 8 inches; temperature 0
-
35 C ; Surfaces

such as
carpet, tile, wood, concrete, grass

Power In


120VAC 60Hz for charging batteries; Battery indicator for
charging and current monitoring

Drive Movement

Forward, reverse and rotational movement at variable
speeds no greater than 6 feet per second
for maneuvering. 2
motor rear wheel drive; One motor to drive flippers for
clearing obstacles; each motor capable of variable speed.

Arm Movement

Radial motion from a rotating base with 2 rotations max at
720 degrees; pan and tilt X and Y directions of 90

degrees
and 180 degrees respectively

Video/Audio

640x480 video; 30fps; Infrared visible to 5ft; 20Hz
-
16kHz
audio one
-
way.

ECE Group 4

SWAT Reconnaissance Vehicle

ME Group 10


Page
30

of
78



The simple lifting arm scored the highest at 4.28 and was the selected design for the SWAT robot
. Five
of these categories were weighted at or above 10% for the system: space efficiency, simplicity to
design/build, possible damage to system, control system design, viewing positioning, and a bonus
factor. Space efficiency is of high important becaus
e this camera articulation system cannot be in the
way of the rest of the robot. The design selected cannot be too complex to fit within the scope of the
project
. I
ts overall simplicity is of much importance for both the mechanical design team
,

as well as for
the electrical design team. A large arm system could easily be damaged so its ability to be stowed
while the robot is in motion is critical. Also a bonus factor was included which was scored based on
the perks of each design, for example
the jointed lifting arm would be a good project for a future senior
design project.


With multiple pivoted treads the stair climbing capabilities of this device will be much greater than the
simple tank tread system while remaining cost effective. If the