ROV Snoopy and ROV Woodstock
Cambridge Underwater Robotics Team
Cambridge Rindge and Latin School
Cambridge, MA 02138
E. Huo (2004), C. Grouard (2004), C. Katrak (2004),
C. Stefanov-Wagner (2006)
Mr. Paul McGuinness
June 11, 2004
Abstract
The team is now competing in the MATE ROV Competition for its
third year. The three members from last year recruited a new member,
and began to meet after school once weekly in November. Until February,
meetings consisted primarily of brainstorming, considering various
strategies and designs, which could best complete the mission tasks. The
decision was made to attempt all mission tasks, and a rough design was
chosen of two small robots, controlled separately, with high
maneuverability and mostly passive devices. Designs were then refined,
with construction beginning in early March. In the final two weeks before
Regionals, the vehicles were tested in the school pool and modified as
necessary.
Fol lowing regionals, we disassembled the robots, planning to
reassemble them on frames more suited to their specific tasks. During
rebuilding, we added an additional vertical thrust motor to each vehicle,
improved several task devices and made the motor and camera mountings
more secure. Rechristened Snoopy and Woodstock , the vehicles are now
ready for pool testing and competition.
Contents
1. Design Rationale 1
2. Electrical Schematics 4
3. Description of At Least One Challenge 5
4. Troubleshooting 6
5. Future Improvements 7
6. Lessons Learned or Skills Gained 9
7. Use of ROVs in National Marine Sanctuaries 10
8. Acknowledgments 13
9. Budget and Expense Sheet 14
1. Design Rationale
There are several key design features to the mission system we have
constructed. The most important of these features was the decision to
construct two ROVs to complete the array of tasks set out before us. There
are several advantages to using two ROVs rather than just one. First, each
ROV is simpler, because it has fewer tasks to complete. Second, the
smaller number of tasks per ROV allow for larger devices while
maintaining the general design of a small, highly maneuverable vehicle.
By having two vehicles completing tasks at one time, controlled by
separate drivers and independent control systems, less time is required to
complete as many mission tasks as possible. With seven tasks to
complete, time becomes an important consideration. (N.B. We received
off icial confirmation that having two ROVs, together operating under the
25-amp limit, i s allowed.)
ROV Snoopy ( l eft) and ROV
Woodstock (right)
Another notable feature
is the passive nature of most
of our task devices. By using
non-powered devices to
complete most tasks, we were able to limit current draw and remain
within limits even with two vehicles. For the patch placement, fish
Page 1
col lection, and bell identi fication, simple, non-powered devices complete
a task that would otherwise require a very complex powered device. For
the releasing of the towfish hook, dropping the tags, and the collection of
rocks, we could not think of any very reliable passive device, so these
three devices are actuated. With our effort to limit current draw, even
these devices are very simple, largely passive devices with a control led
trigger, relying on gravity and maneuvering to complete the remainder of
each task.
ROV Woodstock ’s tag-
dropping device is powered
by two solenoids, taken from
an old doorbell , wired in
parallel but with oppositely
facing diodes. When the
actuator DPDT switch is
flipped to the left, the left
solenoid pulls the core back,
releasing the left tag. When
the DPDT switch is fl ipped
to the right, the right tag is
released. On top of each tag
is a micro-Mag-Lite that
shines straight down. This
Mag-Lite shines a spotlight
over where the tag is supposed to land.
Page 2
Another powered
device is the rock-collection
system. After testing a
passive system which
consisted of a sharp scoop,
we concluded the ROV
Snoopy did not meet the
power requirements
necessary for reliable retr ieval of the rocks. Thus, an active system was
developed that use a bilge pump to act as a vacuum for rocks. The bi lge
pump just has i ts protective covering removed, and instead of one exi t for
water, a grill of holes were dril led on the side casing for water to exit.
This allows for greater water flow and no exertion of a lateral force by the
bilge pump. In addition, a chicken wire
covering protects the rock sample from
col lision with the bi lge pump’s propellors.
When the rock is retr ieved into the main
tube, the bilge pump is turned off so that the
rock can fall into a bottom pocket. Once the
ROV has resurfaced, a removable cap below
the bottom pocket is taken off for easy
removal of the rock sample.
Page 3
12 V( - ) ( + )25A Fuse (Provided to Competitors)
Left Thruster(Bilge Pump)
DPDT Switch
DPDT Switch
DPDT Switch
Right Thruster(Bilge Pump)
Vertical Thrusters(Bilge Pumps)
Left Thruster(Bilge Pump)
Right Thruster(Bilge Pump)
Vertical Thrusters(Bilge Pumps)
DPDT Switch
Note: Dotted line represents separationbetween the control box and the ROV itself,almost like an imaginary waterline.
Power Switch
A
A25A Ammeter
20A Fuse
PowerSwitch
DPDT Switch
Tag Dropper(Two Solenoids)
Hook Release(Hobby-Kit)
10A Fuse
DPDT Switch
12V DC Out (Camera)
12V DC Out (Camera)
25AAmmeter
( + ) ( - )
( - ) ( + )
2 SPDTSwitches
2 SPDTSwitches
2. Electrical Schematics
Page 4
3. Description of At Least One Challenge
One of the most challenging issues in constructing the ROVs was the
attempt to keep devices simple and passive. Oftentimes, outlandish ideas
are the f irst response to a complicated challenge, such as the seven
mission tasks. However, outlandish ideas are more complicated to build,
and, according to Murphy’s Law, more
likely to fail , while a simple device requires
a less intensive build and is less likely to
fai l . One of the best examples of simplifying
a complex idea was with the fish collection;
original ideas included forks that would lift
up and down picking up the fish, and even a
complicated arm was briefly considered.
However in the end, a simple pair of
parallel rods was our solution.
This simple device for collecting fish also meets our second design
goal of using passive devices. We tried to keep devices passive in order
to conserve current draw— with two ROVs, current draw was a constant
concern. Oftentimes a single motor could have made a task much simpler,
but we kept to our goal of passive devices, successfully limiting each
vehicle to only one powered device.
Page 5
4. Troubleshooting
The cameras we ordered claimed to be waterproof, so a shallow-
water test was done on one of the cameras.
However, the tested camera was not entirely
waterproof, proven by the fact it was not
operational the next day, and when the
camera was opened, much of the circuit
board was covered in rust. Therefore,
another camera was covered in sil icone glue
at all the seams, and was tested in deeper
waters. The camera was sti ll ful ly
operational afterwards.
Whenever a motor stopped working, connections were checked
using the probes of a multimeter. First a voltage was checked to see if the
motor was receiving any signals from the control box. If there was no
signal, then the tether and the control box
had to be checked. I f there was a signal,
then the leads of the motor were probed for
continuity. If there was no continuity, then
the shaft of the motor was manual ly forced
to turn to attempt to clean the brushes. If
that stil l did not work, the case of the
motor was opened to check the status of the
inside components. The motor was then
Page 6
cleaned and lubricated, or replaced if it was too corroded or damaged.
5. Future Improvements
In any engineering project, there are possibi lities for improvement.
This is true of our ROVs. One feature which needs improvement is
construction material for the frame. While
using PVC for the frame is cheap, durable,
and easy to modify, there are drawbacks.
The length of piping that goes into joints
varies widely; meaning that cutting out the
correct pipe lengths is dif ficult , if not
impossible. Due to considerations of tools
and budget, we plan to continue using PVC
while looking for other possible materials.
Another area for improvement is in the control system. While the
current rocker switches we use are suff icient, they do not allow analog
control. An analog joystick would allow improved control , through
integrated controls, and variable motor speeds. Further more such a
setup would easier to use as piloting the ROV becomes more like flying.
Page 7
ROV Woodstock ’s Control Box Layout
Greatest of all improvements would be the use of a fiber optic
control system. The tether could be reduced to a single power cable and
fiber optic cable. One power
cable means that there would
be little if any power lost in
tether. Smaller tether would
result in fewer restr ictions
for a maneuvering ROV. The
fiber optic cable could carry
as many sensory inputs as
necessary allowing it to be re-used year after year . Currently the tether is
Page 8
connected to the control box via DB9 connectors, which offer at least four-
pair of connections.
6. Lesson Learned or Skills Gained
The most important lessons learned during the completion of the
project were to begin earlier and to use more secure building materials.
In the past, we have used plastic cable-ties
to attach motors and task devices.
However, in modifying our vehicles since
regionals, we decided to use pipe brackets
to hold motors and devices. This makes the
propulsion more efficient due to reduced
vibrations, and makes the task devices
sturdier and more effective. We plan to
continue using this more professional
approach in future ROVs.
This year , we began meeting about six months before the regional
competition but did most our construction in the final three or four weeks
before competition. This was because we spent too much time earlier in
the project designing speci fic devices without real ly knowing the exact
dimensions of the tasks. All of this early brainstorming had to be revised
when we had more information. Also, we were designing primarily in
our heads and on paper, rather than with real objects. In the future, we
plan to work more with three-dimensional models in the early stages of
Page 9
design, and to build a functional vehicle for practice during the fal l .
Later in the project, when we have more information, we can bui ld the
actual vehicle with more knowledge and experience.
7. Use of ROVs in National Marine Sanctuaries
ROVs are used in National Marine sanctuaries to enhance the work
done by scientists. They take pictures of ocean li fe around to help us
understand what functions occur in sanctuaries and how they can be
protected. They are used to videotape ocean life or as l ive camcorders to
further aid the understanding and exploring of marine sanctuaries.
When there are unfavorable and hazardous diving conditions such
as during the night and exploring the ocean f loor, ROVs are very useful.
They can stay down for many hours since they are not limited to an
oxygen as a diver would be. ROVs have plenty of l ight and are very
maneuverable. They can hold a certain location for hours at di fferent
depths. More importantly, ROVs are used to collect biological and
geological samples and to understand and characterize coral reefs. ROVs
also measure distance, which helps scientists evaluate conditions at
certain depths and potential problems.
ROVs were deployed in the Olympic Coast National Marine
Sanctuary to explore the organism activities late at night on the ocean
floor. Side-scan sonar and multiyear bathymetry (high resolution seafloor
contour mapping) were among the instruments the ROV used. From this
expedition, scientists were surprised to discover a sand and cobble f loor,
Page 10
scattered with boulders. Brittle stars (pictured below) covered almost the
entire ocean floor. The data and images collected were used to predict
how rich the environment analyzed was. Scientists used this information
to compare it to other places in order to characterize the seaf loor
community.
(The image above is of bri ttlestars, which often gather on the soft
substrate.)
Page 11
The Phantom III S2 ROV (shown above) was about to be launched in
the Thunder Bay National Sanctuary. In this sanctuary lies over a
hundred shipwrecks. The ROV provided video documentation for the
understanding of preservation of submerged cultural resources. This
project was for outreach and educational purposes: the video was shown
to students, sparking interests in oceanography. The video greatly
fascinated students, and made them feel connected to the ocean, as if they
were actually in the ocean themselves. Also, the ROV documented the
settlement of the zebra mussels and round gobies, two invasive species.
In 1999, the locations of the spawning corals were observed. A ROV
was equipped with a video camera and skimmed the seafloor . A scientist
then used a multiyear bathymetry to track the ROV_s location. This was
the first time scientists were able to identi fy the location of the seafloor
features at the exact moment they were perceiving and videotaping them.
Sources:
NOAA Ocean Explorer: Sanctuary Quest
http://oceanexplorer.noaa.gov/explorations/02quest/sanctuaryquest.html
Page 12
CMG Multibeam Map Guides Sustainable Seas Exploration in Flower Garden Banks
National Marine Sanctuary
http://soundwaves.usgs.gov/1999/10/fieldwork.html
• NOAA Partnership Conducts Live Webcast from The Thunder Bay National Marine
Sanctuary and Underwater Preserve
http://www.nurp.noaa.gov/Spotlight%20Articles/underwaterweb.html
NOAA Ocean Explorer: Santuary Quest
http://oceanexplorer.noaa.gov/explorations/02quest/logs/jun16/jun16.html
8. Acknowledgements
We would like to thank Paul McGuinness for being our team
advisor, for being our supplier of tools and materials, and for keeping us
focused on the tasks at hand. Thanks to Peter Kerebruk from Draper
Laboratories for fitt ing propellers onto our thruster shafts. We appreciate
team graduate Thaddeus Stefanov-Wagner for bouncing our ideas back to
us and giving us feedback on our design. We would also like to thank the
Boston, MA Piledrivers Union and the Cambridge Public School
Department for their support of the team. Lastly, thanks to Nina
Alexander and Amanda Shing for giving us moral support and distracting
us when we got too serious and stressed, to Hummy Song, Paul Boudreau,
Catherine Liu, and Sara Gaynor for help with construction,
documentation and random comments, and to our parents for giving us
bui lding space and tolerating those late-night phone calls.
Page 13
9. Budget and Expenses
Expenses: McMaster (Tether) $46.00RadioShack (Cameras) $360.00Home Depot (Hardware) $150.00West Marine (Bilge Pumps) $100.00RadioShack (Control Box) $113.59Ace Hardware (Hardware) $50.00 $819.59 Funds: Existing from Last Year $500.00MATE $100.00School $300.00 $900.00
Page 14