Subject:
|
Robot contest
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Mon, 16 Jun 1997 11:15:39 GMT
|
Original-From:
|
jmendel141@aol.comSTOPSPAMMERS
|
Viewed:
|
1587 times
|
| |
|
|
Trinity College Fire-Fighting Home Robot Contest
Miscellaneous comments and feedback to questions
***********************************************
Dear contest participant and/or spectator
I am sending you this general message in response to your comments and
feedback about the contest.
After every contest there is always a lot of discussion concerning the
contest, the rules and how to make the event even better. This year there was
more than usual of this type of discussion and I take that as a good sign of
the interest that people have in the contest and their desire to help make it
even better. I greatly appreciate all the feedback and suggestions as
evidenced by the changes I have made in the rules over the years in response
to the comments.
-Jake Mendelssohn
contest coordinator
*************************************************
WHY HAVE A CONTEST?
Because of the large number and diversity of comments that came in this year,
I think that it might be appropriate that I explain what my concept is of the
contest.
I originally developed this concept of a contest because I was rather
disappointed with the slow progress that I thought Robotics was making. 20
years ago I started a Robotics company because I really believed that the
"Age of Robotics" was just around the corner. The computer people had had
their revolution and I thought that Robotics was next. Well, it turns out
that Robotics is many orders of magnitude more difficult than computers, thus
the need for this contest.
The goals I have set for the contest are:
1. Provide an incentive for the Robotics community to develop what I hoped
would be a practical application for a real-world Robot. This "Killer App"
would be the means by which Robotics would finally break out and become a
useful technology in everyday life.
2. Promote the technology by giving an inducement for Robotists to try new
techniques to solve a real-world practical problem
3. Demonstrate to government, business and the public that this technology
is attainable and can have practical applications and should be encouraged
and supported.
4. Encourage younger students to get involved in Robotics because I truly
believe that the next breakthrough will not be made by anyone my age, but by
the students still in college and high school. I also believe that building a
Robot is a great educational exercise for a student by giving them hands-on
experience in physics, mechanics, hardware, software, logic. They will also
have to learn how to overcome failure and to plan and organize a long term,
multifaceted project. Robotics is applied everything and there is no better
learning tool that building a Robot.
I designed the Fire-Fighting Home Robot Contest, because putting out a fire
was something that everyone could relate to and appreciate the need for. It
was an activity that had movement, action and suspense. It could be
understood by the layman and yet was intriguing enough to entice the expert.
I believe in incrementalism and do not think that the first practical
application of this technology actually be used in a home. The home
environment is too crowded, variable and the average home owner could
probably not afford the first units that were built. Just like the first
heavier-then-air airplane built by the Wright brother had no real practical
application since it only flew a 150 feet, the first example of this
technology will not look very impressive to the casual observer. But the
future belongs to the person who can look beyond the fog of today and see the
horizon of tomorrow.
Thus even though I call this a HOME Robot Contest I believe that the first
applications will be used in warehouse situations. A warehouse usually
involves straight wide corridors with only occasional obstacles. The cost of
not putting out quickly a small warehouse fire can run into the millions.
Traditional protective fire fighting techniques cost enormous amounts to
install and many can actually do more damage to the stored materials than the
average fire might. This provides a great incentive for the company to invest
in this Robotic technology. After a few years of running successfully in
warehouses, this technology could then move into the more complicated office
building setting. Finally this technology will be efficient enough, effective
enough and cheap enough to move into the home.
However, the first step is that someone needs to be able to make it work. In
the past few years since we have started this contest, I have seen tremendous
advances in the effectiveness of the Robots in the contest. In the first
year, although we had 26 entries, only 3 entries put out fire twice and no
Robot put out the fire all three times. This year there were 10 entrees that
put out the fire 2 or more times even when running in the more difficult
furniture mode. I believe that is a few years some Robot will be good enough
to win the contest and then go right into production as a commercial unit.
***********************************************
THE MOST COMMONLY ASKED QUESTIONS
WHY DOES THE ARENA HAVE ONLY 90 DEGREE ANGLES?
I agree that having only 90 degree angles in the arena is very artificial and
simplistic, but it is also a good representation of the real world where our
buildings are mostly 90 degrees and very artificial. Also remember that I
believe that the first actual use of these sorts of Robots will be in a
warehouse situation which are very "90 degree-ish". Besides, it is hard
enough to get these devices to work in this "easy", "simplistic" and
"artificial" environment, putting in 56 degree turn would be even harder.
When we reach the day when most of the Robots have no trouble with a 90
degree turn, then I will consider using other turns.
WHY IS THE ARENA ALL ON ONE LEVEL? WHY NOT HAVE STAIRS?
Practically the same answers as to the previous question. While houses do
have multi-levels, warehouses do not. And once again it is hard enough to
move on one level. Putting in stairs would make it practically impossible.
WHY NOT MAKE IT A TRUE MAZE WITH THE ROOMS AND WALLS UNKNOWN UNTIL THE
CONTEST TIME?
I do not envision that these Robots will first be used to go into unknown
areas. These Robots will be permanently stationed to protect a specific
warehouse or later, a specific home. The floor plan of that warehouse or home
is a known quantity and can be put into the Robot's programming, if
necessary. Once again, not every Robots design needs these kind of a priori
knowledge, but many do, and we are trying to get the widest participation
possible.
WHY HAVE A RANDOM PLACEMENT OF THE CANDLE WHICH RESULTS IN SOME ROBOTS HAVING
AN "EASIER" TASK THEN OTHERS?
It a real world situation the location of the fire would be unknown and
somewhat "random". A real fire-fighting Robot would have to prove its ability
to search for the fire, rather then just go to it. If the location of the
candle were known to the Robot before it started, it would be a rather easy
task just to go directly to the correct room and flood it with CO2. The goal
of this contest is not to make a Robot that can go to a specific place, but
to make a Robot that can find and extinguish a candle. That is quite a
different matter and a much harder one at that. But once you do it you have a
really valuable device.
Yes, there is a bit of unfairness in that some Robots may accidentally get
"harder" rooms than others, but we try to even things out somewhat by taking
the 2 best of the 3 trials.
WHAT ARE THE ROBOT PENALIZED FOR HITTING THE WALL?
In a real world situation, moving through a house by ricocheting off the
walls is not a very practical means of locomotion. Yes, it might work in a
few situations and yes, it does work in the mostly sterile environment of
this contest, but in the real world, it is very impractical. In a real-world
warehouse with its stacks of merchandise, this is not a realistic way to
travel.
WHY IS DEAD RECKONING ALLOWED SINCE IT IS NOT A VERY REAL-WORLD WAY TO MOVE?
Yes, this is correct, dead reckoning (the adding of steps and turns to your
original position in order to calculate your new position) is not a very
effective way to maneuver in the real world. The reason it is allowed in the
contest is because it is the easiest way for a beginner to maneuver their
Robot. If we make the initial contest requirements too high, then novices
would never be able to enter. In the 1998 contest however, there will be a
special option which will give advantages to Robots that can truly determine
their position in the arena without the use of dead-reckoning.
LARGER ROBOTS WOULD BE EASIER TO SEE AND WOULD BE MORE FUN. WHY CAN'T THE
ARENA BE MADE BIGGER?
The goal of this contest is to demonstrate the application of a technology.
This is done just as easily with a small Robot as with a large one. I agree
that a larger arena and large Robots would have certain advantages, but it
would also dramatically increase the cost of building a Robot. It is
expensive enough to enter the contest and most people without corporate or
university funding would find the cost of a large Robot prohibitive.
MY ROBOT CAN DO "X" SO WHY CAN'T THE CONTEST RULES BE CHANGED TO ACCOMMODATE
ME AND MY ROBOT?
I get this question or some variation of it, all the time. The "X" in the
question is usually some very specific thing like wanting mirrors hung above
the arena so the Robot can look up to see the reflection of the candle, or
wanting reach room pained a different color, or having clear walls in the
rooms. The general answer to all these questions is that it is not a
situation that exists commonly in the real world and we are trying to build
Robots that can adapt to the real world and not visa versa.
WHY IS SPEED A CONSIDERATION FOR JUDGING THE WINNING ROBOT?
Obviously in putting out a fire, speed is important, but it is even more
important in making the Robot contest interesting and fun to watch. It is
very painful to watch a Robot take the full 6 minutes to find and extinguish
a candle. However, reliability is also important and the current rules also
encourage reliability.
WHAT CHANGES AND MODIFICATIONS ARE PLANNED FOR THE FUTURE?
We are working on that right now and you will see a lot of the suggestions I
received incorporated into the modified rules to make the contest even
better. I really appreciate and depend on all of your comments and
suggestions. The contest has evolved and changed for the better each and
every year because of the feedback I have gotten from all of you. Thank you
and keep up the good work!
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
