Re: Great opportunity for Ada application

["James S. Rogers" <jimmaureenrogers@worldnet.att.net>]

"David Marceau" <davidmarceau@sympatico.ca> wrote in message
news:3E4AAD59.BB77642@sympatico.ca...
> It seems the best horses are on 1)Lockheed Martin and 2)Boeing without
> any competition.  I saw some fly by wire stuff uav/utv on tv the other
> night on discovery concerning the making of the X-planes.
>
> Whatever is competing, I think the config with the most strategic impact
> will win and not the cheapest to mass-produce price.
> I also think the config with the free non-used non-competition bonus
> features will also add value if there is a tie :)  The rules are not
> clear as to what happens if there is a tie.
>
> Notice the section of COMPETITION AND REPAIRS:
> "
> Competition
> The goal of this competition is for the race vehicle to complete the
> course as fast as possible.
> Interfering with other race vehicles will not be allowed.  The following
> will result in the
> disqualification of a contestant:
> Intentionally interfering with competing Race or Chase vehicles
> Intentionally interfering with competing operators
> Intentionally interfering, spoofing, jamming, or disturbing
> communications used by
> competitors or the race organizers
> Intentionally altering the terrain with the purpose of making it more
> difficult for
> competitors
>
> Refueling and Repairs
> Autonomous refueling and repairs will be allowed at a checkpoint.  That
> is, equipment for
> refurbishing the race vehicle must be set-up at the check point prior to
> the start of the race.  Other
> than the E-Stop, no human (or other biological entity) may send commands
> to or otherwise control
> the equipment once the race begins.
> "
> Obviously fuel cells are important here :)
>
> Can't interfere with other cars....hmmm I thought all is fair in love
> and war.
> Zapping another car before, during and after the competition without
> getting caught seems to be still legal.
> It's like doping in the olympics :)  All the intention disqualification
> capabilities could be considered as an added-value bonus feature if
> there is a tie.  If there is a tie, these added features could prove
> their worth in the second round :)
>
> A Top of the line utv system
> ----------------------------
> -spark, openravenscar(pedal to the metal solution foggy for me but I'm
> sure ada gurus out there will opt for this)
> -uav-lets or robofly-lets spurting out of the atv all individually
> monitored via the atv to web-front-end.  This could be critical for the
> last five seconds of allowed off-ground uav time if we consider a
> uav-let/robofly as part of the uav :)  Nothing stops us from propelling
> the uav-let as a projectile for the last five seconds to get to the
> destination target :)
>
> A not so thorough vision of what a econo-system could have
> ----------------------------------------------------------
> aws(web front end).  The more creative/more value/better looking, the
> more probably the ada team will win.  I have ideas up the yin yang for
> this.(although most communication is not allowed does not mean it cannot
> be implemented and not used during the competition).
> or gtkada(non-web front end)
> ada2nqc(control on-vehicle robot adamindstorm for wheel turning and
> other motor controls maybe)
> ada2nqc(maybe also for the wireless communication between on-vehicle and
> web-front-end)
> A#(.NET)/jgnat/j2me, adaCOM (control wireless communication between
> on-vehicle and web-front-end) mix of commercial off the shelf and
> open-source here maybe :)
> Obviously other sensors/actuators will be here to.
>
> That said I'll volunteer and contribute to any solution an ada team
> would propose and learn from it as a gesture of good will to get an ada
> coder job as an end result.  One catch: I am Canadian citizen.

Having done some Ada work on teleoperated ground vehicles I
will venture the following opinions:

UAV experience will help (such as that available to Lockheed and
Boeing) but is not critical.

It will take more than one year to develop the software and hardware
for such an effort. Note that the primary difficulty is not in the
vehicle control software. That can be handled relatively easily.
The difficulty is in the autonomous control system generating the
vehicle control inputs, and reading the vehicle status outputs.

Sensors needed for this include (as a minimum) a gps receiver,
video sensors to determine the edge of a road or path,
ladar or sonor detector to identify obstacles, and a pan/tilt/bearing
sensor to determine vehicle orientation.

You need to be able to accurately describe the course to the
vehicle, giving it enough intelligence to find an optimal path
through the terrain. You need to be able to accurately identify
the vehicle's current position at all times during the race.
You will need to employ a Kalman filter merging all input data
(position, heading, terrain, obstacles, fuel consumption, engine
temperature, etc.) to determine best possible course.

I assume this will be an off-road race to avoid hazards with
non-race participants. If not, you must be able to program
your vehicle to obey all applicable traffic rules and regulations.

You will need to determine how you are going to control and
instrument your vehicle. Most existing ground vehicles are not
yet completely drive by wire. This means you may have to
install and calibrate actuators. Actuators that have been proven
for ground vehicle control are available from Omnitech Robotics,
Inc. Those actuators communicate using the CAN (Controller
Area Network) bus developed by Bosch in Germany.

You will need time to test your system. There are a lot of
variables to get right in an autonomous control system.

Your vehicle will probably need a reliable 4 wheel drive an
at least a 30 gallon gas tank. Your vehicle is allowed to
continuously report to you during the race. This would be
a "good idea". You cannot send it commands except for
the E-Stop (emergency stop) command, which is included
for safety reasons. Note that any repair or refuling must be
handled autonomously. This provides a challenge. Your
autonomous system may need to turn off the engine
during refuling and repairs. It will need to continue enough
capability to restart the engine at the end of the pit stop.
The vehicle must be able to sense end of pit stop without
a discrete signal being sent from a biological entity.
This means that your checkpoint must have the ability
to signal the vehicle without your direct action. Of course
it would be best if you can avoid a pit stop altogether.

Anticipate a lot of dust, high winds, rough terrain and hot
weather for this race.

I estimate that the materials needed to create and operate
this vehicle will cost at least $200,000.00. This cost does
not include engineering effort or manufacturing/assembly time.

Jim Rogers