rec.autos.simulators

The races went off as planned, with a small audience watching them
"live".  The robots are now on the ftp site, so anyone can race the same
cars on the same tracks on their own computers.

The overall winner was Randy Saint of Texas, U.S.A., with his latest
version of "Ramdu".  A close second was Jussi Pajala of Finland with
"WappuCar".  Third was Mike Inman of Florida, U.S.A., with "Indretti".
Safai Ma's "Burns" took fourth and Bill Benedict took fifth with "Rusty".

Wappucar was dominant on the tracks with the most curves and least
straights.  It won on stef2 and zandvort.  Ramdu won most of the races on
the other tracks, but not all.  The point totals after 12 races were:

    Ramdu       Wappucar        Indretti        Burns        Rusty
     96           86              45              37           24

The prizes have been spoken for.  Randy chose the "C++ Power Paradigms"
book by Mark Watson, Jussi chose Mark's other book, "Common LISP Modules:
Artificial Intell........), and Mike Inman got what he wanted, the CD-ROM
of Street Wizard maps and software.

Below are the detailed results as output by the host program for the six
races.  (i.e., a concatenation of the race.out files)  This is also the
order in which the races were run.  The authors of the various robots are:

Randy Saint     - Ramdu (Texas)
Jussi Pajala    - WappuCar (Finland, Wappu = Mayday)
Mike Inman      - Indretti (Florida)  
Safai Ma        - Burns (Canada)
Bill Benedict   - Rusty (U.S., at Rose-Hulman Tech)
Grant Reeve     - Grant1 (New Zeeland)
Oscar Gustavsson- OscCar2 (Sweden)
Tristrom Cooke  - Bingo (Australia)
Patrick Tierney - Heath (Australia)

All of these new robots are much better drivers than anything we've seen
before, with average speed increased by 5 to 15 percent!
----------------------------------------------------------------
9 cars for 15 laps.  The track was STEF2.trk.  The drivers were:
Bingo, Burns, Grant1, Heath, Indretti, OscCar2, Ramdu, Rusty, and WappuCar.

   results of race 1:
starting positions:
Ramdu, Heath, Burns, Indretti, Bingo, Grant1, Rusty, OscCar2, and WappuCar.
place 1  WappuCar   average speed 54.12 mph   10 points accumulated
place 2  Ramdu      average speed 52.23 mph   6 points accumulated
place 3  Indretti   average speed 50.4 mph   4 points accumulated
place 4  Burns      average speed 49.59 mph   3 points accumulated
place 5  Grant1     average speed 48.51 mph   2 points accumulated
place 6  Rusty      average speed 47.64 mph   1 points accumulated
place 7  OscCar2    average speed 46.98 mph   0 points accumulated
place 8  Heath      average speed 46.22 mph   0 points accumulated
place 9  Bingo      average speed 46.02 mph   0 points accumulated

   results of race 2:
starting positions:
WappuCar, OscCar2, Rusty, Grant1, Bingo, Indretti, Burns, Heath, and Ramdu.
place 1  WappuCar   average speed 53.38 mph   20 points accumulated
place 2  Ramdu      average speed 52.69 mph   12 points accumulated
place 3  Indretti   average speed 50.56 mph   8 points accumulated
place 4  Burns      average speed 49.84 mph   6 points accumulated
place 5  Grant1     average speed 48.41 mph   4 points accumulated
place 6  Rusty      average speed 47.61 mph   2 points accumulated
place 7  OscCar2    average speed 47.04 mph   0 points accumulated
place 8  Heath      average speed 46.42 mph   0 points accumulated
place 9  Bingo      average speed 45.78 mph   0 points accumulated
----------------------------------------------------------------

9 cars for 20 laps.  The track was OVAL2.trk.  The drivers were:
Bingo, Burns, Grant1, Heath, Indretti, OscCar2, Ramdu, Rusty, and WappuCar.

   results of race 1:
starting positions:
Bingo, WappuCar, Ramdu, Rusty, Grant1, OscCar2, Burns, Indretti, and Heath.
place 1  Ramdu      average speed 70.83 mph   10 points accumulated
place 2  Indretti   average speed 70.31 mph   6 points accumulated
place 3  Burns      average speed 69.65 mph   4 points accumulated
place 4  WappuCar   average speed 69.39 mph   3 points accumulated
place 5  Heath      average speed 69.05 mph   2 points accumulated
place 6  OscCar2    average speed 68.64 mph   1 points accumulated
place 7  Grant1     average speed 67.45 mph   0 points accumulated
place 8  Rusty      average speed 67.31 mph   0 points accumulated
place 9  Bingo      average speed 66.61 mph   0 points accumulated

   results of race 2:
starting positions:
Heath, Indretti, Burns, OscCar2, Grant1, Rusty, Ramdu, WappuCar, and Bingo.
place 1  Indretti   average speed 70.86 mph   16 points accumulated
place 2  Ramdu      average speed 70.62 mph   16 points accumulated
place 3  Burns      average speed 69.24 mph   8 points accumulated
place 4  WappuCar   average speed 69.03 mph   6 points accumulated
place 5  OscCar2    average speed 68.35 mph   3 points accumulated
place 6  Grant1     average speed 67.36 mph   1 points accumulated
place 7  Heath      average speed 66.77 mph   2 points accumulated
place 8  Bingo      average speed 66.62 mph   0 points accumulated
place 9  Rusty      average speed 66.77 mph   0 points accumulated
----------------------------------------------------------------

9 cars for 15 laps.  The track was V03.trk.  The drivers were:
Bingo, Burns, Grant1, Heath, Indretti, OscCar2, Ramdu, Rusty, and WappuCar.

   results of race 1:
starting positions:
Heath, Rusty, Indretti, WappuCar, Burns, Bingo, OscCar2, Ramdu, and Grant1.
place 1  Ramdu      average speed 71.29 mph   10 points accumulated
place 2  WappuCar   average speed 68.41 mph   6 points accumulated
place 3  Burns      average speed 67.3 mph   4 points accumulated
place 4  Indretti   average speed 67.26 mph   3 points accumulated
place 5  Grant1     average speed 63.58 mph   2 points accumulated
place 6  OscCar2    average speed 62.28 mph   1 points accumulated
place 7  Rusty      average speed 61.98 mph   0 points accumulated
place 8  Bingo      average speed 61.4 mph   0 points accumulated
place 9  Heath      average speed 61.28 mph   0 points accumulated

   results of race 2:
starting positions:
Grant1, Ramdu, OscCar2, Bingo, Burns, WappuCar, Indretti, Rusty, and Heath.
place 1  Ramdu      average speed 72.06 mph   20 points accumulated
place 2  WappuCar   average speed 68.16 mph   12 points accumulated
place 3  Indretti   average speed 67.22 mph   7 points accumulated
place 4  Burns      average speed 67.42 mph   7 points accumulated
place 5  Grant1     average speed 63.63 mph   4 points accumulated
place 6  Rusty      average speed 63 mph   1 points accumulated
place 7  Heath      average speed 62.58 mph   0 points accumulated
place 8  OscCar2    average speed 62.21 mph   1 points accumulated
place 9  Bingo      average speed 61.72 mph   0 points accumulated
----------------------------------------------------------------

9 cars for 8 laps.  The track was anew.trk.  The drivers were:
Bingo, Burns, Grant1, Heath, Indretti, OscCar2, Ramdu, Rusty, and WappuCar.

   results of race 1:
starting positions:
Rusty, OscCar2, Bingo, Heath, WappuCar, Indretti, Ramdu, Burns, and Grant1.
place 1  Ramdu      average speed 85.32 mph   10 points accumulated
place 2  WappuCar   average speed 82.94 mph   6 points accumulated
place 3  Burns      average speed 81.68 mph   4 points accumulated
place 4  Indretti   average speed 79.93 mph   3 points accumulated
place 5  Rusty      average speed 79.83 mph   2 points accumulated
place 6  Grant1     average speed 78.31 mph   1 points accumulated
place 7  OscCar2    average speed 78.17 mph   0 points accumulated
place 8  Heath      average speed 77.65 mph   0 points accumulated
place 9  Bingo      average speed 76.45 mph   0 points accumulated

   results of race 2:
starting positions:
Grant1, Burns, Ramdu, Indretti, WappuCar, Heath, Bingo, OscCar2, and Rusty.
place 1  Ramdu      average speed 85.08 mph   20 points accumulated
place 2  WappuCar   average speed 81.55 mph   12 points accumulated
place 3  Burns      average speed 81.25 mph   8 points accumulated
place 4  Rusty      average speed 79.67 mph   5 points accumulated
place 5  Indretti   average speed 78.94 mph   5 points accumulated
place 6  OscCar2    average speed 78.12 mph   1 points accumulated
place 7  Grant1     average speed 78 mph   1 points accumulated
place 8  Heath      average speed 77.11 mph   0 points accumulated
place 9  Bingo      average speed 75.91 mph   0 points accumulated
----------------------------------------------------------------

9 cars for 10 laps.  The track was speed2.trk.  The drivers were:
Bingo, Burns, Grant1, Heath, Indretti, OscCar2, Ramdu, Rusty, and WappuCar.

   results of race 1:
starting positions:
OscCar2, Indretti, Burns, Bingo, Heath, Ramdu, WappuCar, Rusty, and Grant1.
place 1  Ramdu      average speed 107.1 mph   10 points accumulated
place 2  WappuCar   average speed 102.73 mph   6 points accumulated
place 3  Rusty      average speed 98.62 mph   4 points accumulated
place 4  Grant1     average speed 95.92 mph   3 points accumulated
place 5  Indretti   average speed 96.19 mph   2 points accumulated
place 6  Burns      average speed 94.47 mph   1 points accumulated
place 7  OscCar2    average speed 93.8 mph   0 points accumulated
place 8  Bingo      average speed 89.45 mph   0 points accumulated
place 9  Heath      average speed 89.08 mph   0 points accumulated

   results of race 2:
starting positions:
Grant1, Rusty, WappuCar, Ramdu, Heath, Bingo, Burns, Indretti, and OscCar2.
place 1  WappuCar   average speed 105.62 mph   16 points accumulated
place 2  Ramdu      average speed 103.56 mph   16 points accumulated
place 3  Rusty      average speed 99.09 mph   8 points accumulated
place 4  Indretti   average speed 97.38 mph   5 points accumulated
place 5  Grant1     average speed 95.23 mph   5 points accumulated
place 6  Burns      average speed 94.64 mph   2 points accumulated
place 7  OscCar2    average speed 94.16 mph   0 points accumulated
place 8  Bingo      average speed 89.46 mph
...

read more »

Thanks in advance,
        -Colin-

: Thanks in advance,
:       -Colin-

The Robot Auto Racing Simulation (RARS) is a simulation of auto racing
in which the cars are driven by robots.  Its purpose is two-fold: to
serve as a vehicle for Artificial Intelligence development and as a
recreational competition among software authors.  The host software,
including source, is available at no charge.  It currently runs under
MSDOS and UNIX, including Linux, and on the Amiga. This announcement
introduces the sixth release of the software, which we call version
0.50.  New features of ver. 0.50 are summarized below.

The April 30 races will be run on version 0.50.  It has been tested
to assure that the robots have the same lap times as they do on
version 0.39.  (Get race.ann from the ftp site, see below.)

                The Robot Auto Racing Simulation

            A Challenge for Evolutionary Programming

                 A Competition for Programmers

                      by Mitchell E. Timin

Version 0.50 of RARS is ready.  It is written in C++.  The DOS
version was compiled and tested with Borland C++ ver. 3.1.  It should
be easy to port to any C++ that has functions for drawing lines and
arcs, a flood-fill or color fill function, and text output to the
graphic screen.  Hence it will not be difficult to make it run on a
Macintosh or Windows system.  C++ is not required for the robot
"driver" programs.

New features introduced with version 0.45 were:
- The drivers which are to race can be named on the command line.
- The computer can generate a random starting order.
- The race can be run with no graphics display.
- A series of races can be run, on the same track with the same drivers.
- A report is generated in a file, giving the results of all races in
  a series.  Points are accumulated using the F1 scoring system.
- Degrees or radians may be used in any track definition file.
- Clockwise tracks will work now.

Version 0.50 introduces a digital "instrument panel".  Any of the racers may
be cursor-selected using the up & down arrow keys.  Six key variables
for the selected car are shown during the race, updating every .25 sec.

The race tracks are defined by ASCII files.  Many pre-defined tracks
are supplied.  The desired track is named on the command line. Users
can create their own tracks using any text editor, although this is
not a trivial process unless a CAD program is used to find the exact
lengths, angles, and radii for the track segments.  It is possible to
do it by trial and error, however.

There is a RARS anonymous ftp site: magdanoz.mcafee.com in
directory /bin/ftp/rars.  Anyone can get any RARS stuff there,
code, announcements, car controllers, documentation, and tracks.

There is a listserver so that interested parties may discuss
RARS by e-mail.  To subscribe to the list service send e-mail to

    subscribe rars-list
(To unsubscribe, send "unsubscribe rars-list" to the same address.)

We have a page for the www at http://www.racesimcentral.net/~bsr/rars.html.
Ben Rometsch in England is taking care of the www part of the project.

another web page maintained by Dave Gymer, also in England.  Its URL is:

RARS consists of a simulation of the physics of cars racing on a
track, with a simple bird's-eye view of the race.  The unique feature
is that each car is controlled by a separate and independent control
program.  Each car is "driven" by its own control program, which
receives information from the simulation telling it about the car's
local situation.  The "driver" (control program) adjusts the steering
and throttle, and then the physics simulation moves the car a little.
This happens many times per second, of course.  Every car has exactly
the same physical characteristics, only the "drivers" are different.
Hence, the result is a competition between the control programs.
Furthermore, the competition is visible as an auto race, with
acceleration, passing, cornering, braking, etc.

It is intended that many users will write their own robot "drivers".
Thir*** similar, but not identical, examples are supplied.  These
are meant to serve as examples for programmers wishing to develop
their own.  The control programs may be written in other languages if
they are linker-compatible on the the intended platform.  To date
the robot "driver" programs have used C++ or ANSI C.

For genetic programming, the races will be between several programs
selected from an evolving population of programs.  The racing may
take place continuously for long periods of time, with the graphic
display disabled for faster execution.  Of course losers will be
eliminated and winners will breed.  Genetic Algorithm proponents will
probably design robot drivers with a vector of parameters to be
determined by evolution.  Neural nets are also candidate "drivers".
It will be up to the experimenter to decide if human-designed robots
are allowed to compete with the evolving population.

Wanted - People to do or help with any of the following:

     Porting to other platforms
     Testing the software and suggesting enhancements
     Improving the graphics
     Locating a corporate or university sponsor
     Act as a race director to manage a "race meet"
     Reporting, both to academic journals and popular magazines
     Improving the software in any of dozens of ways
     Adding sound effects
     (and of course, building "drivers" to compete in the races!)

: Thanks in advance,
:       -Colin-

                       RARS Version 0.60

            A Challenge for Evolutionary Programming

                 A Competition for Programmers

                      by Mitchell E. Timin

The Robot Auto Racing Simulation (RARS) is a simulation of auto racing
in which the cars are driven by robots.  Its purpose is two-fold: to
serve as a vehicle for Artificial Intelligence development and as a
recreational competition among software authors.  The host software,
including source, is available at no charge.  It currently runs under
MSDOS and UNIX, including Linux, and on the Macintosh and Amiga. This
announcement introduces the seventh release of the software, which we
call version 0.60.  RARS is written in C++.  The DOS version was
compiled and tested with Borland C++ ver. 3.1.  C++ is not required
for the robot "driver" programs.

The race tracks are defined by ASCII files.  Many pre-defined tracks
are supplied.  The desired track is named on the command line. Users
can create their own tracks using any text editor, although this is
not a trivial process unless a CAD program is used to find the exact
lengths, angles, and radii for the track segments.  It is possible to
do it by trial and error, however.

There is a RARS anonymous ftp site: magdanoz.mcafee.com in
directory /bin/ftp/rars.  Anyone can get any RARS stuff there,
code, announcements, car controllers, documentation, and tracks.

The principal file that you probably want is rars060.zip.  Also,
check out the tracks, robots, and digest sub-directories.

There is a listserver so that interested parties may discuss
RARS by e-mail.  To subscribe to the list service send e-mail to

    subscribe rars-list
(To unsubscribe, send "unsubscribe rars-list" to the same address.)

We have three locations on the www:
http://www.cs.bham.ac.uk/~bsr/rars.html
http://www.mal.com/~dgymer/rars/
http://www.cris.com/~Nims/mi/mi_rars.html

RARS consists of a simulation of the physics of cars racing on a
track, with a simple bird's-eye view of the race.  The unique feature
is that each car is controlled by a separate and independent control
program.  Each car is "driven" by its own control program, which
receives information from the simulation telling it about the car's
local situation.  The "driver" (control program) adjusts the steering
and throttle, and then the physics simulation moves the car a little.
This happens many times per second, of course.  Every car has exactly
the same physical characteristics, only the "drivers" are different.
Hence, the result is a competition between the control programs.
Furthermore, the competition is visible as an auto race, with
acceleration, passing, cornering, braking, etc.

It is intended that many users will write their own robot "drivers".
Many robots are supplied.  These are meant to serve as examples for
programmers wishing to develop their own.  The control programs may
be written in other languages if they are linker-compatible on the
intended platform.  To date the robot "driver" programs have used
C++ or ANSI C.

For genetic programming, the races will be between several programs
selected from an evolving population of programs.  The racing may
take place continuously for long periods of time, with the graphic
display disabled for faster execution.  Of course losers will be
eliminated and winners will breed.  Genetic Algorithm proponents will
probably design robot drivers with a vector of parameters to be
determined by evolution.  Neural nets are also candidate "drivers".
It will be up to the experimenter to decide if human-designed robots
are allowed to compete with the evolving population.

New features introduced with version 0.45 were:
- The drivers which are to race can be named on the command line.
- The computer can generate a random starting order.
- The race can be run with no graphics display.
- A series of races can be run, on the same track with the same drivers.
- A report is generated in a file, giving the results of all races in
  a series.  Points are accumulated using the F1 scoring system.
- Degrees or radians may be used in any track definition file.
- Clockwise tracks will work now.

Version 0.50 introduced these features:
- "instrument panel" showing how a selected car is behaving in real-time
- the F and S keys speed up or slow down the simulation during a race
- several additional command line options
- a portable draw_arc function for drawing the track rails

Version 0.60 added many items.  Among them are:
- Fuel consumption and variable mass, with lost time for re-fueling
- command line option to enter the RVG seed, to replay a race exactly
- race report has laps completed, damage and fuel for each car
- optional pre-race practice, for robots which can learn
- The file "WHATNEW.DOC" details all the differences w.r.t. ver. 0.50.

Wanted - People to do or help with any of the following:

     Porting to MS-Windows
     Testing the software and suggesting enhancements
     Improving the graphics
     Locating a corporate or university sponsor
     Act as a race director to manage a "race meet"
     Reporting, both to academic journals and popular magazines
     Improving the software in any of dozens of ways
     Adding sound effects
     (and of course, building "drivers" to compete in the races!)