khong lim Yap | 3 Oct 12:18 2014

How to show the path driven by bots?


i need some guidance on how to turn on path that driven by bots. Thanks
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
Torcs-devel mailing list
Torcs-devel <at>
Mastah_Killah | 23 Sep 18:49 2014

Torcs robot removed from the race.

I've programmed a bot, that is suppossed to drive slow (30 km/h).
After about 5 min : 50 sec, the bot is removed from the race,
I suppose because it is so slow. It is not because of damage or fuel.
Does anybody know how I can remove this limitation?
I'm using torcs 1.3.4. Thanks.
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
Torcs-devel mailing list
Torcs-devel <at>
Michael Xynidis | 4 May 21:16 2014

Have Oculus Rift development kit - Need 360 degree VR

Hi. I've been planning a study on steering systems, and up until now, have been using TORCS. The associated dissertation will include operating a car in reverse--which requires a VR space. I'm doing this research at the University of Central Florida. I've gotten one Oculus Rift kit, and may be able to obtain a second. Ticket #822 (mungewell/rvlander) mentions the need for a dev kit in order to expand SD to support VR via the Rift. I've got one and would like to discuss the issue further with the individuals involved.

Hopefully, moving from the TORCS platform to SD won’t be an issue. The required customizations aren't very complicated: a test track with cones (collision-detecting stanchions) for weaving through, and one-or-two cars where the speed can be fixed at a predetermined rate once entering the slalom. The VR part will essentially be a parallel-parking test into a fixed space. We've already been working at modifying the steering input on the TORCS car--wouldn't expect it to be much different in SD.

Is there anyone I can work with on this? Or, at least, discuss the possibility of Oculus Rift support? Thanks. Looking forward to a good discussion.



Is your legacy SCM system holding you back? Join Perforce May 7 to find out:
&#149; 3 signs your SCM is hindering your productivity
&#149; Requirements for releasing software faster
&#149; Expert tips and advice for migrating your SCM now
Torcs-devel mailing list
Torcs-devel <at>
ravich kurchatov | 20 Apr 20:22 2014

My experience implementing a lidar in torcs (part 1)


This is my experience implementing a lidar in torcs, which I hope will
be useful to anyone following a similar path. 

Partly inspired by Brad Templeton's essay
( on the need
for an open source driving simulator, I decided to investigate more
about bringing this vision to frutition, as a spare time project.

As most roboticists (e.g. Sebastian Thurn in the udacity course) point
out, the major challenge for a robocar is perception. Once a robocar
knows where the road is, where *it* is located and where the other
cars and obstacles are, it is relatively easy to decide what to do and
signal actuators. This insight is validated by how good the built in
TORCS robots are, which have access to all this information. Almost
all of them manage to soundly kick my ass on every racetrack (I'm
playing with a keyboard, maybe I'd do better if I played with a wheel
or joystick).

Simulated Car Racing (SCR) is a good effort to make the robots more
realistic, but only goes a small distance. The track and opponent
sensors that SCR provides (via UDP) are still a bit too much spoon
feeding. Realistic sensors (such as GPS, Inertial Measurement Unit
(IMU), lidar, radar, stereo cameras) used by real self driving cars
are all missing.

The good news is that most of these sensors are easy to implement. GPS
and IMU are trivial as car position and acceleration are directly
available via the `tCarElt* car` class/struct, viz.:


All that is needed for more realism is to add Gaussian noise (with
proper sigma) and random dropouts for the GPS and simulated drift for
the IMU (if it is too accurate).

It is also fortunate that a simple lidar, such as a single beam SICK
type is easy to implement in a few lines of code, using plib's ssgLOS
(line of sight) function (

#+begin_src cpp
void lidar(const tCarElt *car, sgVec3 pos, sgVec3 l_dir, sgVec4 xyzr)
  // pos:   position of the sensor relative to the car->_postMat
  // l_dir: direction in which laser points relative to the sensor

  sgMat4 m;      // sensor position matrix
  sgCopyMat4(m, car->_posMat);
  m[3][0] += pos[0]; m[3][1] += pos[1]; m[3][2] += pos[2]; 
  sgMat4 inv_m;  // inverse of sensor position matrix
  sgTransposeNegateMat4(inv_m, m);

  ssgHit *results;
  int num_hits = ssgLOS(TheScene, l_dir, inv_m, &results);

  if (num_hits == 0) {
    xyzr[3] = 0;

  ssgHit *hit;
  // range initialized to a large number
  float range = 1E9, curr_range;
  for (int i = 0; i < num_hits; ++i) {
    hit = &results[i];
	// divide-by-zero check omitted on purpose
    curr_range = hit->plane[3] / sgScalarProductVec3(hit->plane, l_dir);

    if (curr_range < range)
      range = curr_range;

  sgVec3 xyz;
  sgCopyVec3(xyz, l_dir);
  sgScaleVec3(xyz, range);
  sgXformPnt3(xyz, xyz, m);

  xyzr[0] = xyz[0];
  xyzr[1] = xyz[1];
  xyzr[2] = xyz[2];
  xyzr[3] = range; 

If there is sufficient interest, I'll soon post a part 2 going into an
explanation for the code and how I plan to extend it to simulate a
more sophisticated multi beam lidar such as the Velodyne HDL-64 (which
was extensively used in Darpa GC 2007 and is the main sensor of the
Google SDC).


Start Your Social Network Today - Download eXo Platform
Build your Enterprise Intranet with eXo Platform Software
Java Based Open Source Intranet - Social, Extensible, Cloud Ready
Get Started Now And Turn Your Intranet Into A Collaboration Platform
Alan Ali | 18 Jan 00:22 2014

Initial car positions


Would you please tell me where the initial position (x and y) of the car is initialized?
I can find that the initial inter vehicle distances are equal to 5 meter and i want to change this value, how can i do it?

Thank a lot in advance.

CenturyLink Cloud: The Leader in Enterprise Cloud Services.
Learn Why More Businesses Are Choosing CenturyLink Cloud For
Critical Workloads, Development Environments & Everything In Between.
Get a Quote or Start a Free Trial Today.
Torcs-devel mailing list
Torcs-devel <at>
strandhau biz | 7 Oct 15:15 2013

[Torcs-users] Bug-report regarding torcs-1.3.4 installation and robotgen -- including fixes


I just tried to use the PATCH send by Jasper
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
Torcs-devel mailing list
Torcs-devel <at>
Bernhard Wymann | 4 Aug 13:33 2013

Changes in wheel.cpp, please review/test/comment

Hi all

I changed the wheel forces calculation a bit, I would be happy if you 
review and test the changes. Changelog:
- Fixed car flying off with extreme suspension settings, see
- Fixed applied force when wheel has lost touch with the ground but 
suspension is not fully extended.

Best regards


Get your SQL database under version control now!
Version control is standard for application code, but databases havent 
caught up. So what steps can you take to put your SQL databases under 
version control? Why should you start doing it? Read more to find out.
Michael Sieben | 14 Jun 10:34 2013

Debugging in Visual Studio 2012


i have a question to some more experienced users of Visual Studio.

I got Torcs 1.3.4 and my own robots to compile in VS and can run wtorcs.exe from runtime and/or runtimed.

But is it possible to modify the configuration so i could run and debug torcs with VS? (e.g. use breakpoints, step by step, and inspect variables and classes directly with the VS builtin functions)

Thanks in advance to any help.

best regards
This email is sponsored by Windows:

Build for Windows Store.
Torcs-devel mailing list
Torcs-devel <at>
jlwalter | 22 Apr 05:37 2013

AI research - reinforcement learning

Hello everyone,

My name is Jeremiah and I'm starting a research project on Reinforcement Learning testing different algorithms' effectiveness at 'learning to drive', or dealing with continuous actions. And I'm going to use TORCS as my simulation!

So my question is, has anyone done any similar Reinforcement Learning research and have you any tips or recommendations when it comes to setting the simulation environment up?
I appreciate any help I can get!

Jeremiah Walter
BComp, University of Tasmania
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developers can use
our toolset for easy data analysis & visualization. Get a free account!
Torcs-devel mailing list
Torcs-devel <at>
Daniele Loiacono | 17 Apr 22:48 2013

[GECCO 2013] Simulated Car Racing Competition

Our apologies if you receive multiple copies of this announcement.

               Simulated Car Racing Competition

                  to be held as part of the

  2013 Genetic and Evolutionary Computation Conference (GECCO-2013) 
                July 6-10, Amsterdam, The Netherlands
                  Organized by ACM SIGEVO

 Submission Deadline: June 21, 2013


We are pleased to announce the 2013 Simulated Car Racing Competition.

Get ready to enter the challenge and to submit the fastest driver!

** GOAL ** 

The goal of the championship is to design a controller for a racing 
car that will compete on a set of unknown tracks first alone (against 
the clock) and then against other drivers. The controllers perceive 
the racing environment through a number of sensors that describe the 
relevant features of the car surroundings (e.g., the track limits, 
the position of near-by obstacles), of the car state (the fuel level, 
the engine RPMs, the current gear, etc.), and the current game state 
(lap time, number of lap, etc.). The controller can perform the 
typical driving actions (clutch, changing gear, accelerate, 
break, steering the wheel, etc.) 


The 2013 Competition does not introduce any major change with 
respect to the previous edition. In particular, we will not 
introduce any change to the APIs, such that it will be possible 
to build a controller based on the entries of the past year. 


The championship consists of nine races on nine different tracks 
divided into three legs, one for each conference, involving three 
Grand Prix competitions each. Teams will be allowed to submit a 
different driver to each leg. 

Each Grand Prix consists of three stages: 

    * the warm-up 
    * the qualifying 
    * the race 

During warm-up, each driver races alone for 100000 game ticks 
(approximately 30 minutes of actual game time) on each track. 
Drivers can collect useful information about the tracks and can 
tune their behaviors for the next stages. Accordingly, the 
performance of drivers in this stage is not taken into account 
for their scores. 

During the qualifying stage, as in the previous edition, each driver 
races alone for 10000 game ticks on each of three three tracks of 
the leg (approximately 3 minutes and 20 seconds of actual game time 
per track). The eight controllers that bridge the longest distances 
qualify for the actual Grand Prix races. 

During the final races, these best eight drivers race together. The 
races consist of eight times three lap races on each of the three 
tracks. At the end of each race, the drivers are scored using the 
F1 system: 10 points to the first controller that completes the race, 
8 points to the second one, 6 to the third one, 5 to the fourth, 4 to 
the fifth one, 3 to the sixth, 2 to the seventh, and 1 to the eighth. 
The driver performing the fastest lap in the race will get two 
additional points. The driver completing the race with the smallest 
amount of damage will also get two extra points. The starting grid 
of the first race will be based on the performance obtained in the 
qualifying stage. Each subsequent race, the starting grid will be 
shifted by one so that each driver starts from every position of the 
starting grid exactly once. 

As in the previous edition, the tracks used in each leg are unknown 
to the competitors. 

** WEBPAGE ** 


Submission deadline: June 21, 2013
Conference: July 6-10, 2013


The competition software, including servers for Linux & Windows, 
and C++ and Java clients, can be downloaded from the page: 

For inquiries send an email to scr <at> or visit 
the Car Racing Google Group at 

Please notice that you are *not* required to either attend or 
register to the conferences in order to enter the competition. 


    Daniele Loiacono (Politecnico di Milano) 
    Pier Luca Lanzi (Politecnico di Milano) 
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developers can use
our toolset for easy data analysis & visualization. Get a free account!
Torcs-devel mailing list
Torcs-devel <at>
brian gr | 27 Feb 11:17 2013

brian gr

Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
Torcs-devel mailing list
Torcs-devel <at>