[ros-users] Avoiding Collisions between robots
Brian Gerkey
gerkey at willowgarage.com
Mon Dec 27 19:40:48 UTC 2010
On Mon, Dec 27, 2010 at 10:39 AM, Jack O'Quin <jack.oquin at gmail.com> wrote:
> On Mon, Dec 27, 2010 at 11:48 AM, Eitan Marder-Eppstein
> <eitan at willowgarage.com> wrote:
>> On Sat, Dec 25, 2010 at 7:53 PM, <davidbsp at iol.pt> wrote:
>>> I have been making some experiments on ROS together with Stage, using
>>> multiple mobile robots in the same environment.
>>> All robots run the navigation stack. However, sometimes they collide
>>> with each other.
>>
>> So, its true that the navigation stack doesn't do a great job of handling
>> the multi-robot case, but I'd expect you to get reasonable results in Stage
>> as long as the laser mounted on each robot is capable of seeing the body of
>> the other robot. Have you verified that each robot in stage sees the other
>> in its entirety?
>
> This is definitely worth checking.
>
> I had problems migrating simulations to recent versions of stage.
> Although stage is a 2D simulation, they've added some "2.5D" features
> in stage version 3. Objects now have heights, and the laser simulation
> appears to send out a horizontal planar scan at the Z level of the
> laser's position. A laser mounted on top of a simulated robot may "see
> over the top" of another, similar robot.
To echo Jack: all models in Stage (including lasers and robot bases)
have heights. A sensor will only "see" other models at the same
height. If you copied from standard Stage .world examples, you
probably have some Pioneer-like robots with SICK-like lasers mounted
on their tops in such a way that a robot's laser only sees other
robots' lasers, not their bases. This is the correct result for such
a robot configuration; in the real world, the laser planes don't
intersect the robot bases. If you want to simulate a different
(perhaps less realistic) kind of robot, you can adjust the laser
heights; if you post your .world file (and also any files that are
included by your .world file), we can provide some pointers on what to
modify.
brian.
More information about the ros-users
mailing list