[ros-users] Cliff sensors on the navigation stack
Gonçalo Cabrita
goncabrita at gmail.com
Sat Jul 31 00:00:23 UTC 2010
Thanks Eitan and Dan!
I'll try that first thing on monday and report back what happens.
Our robots, the Roombas, have 4 cliff sensors. I believe using PointCloud is more "legitimate" than faking a laser, so my question is, will a point per cliff sensor suffice, how should I but more per sensor?
Gonçalo Cabrita
ISR - University of Coimbra
Portugal
On Jul 30, 2010, at 7:33 PM, Eitan Marder-Eppstein wrote:
> Gancalo,
>
> It should be possible to make the navigation stack aware of this information. As Dan mentioned, you can publish a LaserScan or PointCloud in-front of the cliff that will keep the robot from driving there. You'll just have to configure the costmap to take in this sensor data, for which the relevant documentation is here: http://www.ros.org/wiki/costmap_2d#Sensor_management_parameters
>
> Hope this helps,
>
> Eitan
>
> On Fri, Jul 30, 2010 at 10:26 AM, Dan Lazewatsky <lazewatskyd at cse.wustl.edu> wrote:
> One solution might be to take your cliff sensor data and publish a fake laser scan when your sensors tell you there's a cliff.
>
> -Dan
>
> 2010/7/30 Gonçalo Cabrita <goncabrita at gmail.com>
> Hi everyone!
>
> I was wondering if it would be possible to easily feed cliff sensor data into the navigation stack (without changing it) so that a robot could avoid stairs for example.
>
> Gonçalo Cabrita
> ISR - University of Coimbra
> Portugal
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
>
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20100731/702ee541/attachment-0004.html>
More information about the ros-users
mailing list