[ros-users] navfn allow_unknown

[Eric Perko]

So, you had to make all the unknown space (-1) in the map free space (0) to
get the planner to plan through it?

That seems like a really hacky thing to do - free space != unknown space
when dealing with a map, even if it is when doing planning in that map. For
example, if I recall correctly, the models AMCL uses treat a scan that falls
in unknown space differently than a scan that falls in free space (or at
least, I could see it making sense to do so). I wouldn't want to have to map
with the parameter set to one value then plan with another value.

- Eric

On Thu, Dec 2, 2010 at 12:11 PM, Christian Verbeek <
verbeek at servicerobotics.eu> wrote:

>  Hi,
>
> That was the correct answer. Thank you very much. I wonder if one should
> make this a parameter in later versions.
>
> Regards
> Christian
>
> Am 01.12.2010 10:33, schrieb Ye Cheng:
>
> Hi,
>
> You make a change in slam_gmapping.cpp to let you send a goal in unknown
> area if you are using gmapping:
>
> change this line:
>
> map_.map.data[MAP_IDX(map_.map.info.width, x, y)] = -1;
> To
>
> map_.map.data[MAP_IDX(map_.map.info.width, x, y)] = 0;I have a screenshot here for the demo. The green area is the explored area by A*:
> http://www.youtube.com/watch?v=CQ_BlPomr9c&hd=1
>
> Cheers,
> Simon
>
>
> On Tue, Nov 30, 2010 at 9:15 AM, Christian Verbeek <
> verbeek at servicerobotics.eu> wrote:
>
>> Dear All (especially Eitan),
>>
>> The allow_unknown parameter in navfn should allow to plan a path through
>> an unknown area. It does not seem to work. If there is an incomplete map
>> I would expect to get a path from navfn if the goal is in an unknown
>> area and the clear_goal_position flag is true.
>>
>> Regards
>> Christian
>>
>> _______________________________________________
>> ros-users mailing list
>> ros-users at code.ros.org
>> https://code.ros.org/mailman/listinfo/ros-users
>>
>
>
> _______________________________________________
> ros-users mailing listros-users at code.ros.orghttps://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/20101202/bdc7a8a9/attachment-0003.html>