Thanks Brian. I think there were problems both with the incoming laser  
scans and the robot's odometry. We had the min and max scan angle  
wrong but the angle increment correct. What's odd is that the laser  
scans looked correct in rviz but were actually skewed oddly.

-Dan
On Jun 3, 2010, at 4:44 PM, Brian Gerkey wrote:

> On Thu, Jun 3, 2010 at 10:07 AM, Dan Lazewatsky
> <lazewatskyd@cse.wustl.edu> wrote:
>> I'm using gmapping to create some maps and getting a lot of messages
>> saying "Scan Matching Failed, using odometry" which is a bit
>> worrisome. Does anyone know what could be causing this and how I  
>> might
>> go about resolving it?
>
> If it only happens sometimes, it might not be a problem.  If it
> happens very frequently, then I would guess that something's wrong
> with the incoming laser data or the associated transforms.  Have you
> tried visualizing the laser data in rviz, in the odometric frame?
>
> 	brian.
> _______________________________________________
> ros-users mailing list
> ros-users@code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users