Planning in unknown space: simulation vs reality [closed]

edit

Hi,

I have implemented a navigation stack in stage. I want to prohibit the robot planning in unknown space, for this reason I set the following parameters:

common_costmap: map_type: voxel track_unknown_space: true lethal_cost_threshold: 100 unknown_cost_value: 255

global_planner (NavFNros): allow_unknown: false

Everything works fine, even if I don't understand why I have to set the map_type to voxel in order to prohibit the planning (i'm working with an Hokuyo laser scanner, so my immediate choice was to set map_type: costmap).

Running the navigation on a real robot I'm not able to issue any goal: the robot thinks to be always too close to obstacles: I get warning about clearing obstacles in the costmap and similar. It's not even capable of performing recovery rotation due to the presence of obstacles. Obviously there are no obstacles around the robot, it's exactly in the middle of the room. Looking at the costmap it's not inside any inflated obstacle radius.

As soon as I set track_unknown_space: false, everything starts working, but obviously I am not anymore preventing the plan into unknown regions.

Has someone experienced a similar behaviour and can help me ?

Thank you