You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been playing around with the local planner and Fetch. I based my code off the base_local_planner and was copying the configuration variables over, when I saw this in fetch_navigation/config/move_base.yaml
pdist_scale: 8.0
gdist_scale: 12.0
However, if you look at the (documentation of these parameters)[https://wiki.ros.org/base_local_planner#Trajectory_Scoring_Parameters]. It says that the maximum value is 5.0 for both of them. I checked and indeed, both of these values are restricted to 5.0 when the local_planner plugin is loaded. This means that both path and goal are weighted the same.
Maybe this is a happy accident, but if that is the case, perhaps these values should both be set at 5.0. Otherwise, if there really is a desire to have a 2/3 ratio between pdist_scale and gdist_scale, it might make more sense to have the values be something like (2 and 3)?
I can happily submit a pull request with corrected values, but I suspect others know the robot much better than I do.
The text was updated successfully, but these errors were encountered:
Hello,
I've been playing around with the local planner and Fetch. I based my code off the base_local_planner and was copying the configuration variables over, when I saw this in fetch_navigation/config/move_base.yaml
However, if you look at the (documentation of these parameters)[https://wiki.ros.org/base_local_planner#Trajectory_Scoring_Parameters]. It says that the maximum value is 5.0 for both of them. I checked and indeed, both of these values are restricted to 5.0 when the local_planner plugin is loaded. This means that both path and goal are weighted the same.
Maybe this is a happy accident, but if that is the case, perhaps these values should both be set at 5.0. Otherwise, if there really is a desire to have a 2/3 ratio between
pdist_scale
andgdist_scale
, it might make more sense to have the values be something like (2 and 3)?I can happily submit a pull request with corrected values, but I suspect others know the robot much better than I do.
The text was updated successfully, but these errors were encountered: