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
Once a path has been generated, the motion profiler in OkapiLib allows the output to be reversed or mirrored.
However, when doing this, the pathpoints' pose coordinates remain in their original state. If one were to want to write a pure-pursuit style controller, the odometry from the robot will not match the coordinates of the (now reversed/mirrored) path.
The text was updated successfully, but these errors were encountered:
dcieslak19973
changed the title
Support for reverse/mirror a path
Support for reverse/mirror a path to update the path's coordinates
Apr 9, 2021
From what I recall, this is the same behavior as Pathfinder. I don't believe that Pathfinder's generator was aware of a path being reversed.
I agree that it would be easier for compatibility with pure-pursuit to have this reversal expressed in the generated coordinates but I don't want to break backwards compatibility any more than necessary. If you can think of a way to have Squiggles do both options then I'm all for it, but I don't want to break anyone's existing code.
One thought I had was to add a pair of functions to squiggles which would take the vector of PathPoints (as returned from generate) and return a new vector of PathPoints that are evaluated as though the input vector's wheel speeds were reversed or mirrored.
Once a path has been generated, the motion profiler in OkapiLib allows the output to be reversed or mirrored.
However, when doing this, the pathpoints' pose coordinates remain in their original state. If one were to want to write a pure-pursuit style controller, the odometry from the robot will not match the coordinates of the (now reversed/mirrored) path.
The text was updated successfully, but these errors were encountered: