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
Build from source of RMF on Ubuntu 20.04 and ROS Galactic
Description of the bug
When an ackermann drive vehicle is required to turn in place for any reason the plugin stops working altogether and the vehicle gets stuck in place.
This means that every time a vehicle ends a trajectory in a corner waypoint, it will likely be stuck there indefinitely if the corner is sharp enough.
I'm not too familiar with the ackermann code but my guess is that this is an edge case where the vehicle is in an area inside the arc between the waypoints that are being interpolated to make a smooth bend.
Steps to reproduce the bug
Create a navgraph with sharp turns (i.e. a square), make the vehicle execute two consecutive trajectories, where the first one ends in a corner and the second one starts from the same corner.
Expected behavior
The vehicle would execute a somewhat sharp turn to proceed to its first waypoint.
Actual behavior
The vehicle turns a bit, gets stuck and never actually executes its trajectory.
The text was updated successfully, but these errors were encountered:
Bug report
Required information:
Build from source of RMF on Ubuntu 20.04 and ROS Galactic
Description of the bug
When an ackermann drive vehicle is required to turn in place for any reason the plugin stops working altogether and the vehicle gets stuck in place.
This means that every time a vehicle ends a trajectory in a corner waypoint, it will likely be stuck there indefinitely if the corner is sharp enough.
I'm not too familiar with the ackermann code but my guess is that this is an edge case where the vehicle is in an area inside the arc between the waypoints that are being interpolated to make a smooth bend.
Steps to reproduce the bug
Create a navgraph with sharp turns (i.e. a square), make the vehicle execute two consecutive trajectories, where the first one ends in a corner and the second one starts from the same corner.
Expected behavior
The vehicle would execute a somewhat sharp turn to proceed to its first waypoint.
Actual behavior
The vehicle turns a bit, gets stuck and never actually executes its trajectory.
The text was updated successfully, but these errors were encountered: