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
Greetings,
I am trying to setup sbpl+teb in stage simulation setup. I am using cfg/diff_drive configuration files as supplied with the code.
I found that robot has higher preference for making backward movements to align robot along the planned global path. Even in place rotation is not prefered for differential drive robot.
I have tried changing the parameter weight_kinematics_forward_drive to higher value. This influences trajectory selection but robot still makes in place back ward movements to correct its heading.
I have also observed that robot gets stuck in a loop at tight places to achieve its goal configuration.
Is this behaviour expected? If not, how can I improve it.
Thanks
Nitin
The text was updated successfully, but these errors were encountered:
Hi, sorry for the late response.
Did you try to tweak the max_vel_{x,y, theta} values? The teb_local_planner seems to prefer directions that allow a faster velocity.
You might also want to try out setting up the global_plan_viapoint_sep parameter of teb_local_planner in order to "force" the teb_local_planner to follow the global plan.
Unfortunately we face the same problems in tight spaces with ackermann driven robots. Maybe tweaking the said parameters (max_vel_*) might help you or the teb_local_planner, respectively to select proper trajectories even in tight environments.
Greetings,
![Uploading Selection_001.png…]()
I am trying to setup sbpl+teb in stage simulation setup. I am using cfg/diff_drive configuration files as supplied with the code.
I found that robot has higher preference for making backward movements to align robot along the planned global path. Even in place rotation is not prefered for differential drive robot.
I have tried changing the parameter weight_kinematics_forward_drive to higher value. This influences trajectory selection but robot still makes in place back ward movements to correct its heading.
I have also observed that robot gets stuck in a loop at tight places to achieve its goal configuration.
Is this behaviour expected? If not, how can I improve it.
Thanks
Nitin
The text was updated successfully, but these errors were encountered: