Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ERROR with roslaunch bezier_application fanuc_m20ia_surfacing.launch #1

Open
oxidation99 opened this issue Dec 3, 2018 · 3 comments

Comments

@oxidation99
Copy link

Hi
I get a error with
roslaunch bezier_application fanuc_m20ia_surfacing.launch surfacing_mode:=true mesh_cad:=plane/plane_defect.ply
at picture below. So how can I fix this ?

image

Thank you

@VictorLamoine VictorLamoine transferred this issue from ros-industrial-attic/bezier Feb 14, 2019
@nlimpert
Copy link
Contributor

nlimpert commented Feb 22, 2019

For me (ROS Melodic, Ubuntu 18.04) MoveIt! usually manages to plan 100% of the 1507 waypoints, however surprisingly sometimes I get a result as follows:

[ INFO] [1550843113.146792250]: Attempting to follow 1507 waypoints for link 'link_6' using a step of 0.050000 m and jump threshold 15.000000 (in global reference frame)
[ INFO] [1550843113.303227998]: Computed Cartesian path with 1422 points (followed 93.245902% of requested trajectory)
[ INFO] [1550843113.306893277]: Execution request received

Successful computed paths result in the following:
tmp

tmp2

So I only started to use Bezier with melodic thus I'm not aware of any previous state running in kinetic. I haven't really looked into the issue but to me either the poses of the waypoints seem to have a rotation twisted by 90° in Y.

Why am I adding this to this issue?
Well I guess I have a similar problem using the painting application. MoveIt is unable to find a plan covering (almost) 100% of the waypoints:

[ INFO] [1550843452.247229475]: Received request to compute Cartesian path
[ INFO] [1550843452.247322502]: Attempting to follow 1055 waypoints for link 'link_6' using a step of 0.050000 m and jump threshold 0.000000 (in global reference frame)
[ INFO] [1550843452.255134687]: Computed Cartesian path with 9 points (followed 0.047393% of requested trajectory)
[ERROR] [1550843452.255365325]: Bezier application: A solution could not be found to move the robot along the trajectory, aborting.
[ INFO] [1550843453.347285872]: Received request to compute Cartesian path
[ INFO] [1550843453.347361612]: Attempting to follow 1 waypoints for link 'link_6' using a step of 0.050000 m and jump threshold 0.000000 (in global reference frame)
[ INFO] [1550843453.355115817]: Computed Cartesian path with 9 points (followed 50.000000% of requested trajectory)
[ERROR] [1550843453.355293204]: Bezier application: A solution could not be found to move the robot along the trajectory, aborting.
[ INFO] [1550843453.356299170]: Received request to compute Cartesian path
[ INFO] [1550843453.356374174]: Attempting to follow 1055 waypoints for link 'link_6' using a step of 0.050000 m and jump threshold 0.000000 (in global reference frame)
[ INFO] [1550843453.364228597]: Computed Cartesian path with 9 points (followed 0.047393% of requested trajectory)
[ERROR] [1550843453.364464429]: Bezier application: A solution could not be found to move the robot along the trajectory, aborting.

So my guess was that if MoveIt is forced to plan for poses pointing into wrong orientations it's unlikely to find a feasible path.

@VictorLamoine
Copy link
Contributor

Two things: one the first test case the orientation of the pose is not the one we defined when testing a long time ago.
We usually made the robot flange flat on the surface. With this configuration the path planning was 100%.

Second test fails because we used a M20iA before, and the M10iA is smaller, it fails to reach the surface.

I've seen these issues but did not solve them because I don't have the time.

Fix for first test: Change the pose orientation in the code, look at the tool orientation and make change the poses orientation so that it matches the tool orientation when the robot flange is laying flat on the surface.

Second test case: Edit the mesh and move it closer to the origin, it should do the job. If the orientation is not correct then the fix is similar to the first test case.

Please remember bezier is very experimental, far from being finished! Lots of things do not work at all 😥

@gavanderhoorn
Copy link
Member

jump threshold 15.000000

that's quite big for a jump threshold.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants