ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
0

end effector orientation groovy MoveIt

asked 2014-02-05 23:29:57 -0600

Equanox gravatar image

updated 2014-02-05 23:35:17 -0600

Hey there,

I'm working with an ur10 arm in groovy using MoveIt. I can plan and excecute the position and orientation of the end effector without problems when i'm using rviz and moving the connectors with the mouse. But when i do the planning from c++ over the move_group interface i'm getting the warning

>"Orientation constraint for link 'right_arm_ee_link' is probably incorrect: 0.482316, 0.611794, 0.626960, 0.345908. Assuming identity instead."

The arm moves in gazebo to the desired position but the orientation is all the time only the identity. Any thoughts?

geometry_msgs::PoseStamped pose;
rnd_pose.header.frame_id = "/ROBOT_BASE";

pose.pose.position.x = sensornode_.translation.x;
pose.pose.position.y = sensornode_.translation.y;
pose.pose.position.z = sensornode_.translation.z;
pose.pose.orientation.w = sensornode_.rotation.w;
pose.pose.orientation.x = sensornode_.rotation.x;
pose.pose.orientation.y = sensornode_.rotation.y;
pose.pose.orientation.z = sensornode_.rotation.z;

group.setStartStateToCurrentState();
group.setGoalOrientationTolerance(0.1);
group.setPoseTarget(pose);
group.move();

Thank you, Matthias

edit retag flag offensive close merge delete

3 Answers

Sort by ยป oldest newest most voted
2

answered 2014-02-06 19:44:53 -0600

fergs gravatar image

Have you tried normalizing the quaternion? Those numbers do not appear to be a normalized quaternion (if it is printed x y z w, then 0.482316, 0.611794, 0.626960, 0.345908 actually has the x y z normalized to a unit vector)

edit flag offensive delete link more
0

answered 2015-09-28 06:49:24 -0600

Diana gravatar image

hello

sorry but how did you normalize the quaternions?

edit flag offensive delete link more

Comments

Please open a new question. If you use Answer or Comment section to ask a new question, people won't even find it.

130s gravatar image 130s  ( 2015-09-28 12:57:51 -0600 )edit
0

answered 2014-02-07 02:20:25 -0600

Equanox gravatar image

Thanks for your advise, When normalizing quaternions the error didn't occured. But the origin of my fault was that i were using a false transformation. Your answer lead me to the right path cause i was wondering why they are not normalized.

Now i can plan to the right position and orientation :)!! Thanks!

edit flag offensive delete link more

Question Tools

2 followers

Stats

Asked: 2014-02-05 23:29:57 -0600

Seen: 2,038 times

Last updated: Sep 28 '15