ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
1

Unable to plan IK using MoveIt!

asked 2018-04-24 15:39:26 -0500

heethesh gravatar image

updated 2018-04-24 15:52:23 -0500

I'm using the LMA kinematics plugin to plan motion for a 6DOF arm. I'm able to use the interactive marker freely in Rviz (although some joints keep going in to invalid states or out of the defined joint limits sometimes and turn pink).

Now I am trying set pose targets from python (my script here). If I use get_random_pose() and set this pose using set_pose_target(), planning and executing works as expected. However, when I use the same random pose target values and create a custom geometry_msgs.msg.Pose() object and the set the pose target, I keep getting the following warning:

Python:

[ WARN] [1524579808.359818083, 4184.338000000]: Fail: ABORTED: No motion plan found. No execution attempted.

Moveit:

Error:   RRTConnect: Unable to sample any valid states for goal tree
         at line 215 in /tmp/binarydeb/ros-kinetic-ompl-1.2.1/src/ompl/geometric/planners/rrt/src/RRTConnect.cpp
Info:    RRTConnect: Created 1 states (1 start + 0 goal)

Could it be an issue with my python script or the inverse kinematics? How is planning working with random pose and fails to plan with the same position and orientation values when I create a custom pose object and assign these values? Tried increasing the planning time, attempts and goal tolerance with no luck!

PS: I was unable to use IK fast as it kept failing to solve the IK and with KDL, the interactive marker does not move properly (joints seem to stick at 90 degrees and movement is not free). Actually my arm is 5DOF and I had to include a dummy revolute joint with ilmits (0-0.01) to make the interactive marker work with LMA plugin. IK fast failed to generate for both 6DOF and 5DOF (Raghavan Roth equations too complex error).

URDF file here.

My MoveIt Config package here.

Image of my robot for reference:

image description

edit retag flag offensive close merge delete

Comments

Have you found IKFAST solution for 5DOF? we are also getting same error like, 'Raghavan Roth equations too complex error'

nd gravatar image nd  ( 2018-05-02 08:26:07 -0500 )edit

Nope. You could try adding a dummy joint to make it a 6DOF manipulator and then try solving IK using KDL plugin.

heethesh gravatar image heethesh  ( 2018-05-03 16:20:34 -0500 )edit

Hello, OP. Did you manage to solve this issue, by any chance? Or at least to find a decent work-around?

anca.jurjut gravatar image anca.jurjut  ( 2018-06-25 08:32:58 -0500 )edit

1 Answer

Sort by ยป oldest newest most voted
1

answered 2019-05-17 13:20:07 -0500

Bonadio gravatar image

Hello

I was able to generate an IKFAST for my 5dof, after trying several urdf configurations I found that the error 'Raghavan Roth equations too complex error' happened when my last link was not pointing on the X axis, I looked at your URDF and the last joint is on the Z axis

[]s

edit flag offensive delete link more

Comments

Hi, apologies for resurrecting such an old thread! I am running into the same problem and am wondering if you could explain / remember what you mean by 'pointing on the x axis'? I have added some rotations to my URDF so that my final joint points along x but receive the same error. However as this is not the home position, the y axis of that joints TF points along the global x. Does the x axis of the joint's TF have to be pointing along global x too? If it is more convenient / neater to reply I created a thread here https://answers.ros.org/question/3458...

h_riverside gravatar image h_riverside  ( 2020-03-05 03:05:39 -0500 )edit

Question Tools

2 followers

Stats

Asked: 2018-04-24 15:39:26 -0500

Seen: 1,252 times

Last updated: Apr 24 '18