Suitability of 5 axis robot for using IKFast

edit

Hi

• Openrave version:https://github.com/rdiankov/openrave (master) built with advice from https://scaron.info/teaching/installi...

  • ROS version: kinetic

  • Linux Distro: Ubuntu 16.04

I am a newcomer to openrave and ikfast. I am working on a robot that is very similar to a 5 axis mill in design, but with some unusual offsets in the final two rotational axes. Here is a link to a simplified kinematic diagram (using pasteboard as I do not have enough karma to post the jpeg):

Diagram

It is basically a PPPRR robot. I extracted a URDF from solidworks for our robot, and the URDF works well in ROS / RViz. I was able to build a moveit package and via KDL, was able to plan and move correctly.

However when using the latest stable version of IKFast from the github, I am repeatedly coming up against errors:

python `openrave-config --python-dir`/openravepy/_openravepy_/ikfast.py --robot=saws_urdf.dae --iktype=translationdirection5d --baselink="0" --eelink="5" --savefile="$IKFAST_OUTPUT_PATH"
INFO: moved translation [0, 0, 41/200] to right end
INFO: moved translation [0, 0, 0] to left end
INFO: moved translation on intersecting axis [6139/12500, -43143/100000, -40543/100000] to left
INFO: [[0, 0, -1, 0],[0, 1, 0, 0],[1, 0, 0, 0]]
INFO: [[1, 0, 0, 0],[0, 1, 0, 0],[0, 0, 1, j0]]
INFO: [[0, 0, 1, 0],[0, 1, 0, 0],[-1, 0, 0, 0]]
INFO: [[1, 0, 0, 0],[0, 1, 0, 0],[0, 0, 1, j1]]
INFO: [[1, 0, 0, 6139/12500],[0, 0, 1, -40543/100000],[0, -1, 0, 43143/100000]]
INFO: [[1, 0, 0, 0],[0, 1, 0, 0],[0, 0, 1, j2]]
INFO: [[1, 0, 0, 0],[0, 1, 0, 0],[0, 0, 1, 0]]
INFO: [[cos(j3), -sin(j3), 0, 0],[sin(j3), cos(j3), 0, 0],[0, 0, 1, 0]]
INFO: [[0, 0, 1, 0],[1, 0, 0, 0],[0, 1, 0, -61/1000]]
INFO: [[cos(j4), -sin(j4), 0, 0],[sin(j4), cos(j4), 0, 0],[0, 0, 1, 0]]
INFO: [[0, 0, -1, 0],[0, 1, 0, 0],[1, 0, 0, 41/200]]
INFO: ikfast translation direction 5d: [j0, j1, j2, j3, j4], direction=[0]
[0]
[1]
INFO: computed in 0.536197s
INFO: attempting li/woernle/hiller general ik method
WARNING: symbols [j2, cj3, sj3, cj4, sj4] are not all rotational, is this really necessary?
WARNING: CannotSolveError: symbols [j2, cj3, sj3, cj4, sj4] are not all rotational, is this really necessary?
INFO: computed in 0.415236s
INFO: attempting li/woernle/hiller general ik method
WARNING: CannotSolveError: Li/Woernle/Hiller method requires 3 unknown variables, has 2
INFO: attempting kohli/osvatic general ik method
WARNING: CannotSolveError: Kohli/Osvatic method requires 3 unknown variables
INFO: attempting kohli/osvatic general ik method
WARNING: CannotSolveError: Kohli/Osvatic method requires 3 unknown variables
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/openravepy/_openravepy_/ikfast.py", line 9521, in <module>
    chaintree = solver.generateIkSolver(options.baselink,options.eelink,options.freeindices,solvefn=solvefn)
  File "/usr ...