The ikfastpy can not get the correct ur5 rotation matrix in the world frame.

[sunnymints]

Thank you for your work.
we want to use your code to creat the solutions for ur5. But we get different solutions when we calculate the rotation matrixs between from ROS and your code.
for examp,we get the rotation matrixs from quaternion in ros:
[ 9.44e-05 1 9.762e-05 -1.0915e-01
-1.56e-04 9.76e-05 -9.999e-01 6.04687e-01
-1 9.440e-05 1.5637e-04 4.318e-01 ]

but your code from jiont angle is :
rotation matrixs:
[ 1 9.746e-05 -9.446e-5 -1.0915e-01
9.76e-05 -9.999e-01 1.56e-04 6.04687e-01

• 9.440e-05 -1.56e-04 -1 4.318e-01]
  We can see the two rotation matrixs have same position information，but the rotation part seem to be in other coordinate system