Joint Values Inconsistency on j2n6s300

[rafaelpossas]

Hi All,

I've been trying to do some joint planning using my Jaco j2n6s300 and I came through the problem below.

I am using the kinova-ros package along with ROS Kinetic and Ubuntu 16.04

Whenever I start my simulation with: roslaunch kinova_gazebo robot_launch.launch

and I read the joint states topic (/j2n6s300/joint_states), there seem to be inconsistency on the last 2 joints of the arm (j2n6s300_joint_5 and j2n6s300_joint_6). The problem is twofold:

1. The joints are not even within the lower and upper limits specified within the URDF

2. Everytime I restart Gazebo they seem to get a random values for a same initial pose, See below an example

Starting Gazebo for the first time, look at j2n6s300_joint_5 and j2n6s300_joint_6

name: [j2n6s300_joint_1, j2n6s300_joint_2, j2n6s300_joint_3, j2n6s300_joint_4, j2n6s300_joint_5,
j2n6s300_joint_6, j2n6s300_joint_finger_1, j2n6s300_joint_finger_2, j2n6s300_joint_finger_3]

position: [-3.433988137491184e-05, 2.9003741053466783, 1.2992024657877943, 4.202535734888536, -11.165148707463949, 37.699077393946325, 1.000724306808693, 0.9994006487356604, 0.9993573993723155]

If I restart Gazebo and Run the same command again:

name: [j2n6s300_joint_1, j2n6s300_joint_2, j2n6s300_joint_3, j2n6s300_joint_4, j2n6s300_joint_5,
j2n6s300_joint_6, j2n6s300_joint_finger_1, j2n6s300_joint_finger_2, j2n6s300_joint_finger_3]

position: [-0.0006671249350160835, 2.8976716669402833, 1.3000769288977914, 4.197928347574803, -4.878021615876596, 43.9824150056194, 0.9989748324374386, 1.001878501799836, 1.0015273938192504]

These values are not only pretty high (Its all in radians) but they are different for a same robot pose.

Am I missing something in here or is this a bug?

Thanks in advance.

Rafael

[nsaif-kinova]

Hello Rafael,
Please check this link and let me know it this can help you.
#132
Sincerely,

[rafaelpossas]

Hi Thanks for the Response.

I have checked and tried the solutions proposed, but it still doesnt work. I can't do any kind of planning with such an inconsistency on retrieving joint values.

What is actually causing this?

[martine1406]

hi @rafaelpossas

Unfortunately, we do not know the source of this issue. I would say it is most likely linked with Gazebo... but I am not an expert in Gazebo. Have you tried posting on Gazebo-ros related forums?

[martine1406]

Hi @rafaelpossas

Have you found something related to this issue on the Gazebo forums? The controllers that are defined for the j2n6s300 Gazebo model might not be ideal. I really think this issue is related to #132

[felixmaisonneuve]

This thread has being inactive for over 2 years, I am closing this issue. If there is still a problem, please create a new issue.

Thank You