You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm currently using a jaco arm. With help of this awesome package I can control my arm through moveit!. Thank you!
But I currently meet some problem. That is the real value I measured is different from the value returned by the forward kinematic program, which is calculated based on the joint state value returned by the jaco_arm/joint_states topic. In detail the difference in z axis is 1cm, the difference in x is 7mm, and the difference in y axis changed according to different poses.
Have you meet similar problem? How you solve it?
Looking forward to your response. Thanks in advance.
The text was updated successfully, but these errors were encountered:
We haven't run into this problem, but the first thing to check is if you're using a JACO or a JACO2 arm. They have slightly different link lengths and angles, which the package accounts for by loading in a different configuration file. You can do this by changing the arm_name parameter in wpi_jaco_wrapper/launch/arm.launch from "jaco" to "jaco2".
Other than that, the dh parameters come from the spec sheet Kinova provided for the JACO and the JACO2 arms. You can adjust them in the code at wpi_jaco_wrapper/src/jaco_kinematics, but then the forward kinematics may not match the urdf model provided in jaco_description, which may cause other inconsistencies.
Hi @dekent sorry for the late reply.
I double check the arm_name parameter. It's not wrong. I'm trying to update mine urdf file.
I'm interested about the precision of your JACO arm. After given a cartesian space pose and execute the trajectory. I calculate the arm's pose using forward kinematic program provided by kdl, and find mine arm have error about +-2mm in position.
Hello,
I'm currently using a jaco arm. With help of this awesome package I can control my arm through moveit!. Thank you!
But I currently meet some problem. That is the real value I measured is different from the value returned by the forward kinematic program, which is calculated based on the joint state value returned by the jaco_arm/joint_states topic. In detail the difference in z axis is 1cm, the difference in x is 7mm, and the difference in y axis changed according to different poses.
Have you meet similar problem? How you solve it?
Looking forward to your response. Thanks in advance.
The text was updated successfully, but these errors were encountered: