ROS commands don't match the actual iRobot hands

[osrf-migration]

Original report (archived issue) by Steve Peters (Bitbucket: Steven Peters, GitHub: scpeters).

---

Migrated from pull request #395, originally posted by @mdfloyd

I've been playing with the iRobot hands in Gazebo, and have noticed that the ROS commands don't match the actual iRobot hands. From the real iRobot hands documetation (and experiment): For posititons 1-4: 3500 to put finger approx 90 degrees, 6000 to close gently, 7000 to close finger tightly. For spread (5th position) 8.533 ticks per degree: 768 is 90 degree T grip, 512 rotates the fingers 60 degrees for a spherical grasp. The ROS commands for running the Gazebo iRobot hands are much larger by at least an order of magnitude.

[osrf-migration]

Original comment by John Hsu (Bitbucket: hsu, GitHub: hsu).

---

Thanks @mdfloyd, that's good data. For the existing plugin I made some terrible guestimages on what the transmissions might be for the fingers tendon drives, and ended up with something not surprisingly unrealistic.

[osrf-migration]

Original comment by Marshall Floyd (Bitbucket: mdfloyd).

---

For better accuracy, the 4 tendon motors "24 counts per motor revolution x 300 motor revolutions for one full spindle rotation." I didn't include that before because it doesn't convert to position directly without the geometry of the radius of the tendon on the motor spool, etc. But it can help to correlate to velocity, which is motor ticks per minute. For velocity 1000-2000 is a good range.

[osrf-migration]

Original comment by Nate Koenig (Bitbucket: Nathan Koenig).

---

• changed state from "new" to "on hold"

On hold while we work on Robotiq hand model.