Closed shenlirobot closed 9 years ago
The fingers are underactuated (two joints on each finger, but only one motor). Accurately modeling the fingers in simulation would require actually running a physics simulator with a model of the joint mechanism. This is very complicated and unlikely to produce physically plausible results so, instead, we model the distal joints as a mimic joint. You can see this in the official Kinova URDF file.
This happens on the real robot, too, because there are no position sensors on the distal joints.
I am closing this issue because there is nothing we can do about this - this is a limitation of the hardware.
However, it would be useful to put this information in a more prominent location. @Shentheman could you add a note to the README about this?
Ok! Sure! I will test it again and add some data also.
On Nov 2, 2015, at 9:41 PM, Michael Koval notifications@github.com wrote:
I am closing this issue because there is nothing we can do about this - this is a limitation of the hardware.
However, it would be useful to put this information in a more prominent location. @Shentheman https://github.com/Shentheman could you add a note to the README about this?
— Reply to this email directly or view it on GitHub https://github.com/personalrobotics/ada/issues/45#issuecomment-153221136.
Thank you so much! Sincerely Shen Li
The ada finger is not the same calibrated as the one in simulation. In simulation, if I do MoveHand(1.35,1.35), the two finger will cross to each other. But in reality, the two fingers barely touch to each other. Is this an issue? Do we change it intentionally to protect the finger or some other considerations?