0
\$\begingroup\$

I'm working on a project where I want to very accurately (+/- 1 degree) determine the angle at which a finger is bent.

One of my main constraints is that my sensing mechanism can't interfere (much) with the normal positioning and movements of the hand.

I'm trying to determine the best sort of sensing mechanism for this project. I looked into a lot of rotational encoders but I can't find any small enough to fit between two fingers and this wouldn't work for the first knuckle joint.

My next thought was to use flexible strain gauges across the top of each joint but I don't think I'll be able to get anywhere near the amount of accuracy that I need.

Another thought I had was to place a hall effect sensor on top of the finger joint but I'm not super sure how accurate I'd be able to get with this.

Any thoughts on a good way to do this?

\$\endgroup\$
  • 1
    \$\begingroup\$ What do you ultimately need to accomplish by making these measurements? \$\endgroup\$ – Nick Alexeev Sep 26 '16 at 22:41
  • \$\begingroup\$ Eventually, I want to use this information to restrict the motion of the finger and prevent the finger tip from passing into a certain area. \$\endgroup\$ – Indigo Sep 26 '16 at 22:50
  • \$\begingroup\$ Is a glove allowed ? with low tensile ligament pulleys like than hand strings, so as measure deflection of each degree of freedom. Extension would then be the shortest path length. \$\endgroup\$ – Sunnyskyguy EE75 Sep 26 '16 at 23:34
  • \$\begingroup\$ @TonyStewart Oh, that's actually really clever! Yes eventually it will probably be a glove based system so that would be perfect. \$\endgroup\$ – Indigo Sep 26 '16 at 23:53
  • \$\begingroup\$ A 3 axis accelerometer + 3 axis angular rate sensor (gyro) on each moving bit would produce the joint angles within the precision required while adding basically zero encumbrance. Of course it wouldn't be particularly cheap, as even a ST LSM6DS3 6 DOF IMU will cost about 1.71 dollars in bulk, and you would need 16 of them to sense everything in your power glove. In addition you'd need a fairly hefty microcontroller to aquire all the data, filter it and process the kinematics to produce joint angles. It'd be susceptible to damage as well. \$\endgroup\$ – jms Sep 27 '16 at 0:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.