go away spammer

Author Topic: Designing a Accelerometer Sensor based control for Robotic Arm  (Read 4184 times)

0 Members and 1 Guest are viewing this topic.

Offline lcreasyTopic starter

  • Beginner
  • *
  • Posts: 3
  • Helpful? 0
Designing a Accelerometer Sensor based control for Robotic Arm
« on: February 28, 2010, 06:55:22 PM »
I'm starting the design of a robotics project where I'm developing a hand held controler which I'd like to have drive the motion of a Robotic Arm.  I think I'll primarily be using accelerometer based sensors, but need some advice there.  The idea would be that the controller (AXON II) will constantly monitor the sensors for movement (3 axis) and turn that data into driving data for 2-3 servos.

Has anyone worked with an application like this or have an code examples to get me started on this application?

Thanks!

Offline SmAsH

  • Supreme Robot
  • *****
  • Posts: 3,959
  • Helpful? 75
  • SoR's Locale Electronics Nut.
Re: Designing a Accelerometer Sensor based control for Robotic Arm
« Reply #1 on: February 28, 2010, 11:33:43 PM »
Is this the kind of thing you are looking for? http://www.youtube.com/watch?v=9NEiBDBXFEQ&feature=player_embedded#at=33 Admin might have the code somewhere but i couldn't find it with a search.
Howdy

Offline lcreasyTopic starter

  • Beginner
  • *
  • Posts: 3
  • Helpful? 0
Re: Designing a Accelerometer Sensor based control for Robotic Arm
« Reply #2 on: March 01, 2010, 08:53:32 PM »
Possibly.  Do accelerometers handle angular movements as well as straight cartesian?

Offline Razor Concepts

  • Supreme Robot
  • *****
  • Posts: 1,856
  • Helpful? 53
    • RazorConcepts
Re: Designing a Accelerometer Sensor based control for Robotic Arm
« Reply #3 on: March 01, 2010, 09:01:53 PM »
Yes, in regards to the Cartesian plane it will work, but keep in mind it will only change when it is moving (accelerating). The reason if you tilt an accelerometer and the value stays the same is because of the acceleration of gravity is acting on it... so just moving horizontally will only yield results while it is accelerating.

Offline lcreasyTopic starter

  • Beginner
  • *
  • Posts: 3
  • Helpful? 0
Re: Designing a Accelerometer Sensor based control for Robotic Arm
« Reply #4 on: March 02, 2010, 09:20:19 PM »
Thanks for the insight!

I think I follow you now.  For example, if I accelerate the sensor to 'x' mm/s and then hold at that velocity, my recordable sensor data is only during that acceleration period, thus it would appear sensor data would cease unless I slowed the velocity through decceleration or stopped all together, in which case the sensor data would show a negative, or reverse movement due to the decceleration.  Is this correct?

Also, how would a gyroscope sensor behave in this case?  It is merely providing an angular roation measurement?  Is it also only providing data during acceleration periods, like the example above?

Thanks again!

Offline RAAJ90

  • Beginner
  • *
  • Posts: 3
  • Helpful? 0
Designing a Accelerometer Sensor based control for Robotic Arm
« Reply #5 on: April 19, 2011, 10:41:34 AM »
CAN GET DETAILS HOW TO START WITH ??

Offline vinniewryan

  • Full Member
  • ***
  • Posts: 56
  • Helpful? 2
Re: Designing a Accelerometer Sensor based control for Robotic Arm
« Reply #6 on: April 19, 2011, 02:29:28 PM »
...

Yes, and yes. If the accelerometer is under constant momentum on any given plane registering '0' and you slow or stop the sensor, it will read a negative value which will directly reflect the speed in which that momentum was ceased. A gyro works the exact same way in this respect, although a gyro will never hold its output like an accelerometer will when tilt is introduced. Think of a gyro like a wheel attached to a potentiometer's shaft with a tensioned rubber band holding the wheel in place. If you rotate the wheel in one direction with your hand then let go, the rubber band will bring the wheel back to its original position. If you rotate the "potentiometer" in one direction, the weight of the wheel will cause it to rotate slower than the "pot", this is where your output value increases (or decreases depending on the direction of rotation). Then, the rubber band will bring the wheel back to its original position so your output value will gradually return to zero.
« Last Edit: April 19, 2011, 02:31:52 PM by vinniewryan »

 


Get Your Ad Here