24 thoughts on “four degree of freedom robotic arm with simple haptics

  1. I’ve sent you a …
    I’ve sent you a longer private message which hopefully explains things better.

  2. I really appreciate …
    I really appreciate you helping me out dude. I get everything you said in the first post.

    But I don’t get what’s the difference between a PWM signal and a regular analog voltage signal. According to my understanding, PWM is also generating a voltage but the input is taken in a different way (through pulse widths) and a corresponding voltage is generated at the output. Analog voltage would give the same voltage directly to the servo.

    So what logic am I missing here?

    Thanks dude

  3. I’m not sending PWM …
    I’m not sending PWM to the servo board and commanding a position – i’m sending a voltage to the servo motor which tries to force the operators input gripper open.

    Does that make sense?

  4. when a torque is …
    when a torque is measured on the output arm, its value is used to calculate an analog voltage to apply to the input arm motor directly. I used PWM to generate this voltage. So, when the output arm is commanded to close on an object, the object effectively tries to ‘open’ the output arm and a torque is measured. The controller uses this measurement to calculate a voltage to apply to the input servo motor, where the motor tries to ‘open’ the gripper.

  5. @brandonsenn :
    The …

    @brandonsenn :
    The input gripper can be controlled using the PWM signals from the torque output accordingly. It is not possible to move the gripper servo manually when it is receiving a ‘HIGH’ PWM value. So when the output gripper is holding something, the input gripper stops at some position accordingly. But how do I let it go? The input gripper servo cannot be moved anymore as its receiving a signal. Could you please let me know how did u deal with that issue? Thanks a lot again dude.

  6. @brandonsenn : Hey …
    @brandonsenn : Hey dude thanks a lot for the reply. That didn’t really answer my question, but your on the same track as me. I agree that the signal is a PWM controlled signal. I’l tell you what I was planning to do. So we can see if we’re on the same page or not.

  7. The signal from the …
    The signal from the microcontroller to my input servo is not a traditional servo control signal, but rather a PWM controlled voltage which is proportional to the torque measured on the output gripper. So, there is, and should be some torque resistance on the input servo (this is the force feedback you’re trying to make) but not so much that the servo can no longer be moved. This required some tuning on my system.

    Does this answer your question?

  8. Hey dude its me …
    Hey dude its me again..could you explain me how exactly are you demonstrating the force feedback at the input arm gripper?..I remember you told me that you did not cut the driving gears for the gripper servo and you are moving it against the torque, which is fine..but i noticed that you cannot rotate the servo anymore against the torque once it gets any kind of signal from the microcontroller..could u please let me know how u solved this issue? Thanks a lot

  9. That is an amazing …
    That is an amazing idea! I will look into it and try to make something similar on my own. I am thinking about plexiglass design and force feedback on the gripper input so the operator doesn’t have to move the gears in it. I’ll have to figure out a way to mix the input from both the gripper and input”gripper”. Im looking forward to this project and if i get some HD cam it’s going on YouTube

  10. Thanks man.

    I …
    Thanks man.

    I don’t think I can use the same logic for haptics on the elbow..I’l let you know when my project is done 🙂

  11. since i used …
    since i used relatively low torque hobby servos, the input gripper could be moved even with gears in place due to the leverage of the gripper fingers. A small amount of resistance from these gears could be felt.

    Good luck with your project.

  12. Thanks a lot for …
    Thanks a lot for all the help.
    Yes I intend to build a similar system, but a bit bigger one and haptics feedback at both gripper and the elbow joint to feel the weights lifted.

    Just for one last time,
    If I remove the circuit board and the driving gears for the gripper servo, I should be able to get the feedback from the pwm output which is connected to the motor directly. But how do I move it freely? (as the driving gears are still intact)

  13. yes.

    Do you …
    yes.

    Do you intend to build a system similar to this? If so, let me know how it turns out!

  14. The microcontroller …
    The microcontroller can read the position of each of the input arm potentiometers and send a position command to each of the output arm servos. At the same time it reads the force measurement on the output arm and supplies a proportional voltage to the motor on the input gripper.

  15. Input arm joint …
    Input arm joint servos have their potentiometers wired to the microcontroller. The servo board is removed as well as the driving gearing. Input gripper servo has only the servo board removed. Driving gearing is still in place. Potentiometer is wired to the microcontroller as well as the motor.

  16. But what I don’t …
    But what I don’t get is,

    You cannot remove the wire from the potentiometer as u need the analog signals to be sent to the other arm.
    You are using the signal wire to send in the force feedback. You are also not removing the driving gears. So how will you move the gripper freely AND send a signal for the servo to move?

  17. ok this is what i …
    ok this is what i understand so far..
    1) you modified all the servos on the user controlled arm by soldering a wire at the potentiometer for analog signals, which are sent to the tracking arm.
    2) you cut the driving gears between the motor and the output shaft for 1st and 2nd joints, leaving the connections to the potentiometer, to rotate it freely, but still sending the analog feedback.
    3) the feedback motor is directly controlled by the pwm output/signal wire from the microcontroller.

  18. The “feedback” …
    The “feedback” motor is directly controlled by PWM output from the microcontroller, rather than the standard servo electronics which allows me to vary the motor voltage based on force measured by the Force Sensing Resistor rather than on servo position.

  19. I didn’t get the …
    I didn’t get the last part about force feedback.

    If you are cutting the wires between the motor and servo electronics, and blocking all communication, how are you sending the force feedback to the gripper that you are controlling?

  20. On the input arm, …
    On the input arm, for the 1st &2nd joints i opened the servo housing and removed the driving gears between the motor and the output shaft, making sure to leave the connections to the position sensor(potentiometer) intact. For the ‘gripper’ i cut the wires from the motor to the servo electronics but left the geartrain intact. The leverage of the gripper fingers allows the operator to overcome the servo gearing with a minimum force with the motor still connected to allow for feedback force.

Leave a Reply