Robotic Hand

Hi guys! I'm just starting to work with Arduino as i need it for a science fair project. I think i should be all set on the programming side but i was really hoping you guys could nudge me in the right direction to help me complete my project by January. I'm basing my project off of this abstract I Like to Move It: Motorizing a Robot Hand | Science Project.

What i really need help with is deciding which motors to use and somewhat of an idea of how to breadboard. For this science fair i'm testing 2 different types of motors to see how much pressure the hand can have on a pressure sensor. So i think i need motors with different amounts of torque(?). Also i need an idea of how much power my breadboard would, what type of breadboard, and how to use transistors.

Any help would be appreciated. Links would be great too as long as you explain to me what im looking at.

Thanks again!
-Rohawa

If you use servos instead of motors, you don't need any special driver hardware (they are just controlled from a PWM analog output pin).

Depending how many servos you need to control, it may be difficult to drive them all from a single Arduino (difficult but not impossible - but adding multiple Arduinos might be an easier problem to solve than multiplexing servo outputs).

Alright. What type of servo would you reccomend for this project? i need 2 types for different hands. Also would you think the fingers would bend properly if i linked one motor to 2 fingers?

Some servo hand stuff.

http://www.youtube.com/results?search_query=servo+hand&oq=servo+ha&gs_l=youtube.1.2.0l4.2633.18809.0.24009.8.8.0.0.0.0.2139.3260.0j6j1j9-1.8.0...0.0...1ac.1.Ft7U-yDyJQw

PeterH:
If you use servos instead of motors, you don't need any special driver hardware (they are just controlled from a PWM analog output pin).

Depending how many servos you need to control, it may be difficult to drive them all from a single Arduino (difficult but not impossible - but adding multiple Arduinos might be an easier problem to solve than multiplexing servo outputs).

Or you could use one of these http://proto-pic.co.uk/20-channel-rc-servo-driver/

I am using one for client where the Arduino is tied up doing other stuff that messes with the timers and so the PWM. This solution passes PWM signal generation away from the Arduino and is very stable.

Cheers Pete.

That looks good. Ill check around for those servos. how about bread-boarding? any idea how that would work?

rohawa:
That looks good. Ill check around for those servos. how about bread-boarding? any idea how that would work?

This is a DIL chip and so will fit into your standard breadboard without any problems, I built my prototype on breadboard

I also used an atmega328 chip nativly and not an Ardiuno board as it keeps the cost of production down.

My design also used an SPI capacitive touch sensor breakout from sparkfun and so they both sat on the SPI bus together.

Cheers Pete.

Can you explain that a little more? Im very new to this.

rohawa:
Can you explain that a little more? Im very new to this.

Here are a couple of pictures of the Fritzing layout that replicates breadboard and the PCB that finally went into production.

As you can see the DIL chips fit a standard breadborad. I use standard headers to connect the Servos to

Cheers Pete

Forum2.png

I havent been on the forum in a while but I'm about to order some parts. Im planning to use servos.... specificallly these ones.

Im currently diagramming the breadboard and ordering my voltage regulators etc.
I was wondering if you guys could reccomend to me a decent breadboard that might work well for four servos.
Last but not least I was wondering how i would be able to hold my servos in a position. I want the hand im creating to grip a rod and im wondering how i would be able to make it do that.

Thanks guys Youve been helping alot so far. :slight_smile:

I don't understand why you would want a breadboard. The servos just need ground, power, and the pwm signal from Arduino. Do you really need a breadboard just for that?

To simplify the connections if you want a plug-and-play solution I believe there is a servo shield that provides a set of standard servo sockets so you could plug the servos in directly, but all you'd need was one flying lead per servo and connect the power and ground from each servo to your power supply and the pwm line to the corresponding Arduino output.

I cant find a servo shield thats still in production. Could you please link me to it. :slight_smile:

By the way, would those 2 motors work well for my purpose or would you suggest something else?

I cant find a servo shield thats still in production.

Perhaps because nobody bought one, because they are not needed. Connect the power leads of the three servos to positive side of the power supply. Connect the grounds of the three servos and the Arduino to the negative side. Connect the three control wires to the Arduino. No shield needed.

rohawa:
would those 2 motors work well for my purpose or would you suggest something else?

If it's only for a robotic hand and not for a rock crusher, you probably don't actually need a high torque metal geared servo and could get away with something much cheaper. To know for sure you're need to know what sort of force, travel and speed you need from your servos. Then you can compare those requirements against the speed/torque ratings to see which servos would be adequate.

Well the point was to do a pressure test with the hand. so thats what i was thinking.

Well the point was to do a pressure test with the hand. so thats what i was thinking.

A robotic hand might not be the best fit for pressure testing.

rohawa:
Well the point was to do a pressure test with the hand. so thats what i was thinking.

I don't know what you mean by 'pressure test'. You want your robotic hand to squeeze some sort of force sensor to measure how much force it is applying? Is this the maximum force it can generate, or do you mean that you want to control the applied pressure as well as the finger positions? It's important to be clear about just what you're trying to achieve.

In any case you will need to decide what maximum force you're designing the hand for, do the mechanical design and calculate the torque required from the servos. This will tell you which servos you need. Or, if you just want to show that a bigger servo produces more force (not sure what the point of that would be) then just buy two sets of servos with different torque ratings. But it seems to me that a working articulated hand is a thoroughly worth while project in its own right and trying to measure or vary the output force would best be handled as a separate project once you have the basic hand working.