Making a robotic arm which will collect vegetables

Now i got a stable base that can run on dirt and can bring a payload of 2kg, remote controlled and a good brain which can get where are the veggies.

The robot base is controlled with arduino, and i want it to control a robotic arm to collect things.

What is the cheapest way to get an arm that can lift at most 1kg, grab firmly but not squeeze?

I was thinking to use a servo controller board like the ones used with hexapods, driving 4 stardard size servos and 3-4 micro servos to cut tomatoes or cucumbers from the plant and grab it. Maybe with some sort of pressure sensors to be more gentle.

Weight and torque

CrustClaw arms met weight and grip requirements, but i fear the robot would tip over if the arm moves too fast.

About lifting i was thinking to have a rotating joint in the base, a 360 continuous servo pulling a cable to move the first part of the arm and a second to move the second part of the arm. Then the gripper hand.

That way i would have actuators near the center of weght and can use a custom gear ratio.

The arm doesn’t need to be longer than 60 cm, even if longer it’s better

What you are trying to do is

What you are trying to do is actually a pretty difficult thing and requires a lot more power than you would think.

A 60 cm arm picking up a 1 kg item is going to require a 60 kg * cm motor.  No servo under hundreds of dollars is going to do that for you.  Roxanne pushed me toward some motors that are designed to open skylight windows that are 12v and around 30 kg*cm for around $15/motor on ebay and do about 1-2 rpm.  You will need something able to switch higher voltage and amps so a standard board isn’t going to cut it.  You will need to have variable resistors on the joints going to analog to get your position for the IK calcs you will need to do.

The arm I played with was 40 cm and I was trying to pick up 2 kg.  What I did was use a couple Dagu Mini driver which have a built in mosfet h bridge that will do I think 3.5 amps for a short period.  I used a 2 kg counter balancing weight (like a teeter totter) to reduce the high torque situations for the joint closest to my base.  I got it working but was unhappy since it wasn’t stiff enough and would bounce all over the place.  So, make it as stiff as you can especially with an arm that long and make sure all the joints have no slack at all.  I have a revised design but haven’t gotten to building it yet.

Good luck.

Regards,

Bill

 

 

I know this is just me…
When I’m reading this while tired I keep reading the title as “Making a robotic arm OUT OF vegetables.” :slight_smile:

The arm which I have

The arm which I have installed on my robot is rated for 500 grams payload. It uses servos with 5:1 gearing which then drives some very large gears. In my page I have a link to a video  http://youtu.be/HXhoDoeStPI which shows some very first moves of the arm controlled manually by me.

https://www.robotshop.com/letsmakerobots/ex1-earth-explorer-one

I purchased that arm from www.robotshop.ca

Earth explorer one

Your robot is maybe one of the first robot i taken for inspiration:)

I’ll try to build the arm, as i’ve got more time than money. The arm on EX-one are just fine for the task, but quite expensive and i’d need a stronger base. How much does that arm weight?

 

Next challange!

Now that i think, i can grow bamboo and use sticks as bones!

Rethinking it

That’s why i opened this thread. I read here about many legged bots struggling for power and industrial arms require too much power. But i feel there is somewhere a way to get something efficient enough.

I’ll look on e-bay for these motors also. Maybe using a combo of high torque brushless and precise servos could achieve it for lesser price.

Arm payload capacity

Zucchini and cucumbers weight about 200~300mg and grow from ground height to 60cm(cucumbers can grow way higher, but i’m grabbing low ones).

The robot will get a vegetable at once, and then discharge the container when the base has reached max payload or capacity. Reading around i understood that arms managing Kilograms are too slow, bulky, power hungry or expensive, so a goal of 350mg for 50cm is already a good success!

At the beginning i was thinking to give to the arm extra strength because it could get stuck in wines and maybe move small pots, but that can be achieved with way shorter arms or a fork lift.

I can make custom iron parts, but i prefer to use plastic. I’ll try to draw up something.

So far i got that it needs a counter weight, standard size servos with low gear, a custom gripper and a dedicated controller for servos. The first prototype will have an extensible forearm so i can test with reach/payload. Any advice where to buy servos or controller boards?