Presence/Absence Detection System

Hey all,

I’m new with Arduino and I want to learn if it’s feasible to perform my project with arduino R3 & its shields.

I want to check the presence/absence of components on a part.

Step by step, machine need to be worked like below;

  1. Push start button and Fix the part on a jig with a linear servo or pneumatic controlled clamps.

  2. Detect missing components:

a. If there’s any missing component, Show the missing component with red ligh on picture which has LEDs for components and buzzer will make a sound for that.
Press “missing component button” to release the part from jig.
Put missing component on main part and push start button again.

b. If there no missing component, put a mark on the back side (like put a dot with a needle or put a sticker or etc…) and all LEDs will have green light then release the part fixations.

Please find the image below for my project b. I’ll kindly request you to help me about the feasibility of this project. If you’ve any question, please do not hesitate to ask me.

Thanks in advance,

Presence and absence detection.png

For the future it will be easier for people here if you attach your images as jpg files so we don't have to download anything. See this Image Guide

Are you expecting the Arduino to attach the fabric to the jig or will that be done by a person (or an existing machine).

Or is it the components that will be in the jig?

Will the components always be in the same place?
Are all the components identical?

if the components are metal would it be possible to touch the place with a pair of metal probes and pass a small current to detect the part?

I think you need to provide a lot more information. Do you have any photos of the actual thing you want to check?

Would it be an option to use a camera and image recognition software to do the job? However an Arduino would not be powerful enough to do that.

...R

The only universal recognition method is by image processing, on a powerful computer. Else you have to find sensors which react on the various materials, or on the height of the attached components. Precise measuring of 0.5mm differences in height is very expensive. Eventually you can use kind of wipers, that are blocked when touching the side of a component, but this again will be somewhat expensive, requiring a motor for each wiper and a motor stall detection.

First of all thank you for your quick reply and for your warning about image upload. I modified my post.

About your question, please find my answers below;

  1. I'm not expecting to assemble components with Arduino. It will done by a person.

  2. Components will always be in the same place.

3.Plastic, metallic and fabric materials are identical.

4.Metallic components can be detected with magnet but can not touch with metal probe. (Think that they're covered with non-conductive material.

  1. As far as I know, Camera is not a good option. However, my main part is glossy black in actual condition. Metal parts and fabric clothes are mat black.

I think they will not let me to send the actual part image.

DrDiettrich:
The only universal recognition method is by image processing, on a powerful computer. Else you have to find sensors which react on the various materials, or on the height of the attached components. Precise measuring of 0.5mm differences in height is very expensive. Eventually you can use kind of wipers, that are blocked when touching the side of a component, but this again will be somewhat expensive, requiring a motor for each wiper and a motor stall detection.

Thank you DrDiettrich for your reply.

I think we can detect orange plastics using colour sensors like in examples below;

Example-1Example-2

For metal nuts, we can detect them like in examples below;
Example-1
Example-2

For cloth fabric, I think we can use the reflectivity of the surface. Cause base metarial is very glossy but cloth fabric is mat.

Any suggestions else?

Eventually light barriers can be used, built from cheap laser diodes and photo transistors, if the base material reflects the beam well. Then an obstacle or mat surface may deviate, dim or displace the beam, when the beam is applied in a flat angle, grazing the surface.

deliibram:
I think they will not let me to send the actual part image.

That makes me wonder if this is a commercial project and it does not seem fair to be expecting free advice for a commercial project. The Arduino is an Open Source system and I expect people to be open about their projects.

There are plenty of professionals who would be glad to give you advice for a fee.

...R

Hi Robin,

It's not commercial issue. We're a bunch of students and we're visiting companies and looking for the need of industry. 3 weeks ago, we visited a company which is running in automotive sector. They're checking the parts & subparts with automation system. They're using PLCs, touch screens, etc.. which are expensive. So, I think that it can be a challenge for me to find a cheaper solution.

I mean "they" cause I don't have any relation with them and I don't think they will give the part to me for free without explanation. Cause they are not aware of my project.

That's why I intended to use arduino. If you're still thinking, it's a commercial use it's up to you to not to guide me.

Thanks,

So, I think that it can be a challenge for me to find a cheaper solution.

I certainly agree. Arduino is not the proper place to start; you should be studying detection methods.

deliibram:
I mean "they" cause I don't have any relation with them and I don't think they will give the part to me for free without explanation. Cause they are not aware of my project.

Without specific examples of the things you want to detect it is hard to help.

Maybe there is a good reason why the system they are using is expensive.

One part of the expense is the ability to call in a professional to fix it when it breaks down. Another part is the ability to claim substantial damages from the manufacturer if the product causes a significant problem. Think of the cost to Volkswagen of the recent scandal about faking compliance with emissions regulations (or whatever it was).

There is an enormous difference between the quality of a hobby solution and an industrial solution and if you are planning to work in industry it will be important for you to learn what those differences are and why. And I am not trying to suggest that there is no scope for cost reduction. Just don't expect to go from £50,000 to £500.

...R