How do identify/get the closest metal object

[akkon83]

Hello,
I'm a junior engineer, and I'm facing a physics/maths problem, for my new project.
I don't know all the technologies at the moment, and don't know all the tools available. I would like to calculate/get the closest Metal object from an other object (called x).
x could store a sensor or a microcontroller to proceed in the calculation, but what do i need, to get the distance(dont need the distance in cm or m but more the closest object).What will be the easiest method to use? Ulstrasound? the laser meter is not an option as I'm in a 3D environment.
Hope I was clear in my explanation you help will be greatfull.

ps: this was already posted in General Engineering but didn't get any response so I'll try here ;)

 

[Unrest]

Radar?

Inductive (metal detector)?

Ultrasound and laser won't distinguish metal from non-metal.

What other characteristics to the metal object and non-metal objects have? Are the metals always shiny? Ferrous?

 

[akkon83]

The object will all be made of stell, the sensor/radar whatever I don't know all the tecknologies would have to find out the closest one to him selft out of 5,6,7 object (made out of iroon).

 

[akkon83]

All the programming stuff I can handle but I don't know what to use, to detect that is my main problem ;)

 

[Unrest]

The object will all be made of stell, the sensor/radar whatever I don't know all the tecknologies would have to find out the closest one to him selft out of 5,6,7 object (made out of iroon).

OK, can you provide more details? How about the non-metal objects? If they are obstructing the closest metal object do you still have to detect the metal one? What else distinguishes the metal objects from the non-metal ones? Can a camera show the difference? How about accuracy for more than one closest object? What do you actually need to find out? Just the direction in 3D? To what accuracy?

How far away are they? If close enough you might be able to use change of inductance of a coil (metal detector), or even measure the magnetic field (of the earth, altered by the objects). Might require some powerful software/math.

How about swing a magnet on a string, and reel it out till it grabs something? Or just have powerful magnets attached to the device and ramp up the power till they grab something. These might sound silly, but without knowing much about your requirements they could be the best solution.

If they're far enough away, maybe a radar.

If they're in thermal equilibrium with everything else, and are shiny, perhaps use a scanning IR temperature sensor or IR camera to detect different emissivities.

Why rule out laser distance measurement? You can do 3D laser scanning.

 

[akkon83]

cheers for your respond, I'll try and be more precise on my requieremnt.
first it won't really be a 3D environment but kind of, it will be on a plain surface.
Normally there shouldn't be any obstacle, only metal object will be near it.
Then the maximum distance will probably be 3 metre, and minimum distance will be touching it. Generally my dtecter will be in the middle of theses 5-6 object. After the detector musn't move the object, I just need to know which one is the closest, I will then do know yet how but indicate the location of the nearest object with a led or something like that not sure yet. I would probably need to make lots of calculation as the object can all move and then will need to find out which one is the closest again, this action is supposed to be quick enought, I think that will depend on my program I guess, if my detector isn't able to find it in half a second it is useless.
Hope I was clear enough don't hesitate if you need more details.
I supposed that laser meter wasn't an option but maybe it is as my only knowledge on this equipment was only was you give the direction and you mesure a wall or something like that .

 

[Unrest]

In this case it sounds like the traditional approaches of ultrasound or laser would be most suitable. The 'metal' aspect was a bit of a red herring.

If you require high directional resolution then you might need a rotating scanner (like http://www.lasersforindustry.co.nz/products/lasertech_gs50" ). For a more approximate direction with objects that subtend large angles at x, it might be easier to install multiple fixed sensors pointing in all directions (like parking sensors on car bumpers). You'll probably have to look into several methods to really find which is "easiest" or "best".

 

[akkon83]

I think the laser approach isn't possible, this method will work if we place the laser in the correct possition etc.. in my project we don't know how the laser will end up that why I was mentionning a 3D environment, and also I requiere something small and it seems that the laser meter is to big, I will dig into the ulstrasound method, it was the only solution I found on the net so I think I look into that. thanks for your help