Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Radar Controller Airpowered Gunnery

Status
Not open for further replies.

Ron Jeremy

New Member
I'm brainstorming for a senior project and wondering if I could bounce an idea off the board.

I'm interested in model warship combat (see youtube example here). As a first pass, my idea is a fixed shore battery consisting of a single air-powered bb gun. The electronically interesting part is a radar that would determine azimuth and range to a target on the water, then aim the gun and fire.

I've never worked with radar as at a hobbyist level. Does anyone have experience with a project like this?
 
Assuming they let you build weapons for your project (at my university they certainly don't) radar is usually far beyond the means, budget, and expertise of undergraduate students and hobbiests. Radar units just don't exist at non-commercial prices, and the expertise and parts required to build them is...a lot to say the least.
 
Last edited:
It seems easy initially - however radar is as complicated as computer vision if you want to provide an acquisition and tracking system (identifying what is a target for example).
You'll need a good background in signal processing.

Another option is to look at vision based blob tracking. This means using a computer to detect a blob on a video feed, when the blob moves use optic flow to pass commands to the X/Y servos to maintain the blob in the centre of the screen. The contrast of the blob in relation to the background may need to be high.

If this is for public displays then I would avoid using high energy beams such as lasers to provide the high contrast blob for ranging. Alternatively paint the ship a paint that will fluoresce at a wavelength to create the blob. From one side it'll look like a ship, from the other it'll look like it's been lit up like a christmas tree.
Just filter the image to give a blob - not it will be a blob, so targetting will still need some work. Ranging? if the light is parallel source you could use the position in the frame to range but the fact the blob shape will change depending on the angle and range of the ship from the camera would make that tricky.

Estimating vector of the ship if it's moving will also prove tricky.

You'll also need to program in safety too as you're using projectiles. Program in a firezone and ensure that BB pellets will be fired within that zone. Nothing like putting your hand infront of the active system only to be targetted with a hail of BB pellets.. ow!
 
Thanks for all the feedback! Video might be the way to go. I wonder about the contrast if it's supposed to operate in daylight. Maybe if the target was of a known dimension and had bright LEDs on the bow and stern, and amidships on either beam, then I could get a better tracking versus a blob.?
 
Thanks for all the feedback! Video might be the way to go. I wonder about the contrast if it's supposed to operate in daylight. Maybe if the target was of a known dimension and had bright LEDs on the bow and stern, and amidships on either beam, then I could get a better tracking versus a blob.?

Yes an active target would be easier.

Remember the blobs will be non-uniform and fuzzy however you should be able to get a reasonable guess of the centre point.

A couple of examples of this I've worked with:

**broken link removed** - I have the CMUcam2+ which suffers from low processing power and frame rate - it's also not programmable - the later version looks more programmable but not stellar in performance. It will do blob tracking out of the box but I didn't use it.

Surveyor - this is a much higher performance platform and is very capable of providing blob detection. I helped these guys develop the stereo vision platform - my SRV1 as two cameras and blackfin processors.
The quality of video feed is very good too.
**broken link removed**

You'll need to do some programming using the gcc compiler chain but the folks on the Surveyor forum are helpful in that aspect.
 
I've written a response but I think because it had a few links that it's gone to the moderators for moderation..
 
Status
Not open for further replies.

Latest threads

Back
Top