richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Sun Dec 24, 2017 4:59 pm

I had a look at this, using peizo contact mic's on a steel plate.

I found that there was a phase shift which made detection of the initial impact tricky.

I did wonder if there was a way to use a pair of linear ccd's from scanners to capture the pellet/bullet doing past. One for x and one for y, but the frame rate required is very high for firearms. Not so bad for airguns.

Using a camera to pick up the hole is a bit pointless as I want to avoid using up paper targets and it also fails when you shoot really accurately!


A simple, low cost solution would be a great bit of kit for shooters, but it does seem to be harder than you'd think to get right.

This video shows something working, but not with pellets!

https://www.youtube.com/watch?v=_1xf1fqT7eU

PaulCantrell
Posts: 1
Joined: Tue Jan 23, 2018 10:53 pm

Re: Need High Accuracy Electronic Target using triangulation

Tue Jan 23, 2018 11:05 pm

jim_hahn wrote:
Tue May 03, 2016 1:46 pm
I may have misread the original post too.
--jim
Jim,

I was shooting at Reading when you were running High Power... I just started up again and people were mentioning that you were working on a target system. Can you pm me?

Paul

shwas
Posts: 1
Joined: Sun Apr 28, 2019 9:50 am

Re: Need High Accuracy Electronic Target using triangulation

Sun Apr 28, 2019 10:02 am

Hi,
I am rifle shooter and I read your all post. Can you help me to built this circuit board. I don't know about electronic and programming.
Please help me.
My email id is [email protected]

Thank you
Shwas

Andyroo

Re: Need High Accuracy Electronic Target using triangulation

Sun Apr 28, 2019 3:58 pm

Given the last TWO posters asking for help have not been on the forum for a year and the one before that has not been seen for three years I would start a new thread explaining what you want to do and where you are stuck.

It is also not good manners to put your email on a forum asking for private help as the aim here is so that everyone can learn how to do things in open discussions rather than outside.

richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Sun Apr 28, 2019 5:20 pm

Just to update anyone on where I got with this.

Using an Arduino, which has the timer counters to do all the measurements in hardware, a really thick steel plate (important to avoid ringing) and acoustic pickups (very simple sound detector modules for Arduino), I was able to get a resolution down to 2mm with airgun pellets.

As the pellets are subsonic, I was detecting the impact with the plate. Initial experiments with piezo's on the plate is not much use as sound travels fast in the steel.

You can't use a thin plate as it rings and that is louder than the pellet impact.

To get better than 2mm you run into physics problems!

Basically the impact of the pellet takes some time from the tip hitting to the whole pellet squashing flat. This time is longer than the time needed to resolve down to 0.5mm.

The sound wave of the impact at this resolution is a slow ramp up, so it's hard to get a repeatable measuring point.

I also tried ultrasonic pickups from ultrasonic range finders and these worked better with faster response.

Roughly speaking, to resolve to 0.5mm at speed of sound, you are looking at 1.5uS.

A pellet travelling at 200m/s takes 25us to impact from first initial touch to completly squashed flat. The sound wave comes out during that 25us.

If the pellet was going a LOT faster, or it was a lot harder, making a more abrupt impact sound, it might be possible.

This is why I started looking at optical beam detection instead. Using a line scanner from a desktop scanner. Then I ran into a problem of not having a collimated line beam to illuminate it. Not without some very fancy, expensve optics!


Bottom line, is that you need a nice clean signal, optical or acoustic, that has a nice clean edge to measure accurately to 1.5uS. Pellet impacts don't give you this.

Shockwaves from a supersonic bullet may well do, but then you probably don't need sub millimeter accuracy unless you are talking .22 cal match target shooting.

The timing hardware was the easy bit. Plotting the impact from the timings of three pickups is an interesting task to do in software. Not hard, but does need careful thought!

This is also quite difficult to test, as you want to get impacts to occur at a known position to within 0.5mm, to check your software against.

I tired a pendulum striker to hit the plate at exactly the same place every time as it got tedious finding and measuring pellet impact points to 0.5mm


Also interesting to note that different pellet shapes give different acoustic signals to process. Wad cutters, not suprisingly giving better sound signals to work off of.

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Sat Nov 30, 2019 12:16 am

Very interesting read! I was wondering yesterday how it was that nobody had come up with a reasonably-priced target system yet, given how cheap technology has become in recent years. Did you ever try the system you'd come up with with .22LR rounds? Or just pellets? I know Jeff, the guy that writes the OnTarget software in the video you posted at the top of the page. A problem with using paper impact sounds is obviously the paper could be gone for the next shot. A steel plate obviously gets around that problem. Some commercially available targets solve that with a rubber roll that you have to replace now and then. I see there are also laser targets available now, but I'm not sure how those function. I'd really like to know which targets Eley uses at their ammo test ranges, as it sounds like they're using something similar to the scanner idea.

richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Sat Nov 30, 2019 6:34 pm

_Shorty wrote:
Sat Nov 30, 2019 12:16 am
Very interesting read! I was wondering yesterday how it was that nobody had come up with a reasonably-priced target system yet, given how cheap technology has become in recent years. Did you ever try the system you'd come up with with .22LR rounds? Or just pellets? I know Jeff, the guy that writes the OnTarget software in the video you posted at the top of the page. A problem with using paper impact sounds is obviously the paper could be gone for the next shot. A steel plate obviously gets around that problem. Some commercially available targets solve that with a rubber roll that you have to replace now and then. I see there are also laser targets available now, but I'm not sure how those function. I'd really like to know which targets Eley uses at their ammo test ranges, as it sounds like they're using something similar to the scanner idea.
With supersonic rounds, you can pick up on the shockwave as the bullet passes a microphone array. Air gun pellets aren't going fast enough so I used the impact sound.

The math is all the same, it's not that difficult and the processing isn't that tough especially if the hardware is doing all the timing for you.

I just ran into the issue of physics! Accuracy of under 1mm means measuring small time differences of an 'event' that in comparison, takes a long time to occur.

It may be possible to use faster processing to record the entire pellet impact waveform insead of an edge trigger and from those waveforms make a more accurate time delay measurement.

I see that there is a Teensy that runs at 600mhz, so it should be able to sample the pellet impact at ultrasonic frequencies,

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Sat Nov 30, 2019 11:49 pm

Yeah, I shoot subsonic target rounds in 22LR, so the supersonic method wouldn't be of any use in that case. How thick of a plate did you find was sufficient for pellets in order to stop the ringing? I imagine a sheet of Dynamat from a car audio supplier would also help in that regard. Stick it to the back of the steel. They're designed to stop the sheet metal in a car from resonating and adding/altering the sound you hear from the stereo.

As for ultrasonic sampling, I imagine you could also run into limitations of the microphones, depending on what you use.

richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Sun Dec 01, 2019 9:04 am

_Shorty wrote:
Sat Nov 30, 2019 11:49 pm
Yeah, I shoot subsonic target rounds in 22LR, so the supersonic method wouldn't be of any use in that case. How thick of a plate did you find was sufficient for pellets in order to stop the ringing? I imagine a sheet of Dynamat from a car audio supplier would also help in that regard. Stick it to the back of the steel. They're designed to stop the sheet metal in a car from resonating and adding/altering the sound you hear from the stereo.

As for ultrasonic sampling, I imagine you could also run into limitations of the microphones, depending on what you use.
Yes, you could damp the plate, I just used a really thick plate I had to hand, 1/2 inch I think.

I started with audio mics and then moved to ultrasonic transducers.

I am sure there must be an optical solution too. I couldn't find the right parallel light source.

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Sun Dec 01, 2019 9:49 am

I believe Eley's targets are optical, but I can't seem to find out what the brand/model is from just a small amount of googling. They say something about their bespoke software, but don't say anything about who makes the targets. Maybe they're custom, too. I don't know.

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Mon Dec 02, 2019 3:38 am

https://www.meyton.info/en/measuring-pr ... index.html

Too bad the small one is 4700 Euros, haha. Interesting concept, though.

richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Mon Dec 02, 2019 9:58 am

_Shorty wrote:
Mon Dec 02, 2019 3:38 am
https://www.meyton.info/en/measuring-pr ... index.html

Too bad the small one is 4700 Euros, haha. Interesting concept, though.
That's similar to what I tried to do, the detector is easy, it's the scanner bar out of a flatbed scanner. In order to cast a shadow that is directly in line with the bullet, the light beams all have to be exactly parallel (colimated). Thats the tricky bit.

It looks like that's what the device you found does.

Initially I just used a laser line illuminator but this is not colimated so the shadow varies in size and is squewed side to side depending on position.

The ideal solution is an array of lasers. One for each of the scanner sensor pixels, or some clever optical device that projects a flat, wide colimated beam of light.

Such devices do exist but they are very expensive.

I did wonder if some form of rotating scanning line could be used. You'd only need to know WHEN the light beam was broken and the scanner position you tell you where the bullet was. A tiny mirror on a motor with an encoder on. One for each axis. You would have lower resolution, as the distance away from the mirror increases due to fanning out.

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Mon Dec 02, 2019 11:19 am

Yeah, they're using infrared light sources of a type they do not mention. They do tell you how many beams each of the two different target sizes has, though. 192 for the small and 768 for the large. https://www.meyton.info/en/produkte/mes ... index.html

One can dream, hehe.

bachi
Posts: 3
Joined: Sun Jul 21, 2019 1:08 pm

Re: Need High Accuracy Electronic Target using triangulation

Fri Dec 13, 2019 4:34 pm

richardtheboffin wrote:
Sun Apr 28, 2019 5:20 pm
As the pellets are subsonic, I was detecting the impact with the plate. Initial experiments with piezo's on the plate is not much use as sound travels fast in the steel.
A pellet travelling at 200m/s takes 25us to impact from first initial touch to completly squashed flat. The sound wave comes out during that 25us.
I just wonderd what materials would transport sound slower than air and found that sound in rubber apparently has a speed of only 60m/s . Now I imagine getting a clear signal from material like rubber will become the problem - but if that would be solvable, you might come down the required resolution? Maybe other materials with sound speeds much lower than air exist?

Edit: I just found that SIUS uses rubber membranes for accoustinc measuring (https://sius-shop.ch/ls10-hs10-s10/102- ... er-kk.html)

Or in https://sius.com/wp-content/uploads/201 ... S10-de.pdf:
"3 HS10A026 Gummimembrane vorne (dient der akustischen Messung)" -> Rubber membrane front side (used for accousting measuring)

Now I am intrigued by the idea... :)

richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Fri Dec 13, 2019 5:08 pm

If you could find a rubber material, that wouldn't degrade from repeated impacts in a small area. My air rifle will put all it's pellets within 1/2".

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Fri Dec 13, 2019 7:03 pm

That’s why they used a replaceable roll of rubber.

I’ve been thinking about those Meyton targets. I don’t think they’re using lasers or any collimated light source at all. I’m betting they’re simply recessed LEDs so you get a cone of light that only reaches a certain number of the sensors on the other side. Their beams figures likely relate simply to the number of photocells they’re using. I don’t think it is the number of light sources.

richardtheboffin
Posts: 33
Joined: Thu Apr 19, 2012 12:04 pm

Re: Need High Accuracy Electronic Target using triangulation

Tue Jan 14, 2020 12:22 pm

_Shorty wrote:
Fri Dec 13, 2019 7:03 pm
That’s why they used a replaceable roll of rubber.

I’ve been thinking about those Meyton targets. I don’t think they’re using lasers or any collimated light source at all. I’m betting they’re simply recessed LEDs so you get a cone of light that only reaches a certain number of the sensors on the other side. Their beams figures likely relate simply to the number of photocells they’re using. I don’t think it is the number of light sources.
Possibly. I guess if the sensor is down the end of a tube so it can only 'see' directly ahead, it will only detect pellets directly in line with it. Not sure how consistent the shadow being cast will be, when the pellet is near the light source. It would obviously be better the closer it is to the detector.

Actually you can buy lasers for peanuts nowadays so a grid of say 50x50 lasers is not really expensive. (You can get 100 laser diodes for £13!)

Over 100mm thats one beam every 2mm. They'd need to be staggered as the housings are 6mm in diameter.

I may experiment with a single row of 10 lasers and photodiodes, just to see what spacing is required to accurately detect the position to within 0.5mm.

_Shorty
Posts: 8
Joined: Sat Nov 30, 2019 12:07 am

Re: Need High Accuracy Electronic Target using triangulation

Tue Jan 14, 2020 12:52 pm

They certainly could be lasers, but I'm not sure they'd need to be. If you take a look again at their Measuring Principle web page you can see in the bottom picture that they're using a staggered array of sensor holes. And then there are gaps between them, too. I'm assuming the gaps are where the light sources are. 16 holes - gap - 32 holes - gap - 32 holes - gap... The bullet appears to be obscuring five beams/holes to varying degrees. I don't know how big laser parts would be. Could you fit 32 of them in that gap area? Or is their beam illustration merely related to how many detection holes they've got? They claim a resolution of 0.01 mm and accuracy of 0.1 mm over the entire field, with double precision in the middle anyway. "26 light triangles with 768 sensors" seems to indicate it isn't lasers. Recessed/shaded LEDs seem fairly likely. Confine one "light triangle" source to hit only its group of 32 sensor holes.

That link again: https://www.meyton.info/en/measuring-pr ... index.html

n2Jesus
Posts: 1
Joined: Mon Mar 30, 2020 7:00 pm

Re: Need High Accuracy Electronic Target using triangulation

Mon Mar 30, 2020 7:06 pm

I just found this topic trying to do the same thing only with bullets at supersonic speeds.
In my case I know arduinos inside and out, but I'm having trouble finding the "math" needed to pinpoint the location.

Can someone point me at sample code, or the process needed to triangulate with 4 microphones?

thanks,

jim


Return to “General discussion”