displacement measurement

Thread Starter

bug13

Joined Feb 13, 2012
2,002
A Tri-axial accelerometer in the ball and a small uC to act as a data logger (also in the ball) might work. Mems accelerometers are real cheap, so are uC's. Lotsa number crunching ahead.
How would you achieve that? I am also thinking of using a accelerometer, but I don't know how it might work.
 

John P

Joined Oct 14, 2008
2,025
The accelerometer system I mentioned before would tell you the ball's complete state of acceleration and rotation. What you'd get would be 3 dimensions of linear acceleration within the ball's frame of reference, and 3 angular rotation velocities which would let you define the orientation at any moment. At least that's the theory! There's some pretty hairy math between getting the readings and seeing a usable result. But this is a reasonable thing to be attempting, so maybe the work has been done and someone has written it up and made it available on the Internet.
 

sirch2

Joined Jan 21, 2013
1,037
You need gyros as well as accelerometers to get the complete state of rotation (think about rotations around the ceritcal axis) but for this application accelerometers may be enough.

The problem with accelerometers is that to get distance you need to integrate twice and that significantly amplifies the noise, which can give a large degree of inaccuracy.

OP - why does everyting have to be in the projectile? It sounds like you need to give us more detail about where you are trwoing the ball, inside, outside, on the moon?
 

John P

Joined Oct 14, 2008
2,025
The unit I mentioned does include the rate gyros, so in theory it's a complete inertial motion sensor. It's true that double integration is difficult to do without the data being swamped with noise, but that's how inertial navigation works. Note that the time for a complete set of data for this system is extremely short and the user isn't demanding much accuracy, so maybe it can be done with inexpensive hardware.

Edited to say that what I meant by a "complete set of data" was the data derived from one flight of the projectile. Yes, the sensors will drift, but maybe the drift isn't significant over the time that data is being gathered. If you had to navigate a space probe for months then the requirements would be different!
 
Last edited:

ErnieM

Joined Apr 24, 2011
8,377
However, how do you get some reading with a ultrasonic transducer in your suggested solution, I am interested.
A rocket is better as they typically (hopefully?) are pointing UP so your sensor is orientated DOWN.

Basically, a transducer emits a burst of ultrasonic energy. It travels to the ground, bounces back where another transducer senses it, and an amp sends the signal to the micro. Height is then just the time difference times speed of sound over two.

Google "ping))) sensor" to see a bunch of examples. I believe the ping))) itself is only good for a few meters but gives you the basics.
 

studiot

Joined Nov 9, 2007
4,998
Thanks for your reply, all the sensors and electronics need to be in the ball, if that's possible.
If all the works is going to be in the ball, which is in flight, how is anyone going to read it or use it?
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
The accelerometer system I mentioned before would tell you the ball's complete state of acceleration and rotation. What you'd get would be 3 dimensions of linear acceleration within the ball's frame of reference, and 3 angular rotation velocities which would let you define the orientation at any moment. At least that's the theory! There's some pretty hairy math between getting the readings and seeing a usable result. But this is a reasonable thing to be attempting, so maybe the work has been done and someone has written it up and made it available on the Internet.
I can understand the double integration which was suggested by other member in a strange line, but I have never thought of the angular rotation velocities, that looks interesting!

What would you suggest me to search, so I can find out more about this?
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
OP - why does everyting have to be in the projectile? It sounds like you need to give us more detail about where you are trwoing the ball, inside, outside, on the moon?
It would be that I have used the wrong term here, I just assume that if I toss something in the air in an angle of 0 to 90 degree (say 45 degree), that is a projectile.

And I am throwing the ball outdoor.

The idea is, I want my device to do something after I toss/launch it in the air, from a certain distance from me/launch point.

eg. I am standing at Point_Toss, my device/ball will land at Point_Landing, between Point_Toss and Point_Landing, there is a Point_X (still in the air), so I want my device do something (flash a LED?) when Point_X - Point_Toss = Y_meter (+/- error).

Maybe this might help, These need to be in a 3D environment.
 
Last edited:

Thread Starter

bug13

Joined Feb 13, 2012
2,002
Yes on the Moon I think. I have already asked this question, and received no reply.
Well, dip the ball in paint before tossing, then do it on your mother's patio. She will know where it lands.
I'm also quite serious.
You will need to calibrate any measurement system so you will need to think out of the box.
Other possibilities include.
Using a sand pit for landing.
Pegging out with string or grass lining a grid of 1m squares for landing.
Describe the project requirements in more detail for help with the tracking system.
Hi

For second quote above, I got some mixed messages, I am not sure how to response.

For the third quote above, I understand my measurement system need to be calibrate, but I don't understand using a sand pit for landing, and getting 1m squares for landing is going to help me. I want to measure displacement while the ball is till in the air. (or before hitting ground).

If all the works is going to be in the ball, which is in flight, how is anyone going to read it or use it?
I haven't actually thought of it that far yet, I still don't know how to do what I want, but thanks for bringing it up.
 
Last edited:

studiot

Joined Nov 9, 2007
4,998
Say I throw a ball up in the air in an angle (0 - 90 degree), how do I measure the displacement between where I stand to just before the ball land
Your original question.

So let it land and measure the distance to 'just before'.

All the methods I have suggested depend upon marking the landing point in some way.

This will provide accurate calibration. No dead reckoning system ever built has been acurate without such calibration against known points.

I also asked for details of your project. These are the sorts of relevant questions.

Is this just a thought experiment or are you going to build it?

What is the budget?

Is this a one off or what?

If I needed to do this for some sort of work project I would hire or buy some old trisponder or syledis equipement and set it up for the day/week whatever.

Either would track your device in real time, providing the information within the travelling object.

But you have provided few details so others have asked the same question.
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
Your original question.

So let it land and measure the distance to 'just before'.

All the methods I have suggested depend upon marking the landing point in some way.

This will provide accurate calibration. No dead reckoning system ever built has been acurate without such calibration against known points.

I also asked for details of your project. These are the sorts of relevant questions.

Is this just a thought experiment or are you going to build it?

What is the budget?

Is this a one off or what?

If I needed to do this for some sort of work project I would hire or buy some old trisponder or syledis equipement and set it up for the day/week whatever.

Either would track your device in real time, providing the information within the travelling object.

But you have provided few details so others have asked the same question.
Hi,


  • Sorry about the confusion, I think what I really mean was to get the distance while the ball/object still still in the air/mid fight.
  • It started as a thought experiment, but I do intend to built it (or at least it's something I like to mess around), if possible.
  • I haven't thought the budget, my answer: I don't know. maybe something seasonal? I am not rich(just a regular student), but I am willing to spend some money on the project. But I have access to my school's equipment and components (well, that's what I believe anyway)
  • For now, it's one off.
If I needed to do this for some sort of work project I would hire or buy some old trisponder or syledis equipement and set it up for the day/week whatever.
  • That's useful information, I will look into those, but don't think my school have these.
But you have provided few details so others have asked the same question.
  • I don't know what more details I need to provide, but I can answer specify question. (like the ones above)
NOTE: It just started with an silly idea in my mind, I got no plan, and have no idea how to do it, like a blur picture. But the picture is getting clear now, after all those replies and info. And it's getting interesting too, at least that's what I thought anyway. :)
 

wayneh

Joined Sep 9, 2010
17,496
I wouldn't do it with MY iPhone or iPod, but I believe if you chucked one while running the right app, it would come quite close to the answer you seek. It could even display it on the screen.

There are a number of apps - particularly fitness apps - that employ the accelerometer data to help estimate motion when/where GPS cannot do the job. The just-announced iPhone 5S employs a new M chip dedicated to such things.
 
Top