LITTLEBLACKDOG.COM

Posted: Fri Dec 07, 2007 1:31 pm Post subject: Fricken Laser Beams ...

At work, gravel is important. It's expensive, and we use a lot of it (~100,000 yards³ / year). We try and track our gravel volumes pretty carefully, but it isn't an easy thing to do. Piles are hard to measure, loads are hard to measure, and application rates are hard to measure.

When I started working here, we measured gravel piles by walking around them and estimating the area, and height, and then doing some half assed calculations for the gravel volume.

The first thing I did was to add GPS to the picture. We started walking around the piles for the area, and then estimated the height and did some half assed calculations for the volume.

This year we went completely high tech:

We went with a fricken laser beam. It's a laser rangefinder with integrated compass and inclimeter. Each time you pull the trigger, the laser sends the distance, compass heading, and inclination to the GPS unit via bluetooth. The GPS then uses those numbers, it's known position and some basic trig to calculate the position of the laser point in 3 dimensions.

So we mount the laser on a survey pole and start shooting points all over the gravel:

We still walk a path around the pile with the GPS to get a pile outline. Then, at the office, I bring the pile outline and the laser points into our 3D GIS software (ArcScene in this case, although GRASS will do it as well):

After that it is just a matter of connecting the dots, so to speak, and we have an approximation of the gravel pile:

And before you say anything, the pile in background of the shot above, is not the pile I used as an example here. Here is a photo/model example:

End result is, there is nearly 20,000 Yards³ of gravel in that pile. Our old half assed estimate showed nearly 25,000 yards³. According to the crushing/hauling records, it should be somewhere around 19,000 Yards³.

Next year I'm planning to do things a little differently I think. First I'm going to setup the laser in fewer positions, and then shoot more points from each position. Second, I think I will just feed the laser info straight into the field laptop, and then do the trig calculations using a surveyed set of coordinates, instead of using the realtime GPS. I have to discard too many of the points using realtime GPS because of the poor GPS signals.

For the first time out, I'm pretty happy with how well it worked.

Posted: Fri Dec 07, 2007 3:47 pm Post subject:

Thats pretty cool!

Leg Humper Joined: 11 Oct 2003 Posts: 5997

[quote]and then do the trig calculations using a surveyed set of coordinates, [code]

cough cough ... calculus.

Guide Dog Joined: 08 Nov 2003 Posts: 8128 Location: MN

[quote="foxinhand"]

Quote:
and then do the trig calculations using a surveyed set of coordinates, [code]

cough cough ... calculus.

not nesicarially... it you could do it with trig, complex though it be... with enough of the right sets of data points...

Though finding the volume under a 3 dimensional function created by putting those data points against a line/curve of best fit would be remarkably easier. but there is trig involved in creating the line of best fit... more then likely if your using calc, your probably using trig too...

Posted: Fri Dec 07, 2007 8:50 pm Post subject:

[quote="anglachel"]

foxinhand wrote:
Quote:
and then do the trig calculations using a surveyed set of coordinates, [code]

cough cough ... calculus.

not nesicarially... it you could do it with trig, complex though it be... with enough of the right sets of data points...

Though finding the volume under a 3 dimensional function created by putting those data points against a line/curve of best fit would be remarkably easier. but there is trig involved in creating the line of best fit... more then likely if your using calc, your probably using trig too...

No manual trig or calculus necessary. Once the digital model is created wouldn't the software figure out the volume for you?

Guide Dog Joined: 08 Nov 2003 Posts: 8128 Location: MN

gregw wrote:
anglachel wrote:
foxinhand wrote:
Quote:
and then do the trig calculations using a surveyed set of coordinates,

cough cough ... calculus.

not nesicarially... it you could do it with trig, complex though it be... with enough of the right sets of data points...

Though finding the volume under a 3 dimensional function created by putting those data points against a line/curve of best fit would be remarkably easier. but there is trig involved in creating the line of best fit... more then likely if your using calc, your probably using trig too...

No manual trig or calculus necessary. Once the digital model is created wouldn't the software figure out the volume for you?

Yeah, but how does the computer to it?

Posted: Sat Dec 08, 2007 6:08 am Post subject:

anglachel wrote:
gregw wrote:
anglachel wrote:
not nesicarially... it you could do it with trig, complex though it be... with enough of the right sets of data points...

Though finding the volume under a 3 dimensional function created by putting those data points against a line/curve of best fit would be remarkably easier. but there is trig involved in creating the line of best fit... more then likely if your using calc, your probably using trig too...

No manual trig or calculus necessary. Once the digital model is created wouldn't the software figure out the volume for you?

Yeah, but how does the computer to it?

With magic pixie dust of course! Rolling Eyes

Posted: Sat Dec 08, 2007 7:31 am Post subject:

First, it is trig, not calculus that calculates the 3d coordinates. It's very basic.

First step is to calculate the vertical change, and the horizontal distance. This is easy trig because we have the length of the hypotenuse (the laser beam) and the inclination angle. Next we take the horizontal distance, and use it to calculate the delta x and delta y. Again, basic trig because we have the length of the hypotenuse (horizontal distance) and we have the compass heading angle.

Once we have the the three delta's (x,y,z), we take the known GPS position (the right hand point of both triangles) and add the three deltas to get the position where the laser beam hit.

Second, calculating the volume under an irregular 3D model is not easy, even for a computer. Also, I cant just calculate the volume under the model, because the ground isn't necessarily flat. I take the outline polygon and create an estimated bare surface based on that, and then calculate the difference between the bare surface and the gravel pile. The software does this by dividing the models into a series of tiny columns and calculates the volume of the columns. It actually rasterizes the model with the raster pixel value being equal to the volume.

Posted: Sun Dec 09, 2007 11:38 am Post subject:

overlapping the data with results from a different spectrum would show

Guide Dog Joined: 08 Nov 2003 Posts: 8128 Location: MN

hummer010 wrote:
First, it is trig, not calculus that calculates the 3d coordinates. It's very basic.

First step is to calculate the vertical change, and the horizontal distance. This is easy trig because we have the length of the hypotenuse (the laser beam) and the inclination angle. Next we take the horizontal distance, and use it to calculate the delta x and delta y. Again, basic trig because we have the length of the hypotenuse (horizontal distance) and we have the compass heading angle.

Once we have the the three delta's (x,y,z), we take the known GPS position (the right hand point of both triangles) and add the three deltas to get the position where the laser beam hit.

Second, calculating the volume under an irregular 3D model is not easy, even for a computer. Also, I cant just calculate the volume under the model, because the ground isn't necessarily flat. I take the outline polygon and create an estimated bare surface based on that, and then calculate the difference between the bare surface and the gravel pile. The software does this by dividing the models into a series of tiny columns and calculates the volume of the columns. It actually rasterizes the model with the raster pixel value being equal to the volume.

so computer calculus, can't differentiate really, though it can do enough calculations (in thing strips) to give the appearance of it.

unless of course, they did do the calculus already in the app, and just have a generic formula ready to go for it to crunch through...

though one thing is for certain, it would be alot cooler if that was a space based laser that took shots from above before and after the gravel pile to do it's calculations.

Posted: Sun Dec 09, 2007 2:04 pm Post subject:

My father in law started using lasers a few years back to lay all the utility pipes. Water, sewer, gas, etc..... He said it was the best investments he had ever made.

Posted: Mon Dec 10, 2007 7:20 am Post subject:

anglachel wrote:
... it would be alot cooler if that was a space based laser that took shots from above before and after the gravel pile to do it's calculations.

True. When we first started looking at doing something like this, I pitched the idea of acquiring a remote control drone helicopter with a down looking LIDAR unit. We could have even thrown a digital camera on there, and a GPS receiver and captured our own high resolution aerial photographs.

In truth, it had a number of practical uses within our operation, and I had two managers who really wanted it, but neither of them wanted to open up the purse strings to buy it.