All you Math sizes, Here is One For You Revisited

[JÞB]

Inspired by Greg I'll give you all an oportunity to solve my problem for me

Given distance between tracks and normalized forces solve for X, Y and Rho.

Assume I'm writing a simultor for motion of a tracked vechile of some sort

[inniwhoni]

Hi Jeff,

Do you know what this would look like as a mathematical algorithm? I can try and help out with how that’ll look in LabVIEW.

Kevin

[Bob_Schor]

I like math problems, but what is "rho"?  Since it is a "tracked vehicle", I'm assuming it is not a mis-spelling of "Row, row, row your boat."  I'm also unsure that I understand the other parameters (max speed and dt I understand, but not the other three).  How does "Rel Vel" (relative velocity?  relative to what?) correspond to X and Y (I assume it is a derivative, but am still puzzled by "relative")?  Are there other constraints (perhaps having to do with "tracks")?

Bob Schor

[altenbach]

I assume rho is the angle of direction. What are the forces normalized to? Where are the forces entered? is dt fixed or variable? Where do the control values come from? What are the units? Is this a spacecraft (kMPH is pretty fast!!! :D)?

I guess it does the steering by varying the relative speed to the right and left wheel, so the track width is probably constant over the course, right?

We need significantly more information. In any case, using complex numbers would make things easier.

[JÞB]

OK, that Max Speed is in Kilometers per hour   

The distance between the tracks will not vary for a given vehicle but there may be many sizes of these vehicles in the future.

Relative velocity is essentially throttle percent where with both Left and Right tracks at 1 (Max Forward thrust) the vehicle will attain max speed and no rotation will be seen

Connversly, with Left =1 and Right =-1 the darn thing will spin in place Clockwise (as seen looking down) and the distance between the tracks defines the circumference the tracks travel along.

Rho, in this case is 0-360 Degrees and is the change in angle from starting position

dT is how long since the last value was calulated and may not be constant (if we can get the RT code optomized well tighten up our control loop)

In all I just want to know how far it went and which way its pointing (Useful information for any vehicle in motion)

Yes its, fairly simple trig but I figured someone up here would have an elegant solution and I'm slinging "more urgently needed" wires.

For a example you could look here

[altenbach · Accepted solution]

Here's a very quick draft (LabVIEW 2013). Please double-check the math. 😄

(dT is not implemented. A simple solution would be to multiply it with the number going to the N of the FOR loop.)

[jcarmody]

---

@altenbach wrote:

Here's a very quick draft (LabVIEW 2013). [...]

---

That's wonderful!  Please add enemies, guns and a scoring system!

[JÞB]

Absolutely beatiful code!  

At least I was on the right "Basic" math track but bow to my acknwoledged superior in all thinking in complex and multiple coordinate systems at the same time.  

[altenbach]

The first draft had the second element of the state cluster also as complex (speed, direction), but that was dumb because the complex numbers are internally stored in (RE,IM), not (R, theta) of course. This makes it forget the current direction whenever the speed went to zero. Dohh!!!

Would it be better if complex numbers were stored internally as a (R,theta) pair instead? Oh well, too late for that and I am sure it would have other disadvantages.. 😄

[JÞB]

Actually, with the exception of the kPH to m/mSec error all I needed to do was change dT to mSec (I get counts anyway)  D'Oh and multiply by the delta per ticks.

The real machine is a bit slothful so a 40000 m^2 grid was a bit large.

Again- Much appreciated