Even Blinder Optimization

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
I'm working on a Kalman filter application, with enough unknowns that
the tuning process is very much seat of the pants.

I've been getting things close, then using gradient optimization to get
better yet.  I'm thinking of using a genetic algorithm (the one that's
built-in to Scilab), because hey -- genetic = magic, so it's got to
work, right?

Has anyone used genetic algorithms for optimization?  Did they do any good?

--

Tim Wescott
Wescott Design Services
We've slightly trimmed the long signature. Click to see the full one.
Re: Even Blinder Optimization



Quoted text here. Click to load it

It is strange to hear the things like that from you.
By definition, Kalman solution is optimal. How could it be any better?
Gradient optimization = LMS algorithm. Where is Kalman in this picture?

Quoted text here. Click to load it

Genetic algorithms are intended for the discrete problems with
unmanageable number of states. It is an attempt to solve every problem
by brute force without thinking. Sometimes it works, but the
intellectual approach is always better.

If you have a continuous dependence from your parameters, try
optimization from random starting points; this will give the areas of
attraction of the extremums.

Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

Re: Even Blinder Optimization
Quoted text here. Click to load it

The Kalman solution is optimal when your model matches reality well.
I'm working on a hybrid unscented/extended Kalman filter that would have
Kalman hiding his eyes and muttering imprecations under his breath.

Add to that the fact that I'm using sensors that are made by the
customer, and whose performance is not well defined, and GPS data that
is both constantly changing has quality metrics that are both
ever-changing and unknown.

So my model doesn't match reality very well.  Hence, the
seat-of-the-pants tuning.

Quoted text here. Click to load it

I'm using blind optimization to find the best tuning parameters, to make
the Kalman perform as best as it can.  Or at least I'm using it to
refine the tuning parameters.

Quoted text here. Click to load it

The intellectual approach has been breaking down in the face of the
complications, but I'd been kind of thinking that I'd go hit the books
again and see what I could see.

I've got 26 parameters that I'm trying to optimize.  I rarely try to do
all of them in one go, but that's certainly getting close to an
unmanageable number of states for me!

Quoted text here. Click to load it

I may try that, just to map out where the local minima are.  For now,
I've got the computer chewing away on a genetic algorithm (it's low cost
to me -- it only took about 30 minutes to get the wrapper written, so
all I'm paying for is computer time).

--

Tim Wescott
Wescott Design Services
We've slightly trimmed the long signature. Click to see the full one.
Re: Even Blinder Optimization



Quoted text here. Click to load it

For the minimization of the continuous function in multidimensions, you
may want to look at Powell's algorithm (see Numerical Recipies). If the
function is well behaved, the algorithm makes for quadratic convergence
to the nearest minimum. It doesn't take much effort from your side to
employ the algorithm.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

Re: Even Blinder Optimization
Quoted text here. Click to load it
Whoops -- I meant this for the dsp group.

--

Tim Wescott
Wescott Design Services
We've slightly trimmed the long signature. Click to see the full one.
Re: Even Blinder Optimization



Quoted text here. Click to load it

Genetic optimization is intended for discrete functions.
For the minimization of the continuous function in multidimensions, you
may want to look at Powell's algorithm (see Numerical Recipies). If the
function is well behaved, the algorithm makes for quadratic convergence
to the nearest minimum.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

Re: Even Blinder Optimization
Quoted text here. Click to load it

GAs is the absolutely last resort, when you have tried
*everything* else but have not got anywhere near a usable
result.

I'd got to any lengths to *avoid* using a GA. Try something
like simulated annealing first. Far simpler, far faster, and
far more likely to obtain a useful result.

Rune

Site Timeline