intersections of curves



On 7/13/07, sen1 at math.msu.edu <sen1 at math.msu.edu> wrote:
> Hello,
>   Using the function 'find_root' one can find an intersection of
>   two explicit curves
>     x -> [x, g(x)]
>     x -> [x, h(x)]
>
> in an interval [a,b] such that g - h have different signs at the
> endpoints.
>
> Does maxima have an analogous routine to find the intersection of two
> parametrized curves
>
> g:  t -> [x(t), y(t)], 0 < t < 1
> h:  s -> [u(s), v(s)], 0 < s < 1
>
> where it is known that such an intersection exists?

In many cases, I'm guessing the answer is  "yes", for the following reasons.
If x,y,u,v are polynomials in t then Grobner bases will
put these curves in algebraic form. Maxima has some Grobner bases routines.
An example is in
http://cadigweb.ew.usna.edu/~wdj/book/node86.html
Once they are in algebraic form, again Grobner bases can be used
to find 0-dimensional solutions.
Does this make sense?


>
> Note: the parameters are different, so any intersection requires the
> solution of a pair of equations If not, does anyone know a good
> reference for such a routine.
>
> I am interested in the case of transcendental functions or high degree
> polynomials.  In the cases of interest to me, newton's method has degeneracies or
> numerical overflows and does not work.  So, I would prefer some sort
> of 'bisection' method.
>
> One special case of interest to me is the one in which the curves have
> a non-degenerate crossing; i.e., have linearly independent tangent
> vectors at the crossing.
>
> TIA,
>
> -sen
>
>
> --
>   ---------------------------------------------------------------------------
>   | Sheldon E. Newhouse            |    e-mail: sen1 at math.msu.edu           |
>   | Mathematics Department         |                                       |
>   | Michigan State University      | telephone: 517-355-9684                |
>   | E. Lansing, MI 48824-1027 USA  |       FAX: 517-432-1562                |
>   ---------------------------------------------------------------------------
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
>