Prev: Incorporating data from a variable in xlswrite filename
Next: bug in treatment of A(I)=B?
From: eli on 13 Aug 2010 09:02
I am trying to solve an orbit optimization problem using direct method .

basic strategy:
formulate discrete equations of motion using cubic interpulation

augmanting the state vextor: Z=[x u]
where x: is the original state vector (6 variables) and u is the control vector (2 variables)

my question:
I have no rule for creating the u vector- how do i formulate the constrains for the solver ? ( i gave only 6 dynamic relations ie the equations of motion, and also bounds on the control)

Thanks
From: Gene on 13 Aug 2010 10:10
"eli " <elikatan(a)t2.technion.ac.il> wrote in message <i43fob$5uq$1(a)fred.mathworks.com>...
> I am trying to solve an orbit optimization problem using direct method .
>
> basic strategy:
> formulate discrete equations of motion using cubic interpulation
>
> augmanting the state vextor: Z=[x u]
> where x: is the original state vector (6 variables) and u is the control vector (2 variables)
>
> my question:
> I have no rule for creating the u vector- how do i formulate the constrains for the solver ? ( i gave only 6 dynamic relations ie the equations of motion, and also bounds on the control)
>
> Thanks

Hi Eli:

There is a fair amount of literature about this sort of formulation for optimal control problems in general and for space trajectory optimization in particular. You might look at the recent book by J.Betts (2010, published by SIAM). For a related approach google
'dido software'.

Good luck

gene
 | 
Pages: 1
Prev: Incorporating data from a variable in xlswrite filename
Next: bug in treatment of A(I)=B?