Hello Julius (hello List),
my constraint condition is that both point clouds should never intersect
each other. In mathematical terms this is written as: c(k,p,w) = n_i *
(R*p_i - x_i) > 0 or < 0 for all i.
If c < 0 or c > 0 is valid is determined for all i before the
minimization problem is started.
x_i is a point from the model point set, p_i from the data point set
(that is being 'moved' to x_i), n_i are the normals for each point x_i
and R is a standard rotational matrix (R \in SO(3)). k,w,p are the
rotational angles (k = rotation around x-axis, p = rotation around
y-axis, w = rotation around z-axis).
And yes, my initial solution is in the feasible region.
Best regards,
Anja
On 06/03/2013 06:16 PM, Julius Ziegler wrote:
Hello Anja,
On 06/03/2013 06:05 PM, Anja Schäfer wrote:
Hello,
I'm currently implementing a variation of the classical ICP algorithm
and it works fine with artificial data. If I use the real-world data I
actually have, the solution produced by the SLSQP algorithm is
infeasible (although there definitely IS a feasible solution).
I double-checked my constraint functions, I start with feasible data and
"I start with feasible data": Does this mean that your initial
solution is in the feasible region? Otherwise, SQP cannot guarantee to
converge to a feasible solution.
Also, remember that SQP still only works for convex problems. Your
objective function is very likely convex (standard ICP is a least
squares problem), but are your constraints? Could you describe what
your constraints are in your application (I know a little bit of ICP
and registration problems.)
Best regards,
Julius
I tried setting max_eval and max_time but nothing seems to work.
Is there anything else I can do?
Thanks
Anja
--
Anja Schäfer
Interdisciplinary Center for
Scientific Computing - IWR
Heidelberg University
Im Neuenheimer Feld 368, Room 531
69120 Heidelberg, Germany
Email [email protected]
Web http://www.iwr.uni-heidelberg.de/groups/agbock/USER_PAGES/SCHAEFER_ANJA/
_______________________________________________
NLopt-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/nlopt-discuss
