Does one need to normalize nonlinear inequality constraints in fmincon?

1 回表示 (過去 30 日間)
Amir Patel
Amir Patel 2014 年 3 月 20 日
Hi Everyone
I am using fmincon with the 'sqp' algorithm to solve an optimal trajectory. I initially tried 'interior-point' to no avail. My input profile/trajectory is discretised to 60 points. Thus, I have 60 variables for my optimization to run.
I am struggling to converge to a good solution and often get this message during my GlobalSearch run:
Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 3.010150e-19.
> In C:\MATLAB\R2012a Student\toolbox\optim\optim\private\stopTestSQP.p>stopTestSQP at 204
I have normalised and centered by variables but should I be normalising my constraints as well? eg. in my case my angles must be less than 90degree. Should I normalise and make them less than 1?
Or does this error relate to something else?
Thanks in advance, Amir

回答 (0 件)

カテゴリ

Help Center および File ExchangeSolver Outputs and Iterative Display についてさらに検索

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by