Does matlab have any catenary curve fitting toolbox ?

I would like to know the possible options matlab has for catenary curve fitting as toolbox

[VIGNESH BALAJI]

@John D'Errico I understand that the use case is limited. I was just curious whether there was an existing tool which I was unaware of. I have seen users writting tools and promoting it in the community. I saw few tools like -

1. https://nl.mathworks.com/matlabcentral/fileexchange/87584-object-oriented-tools-for-fitting-conics-and-quadrics
2. https://nl.mathworks.com/matlabcentral/fileexchange/38550-catenary-hanging-rope-between-two-points

I understood that it is not excistent, so I am in the process of writting a tool by combining good features of other tools and putting it back in the community.

Answer 1

[Matt J]

fit

lsqcurvefit

[VIGNESH BALAJI]

fit uses the trust region algorithm (kind of damped least squares) and lsqcurvefit uses non linear least squares algorithm which among them is better.

P.S - I have already used fit but I couldn't constrain the length of the fit as fit tries to fit an analytical function and one of the ways to do it is to plot the curve for a specified region in x but that diesn't exactly constrain the length of a curve either.

[Matt J]

but I couldn't constrain the length of the fit

The fact that you have constraints is important information missing from your original post. In any case, as John said, you can use pretty much an y nonlinear optimizer that supports constraints, e.g., fmincon.

[VIGNESH BALAJI]

@Matt J thanks for the suggestion. A quick clarification, fmincon is an nonlinear optimizer with constraints and whereas fit is also an non linear optimizer without constraints right ?

[Matt J]

fit() can process simple upper and lower bounds, but no constraints more complicated than that.

Answer 2

[John D'Errico]

Adding to what @Matt J suggested, you might use nlinfit, lsqnonlin, or any general optimizer, since you can always form a sum of squares.

Why would MATLAB need a specific catenary fitting tool? Note that your last many questions, all about catenary fitting are literally the only ones I recall about fitting a catenary. So why should there be some specific tool for the purpose? You don't write specific tools and toolboxes just in case one person comes along every 30 years. And worse, your specific problem seems to be one that is very non-standard, where the curve is rotated in space. For something like that, you just write code to fit the need, instead of hoping for some never used tool to exist in advance.