How to assign corresponding Z value to gridded X,Y data ?

Hello all,

Having a problem with interpolating the data. I have a live data stream project where data comes from the serial port. After several steps, I want to create a XYZ grid from easting,northing and depth values.

Currently, script creates a surface in real time with "fit" function. 3D surface is visible in the figure and it is updating, so I reckon I have my interpolated surface. Then creating a meshgrid [X,Y] at interested grid size successfully. Then I tried to use "interp2" function to interpolate Z values at grid coordinates(created in meshgrid). But it's not working. Shared the error below as well. How can the matlab produce Z grid corresponding to the X Y grid (which is produced in meshgrid) ?

Kind Regards,

Alper

drawT=0;
while true
 ...
     lla = [lla; Easting, Northing, Depth];

     HRZntl=[lla(:,1) lla(:,2)];
     
     if drawT == 10 % once every 10 input data
        figure(1)
        clf
         
        [X, Y]=meshgrid(min(lla(:,1)):0.1:max(lla(:,1)), min(lla(:,2)):0.1:max(lla(:,2))); %Gridding the data at .1 meters
        
        fitObj=fit(HRZntl,lla(:,3),'linearinterp'); % fitting the surface to my Easting, Northing and Depth data and it works
        
        Z=interp2(lla(:,1),lla(:,2),lla(:,3),X,Y); % Giving Error here
        
        plot(fitObj,HRZntl,lla(:,3));



        drawT=0;
        axis tight
        grid on
    
    end
       
end

Once this code runs, I receive the error below;

Error using griddedInterpolant
The number of input coordinate arrays must match the dimensions of the sample values.

Error in interp2>makegriddedinterp (line 226)
    F = griddedInterpolant(varargin{:});

Error in interp2 (line 126)
        F = makegriddedinterp({X, Y}, V, method,extrap);

Error in RealTime_SBES_DataCollection_and_Mapping (line 108)
        Z=interp2(lla(:,1),lla(:,2),lla(:,3),X,Y);
        
        

Answer 1

[Scott MacKenzie]

Try replacing interp2 with griddata, like this:

[X,Y] = meshgrid(min(lla(:,1)):0.1:max(lla(:,1)), min(lla(:,2)):0.1:max(lla(:,2))); 
Z = griddata(lla(:,1),lla(:,2),lla(:,3),X,Y);

If the original data are in x, y, and z vectors, as is the case here, then griddata is the best choice for interpolation.

[Mustafa Alper Cetintas]

Thanks for persisting Sir! I can not be more appreciated than this for your precious help. This worked well! Attached the latest trial's screenshots here, all components of position matrice are identical-sized now, as the original data is in X,Y and Z vectors.

[Scott MacKenzie]

@Mustafa Alper Cetintas You're welcome. Glad to help. Good luck with your research.