Hi!.Can anyone please help me how to get rid of this errors.I herby attach my code and the errors.

function [xfar_in,yfar_in] = fartheta(y_veh, x_veh)
  % c=0.0100;
  % error_in=0;
  % amplitude=1;
  % xfar_in=zeros(2,1);
  % yfar_in=zeros(2,1);
  % xin_new=zeros(2,1);
  % yin_new=zeros(2,1);
  % slope_in=zeros(2,1);
  % slope_curve_in=zeros(2,1);
  x_save = zeros(2,1);
  y_save = zeros(2,1);
  % coder.varsize('x_save',[1, 2 * length(x_veh)]);
  % coder.varsize('y_save',[1, 2 * length(y_veh)]);
  % error_in = 0.001;
  % xin_new = x_veh;
  % yin_new = y_veh;
  for i = 1:length(x_veh)
    xin_new = x_veh(i);
    yin_new = y_veh(i);
    error_in = 0.001;
    while error_in>=0
      xin_new = xin_new + 0.01;
      yin_new = 1*sin(2*pi*xin_new/25);
      slope_in = (yin_new-y_veh)/(xin_new-x_veh);
      slope_curve_in = (1*2*pi/25)*cos(2*pi*xin_new/25);
      error_in = slope_curve_in-slope_in;
      xin_new = xin_new+0.01;
    end
    x_save = [x_save;xin_new];
    y_save = [y_save;yin_new];
  end
  xfar_in = x_save;
  yfar_in = y_save;
  % xfar_in = xin_new;
  % yfar_in = yin_new;
  % d_in = sqrt(((x_veh-xfar_in)^2) + ((y_veh-yfar_in)^2));
  %  
  % m1=tan(hdg);
  % m2=(y_veh-yfar_in)/(x_veh-xfar_in);
  % farangle=atan((m1-m2)/(1+m1*m2));

[Joe Jones]

Dear

Did you solve this problem ?

I also had this problem recently. "An error occurred while propagating data type 'double' through 'adrc/cg_adrc'". (the last problem in the attached picture 'erroes.JPG')

Hope you can help me if possible. Thank you.

Answer 1

[Walter Roberson]

You need to preallocate the output variable and assign into the appropriate locations in it, instead of dynamically growing your array.

[Sanjal C C]

okay can you please give me an example.

[Sanjal C C]

function [xfar_in,yfar_in] = fartheta(y_veh, x_veh)
  % c=0.0100;
  % error_in=0;
  % amplitude=1;
  % xfar_in=zeros(2,1);
  % yfar_in=zeros(2,1);
  % xin_new=zeros(2,1);
  % yin_new=zeros(2,1);
  % slope_in=zeros(2,1);
  % slope_curve_in=zeros(2,1);
  x_save = zeros(2,1);
  y_save = zeros(2,1);
  % coder.varsize('x_save',[1, 2 * length(x_veh)]);
  % coder.varsize('y_save',[1, 2 * length(y_veh)]);
  % error_in = 0.001;
  % xin_new = x_veh;
  % yin_new = y_veh;
  for i = 1:length(x_veh)
    xin_new = x_veh(i);
    yin_new = y_veh(i);
    error_in = 0.001;
    while error_in>=0
      xin_new = xin_new + 0.01;
      yin_new = 1*sin(2*pi*xin_new/25);
      slope_in = (yin_new-y_veh)/(xin_new-x_veh);
      slope_curve_in = (1*2*pi/25)*cos(2*pi*xin_new/25);
      error_in = slope_curve_in-slope_in;
      xin_new = xin_new+0.01;
    end
    x_save = [x_save;xin_new];
    y_save = [y_save;yin_new];
  end
  xfar_in = x_save;
  yfar_in = y_save;
  % xfar_in = xin_new;
  % yfar_in = yin_new;
  % d_in = sqrt(((x_veh-xfar_in)^2) + ((y_veh-yfar_in)^2));
  %  
  % m1=tan(hdg);
  % m2=(y_veh-yfar_in)/(x_veh-xfar_in);
  % farangle=atan((m1-m2)/(1+m1*m2));

both the inputs are of size [2 1] and I should get two outputs of [2 1] each.The outputs are obtained when the while loop condition values are false .Their is problem is in declaration.

[Walter Roberson]

You have

x_save = zeros(2,1);

so you are starting x_save out is a 2 x 1 vector.

You then do

 x_save = [x_save;xin_new];

so you are growing x_save. You should not do that: assign x_save as large as you need to start and do (for example)

x_save(:,i) = xin_new;

The error message was also about line 28. But line 28 is

xin_new = xin_new+0.01

which is certain to have the same size on both sides, so I think you have posted source that is different than the error message.

You need to have a closer look at

        slope_in = (yin_new-y_veh)/(xin_new-x_veh);

y_veh and x_veh are intended to be vectors, so you have vector / vector . Remember that / is the matrix right divide operator, with A / B being similar to A * pinv(B) . If y_veh and x_veh were both 2 x 1 vectors, then the output of slope_in would be 2 x 2. Perhaps you should be indexing particular elements of x_veh and y_veh ?

[Sanjal C C]

function [xfar_in,yfar_in] = fartheta(y_veh,x_veh,error_in)
% c=0.0100;
% error_in=0;
% amplitude=1;
% xfar_in=zeros(2,1);
% yfar_in=zeros(2,1);
% xin_new=zeros(2,1);
% yin_new=zeros(2,1);
% slope_in=zeros(2,1);
% slope_curve_in=zeros(2,1);
x_save = zeros(2,1);
y_save = zeros(2,1);
% coder.varsize('x_save',[1, 2 * length(x_veh)]);
% coder.varsize('y_save',[1, 2 * length(y_veh)]);
% error_in = 0.001;
% xin_new = x_veh;
% yin_new = y_veh;
for i = 1:length(x_veh)

      xin_new = x_veh(i);
      yin_new = y_veh(i);
      error_in = 0.001;
          while error_in>=0
          xin_new = xin_new + 0.01;
          yin_new = 1*sin(2*pi*xin_new/25);

          slope_in = (yin_new-y_veh)/(xin_new-x_veh);
          slope_curve_in = (1*2*pi/25)*cos(2*pi*xin_new/25);
          error_in = slope_curve_in-slope_in;

          xin_new = xin_new+0.01;

          end
          x_save(:,i) = xin_new;
          y_save(:,i) = yin_new;
         error = error_in; this is line 29

end
xfar_in = xin_new;
yfar_in = yin_new;

I change the code accordingly still I have the following errors.

Size mismatch (size [1 x 1] ~= size [2 x 2]).
The size to the left is the size of the left-hand side of the assignment.

Function 'MATLAB Function1' (#47.762.770), line 29, column 9:
"error_in"
Launch diagnostic report.
Component:MATLAB Function | Category:Coder error
Errors occurred during parsing of MATLAB function 'cvm/MATLAB Function1'
Component:MATLAB Function | Category:Coder error
Simulink cannot determine sizes and/or types of the outputs for block 'cvm/MATLAB Function1' due to errors in the block body, or limitations of the underlying analysis. The errors might be inaccurate. Fix the indicated errors, or explicitly specify sizes and/or types for all block outputs.
Component:MATLAB Function | Category:Coder error
Simulink cannot determine sizes and/or types of the outputs for block 'cvm/MATLAB Function1' due to errors in the block body, or limitations of the underlying analysis. The errors might be inaccurate. Fix the indicated errors, or explicitly specify sizes and/or types for all block outputs.
Component:Simulink | Category:Model error
An error occurred while propagating data type 'double' through 'cvm/MATLAB Function1'.
Component:Simulink | Category:Model error

[Walter Roberson]

As you did not change the calculation of slope_in after my warning about what the / operator means, then it must be the case that you are expecting slope_in to become a 2 x 2 value. Because you use slope_curve_in-slope_in with slope_in now expected to be 2 x 2, then you must be expecting error_in to be 2 x 2 as well. But you initialized error_in = 0.001, so when you assign the 2 x 2 result to it, you are trying to grow the existing variable. You would need to initialize

error_in = 0.001 * ones(length(y_veh), length(x_veh));

Remember, though, that the meaning of

while error_in>=0

is the same as

while all( error_in(:)>=0 )

so as soon as one of the entries in the error_in matrix becomes negative the while loop will exit.

[Sanjal C C]

hey,I tried that but still its the error message what I had got before

[Walter Roberson]

I need your updated code.

[Sanjal C C]

function [xfar_in,yfar_in] = fartheta(y_veh,x_veh,error_in)
% c=0.0100;
% error_in=0;
% amplitude=1;
% xfar_in=zeros(2,1);
% yfar_in=zeros(2,1);
% xin_new=zeros(2,1);
% yin_new=zeros(2,1);
% slope_in=zeros(2,1);
% slope_curve_in=zeros(2,1);
x_save = zeros(2,1);
y_save = zeros(2,1);
% coder.varsize('x_save',[1, 2 * length(x_veh)]);
% coder.varsize('y_save',[1, 2 * length(y_veh)]);
% error_in = 0.001;
% xin_new = x_veh;
% yin_new = y_veh;
for i = 1:length(x_veh)

      xin_new = x_veh(i);
      yin_new = y_veh(i);
      error_in = 0.001;
          while all(error_in(:)>=0)
          xin_new = xin_new + 0.01;
          yin_new = 1*sin(2*pi*xin_new/25);

          slope_in = (yin_new-y_veh)/(xin_new-x_veh);
          slope_curve_in = (1*2*pi/25)*cos(2*pi*xin_new/25);
          error_in = slope_curve_in-slope_in;

          xin_new = xin_new+0.01;

          end

%        error = error_in;
end
        x_save(:,i) = xin_new;
        y_save(:,i) = yin_new;
        error_in=0.001*ones(length(y_veh),length(x_veh));
% xfar_in = xin_new;
% yfar_in = yin_new;
% yfar_in = y_save
% xfar_in = xin_new;
% yfar_in = yin_new;
% d_in = sqrt(((x_veh-xfar_in)^2) + ((y_veh-yfar_in)^2));
%  
% m1=tan(hdg);
% m2=(y_veh-yfar_in)/(x_veh-xfar_in);
% farangle=atan((m1-m2)/(1+m1*m2));

Please let me know were I have gone?

[Sanjal C C]

Yeah really sorry.I could not see your comments and I have attached previously my updated code.

[Walter Roberson]

Maybe you want something like,

function [xfar_in,yfar_in] = fartheta(y_veh,x_veh,error_in)

    nveh = length(x_veh);

    x_save = zeros(1, nveh);
    y_save = zeros(1, nveh);

    for i = 1:nveh
      xin_new = x_veh(i);
      yin_new = y_veh(i);
      error_in = 0.001;
      while true
        xin_new = xin_new + 0.01;
        yin_new = 1*sin(2*pi*xin_new/25);

        slope_in = (yin_new-y_veh(i)) ./ (xin_new-x_veh(i));
        slope_curve_in = (1*2*pi/25)*cos(2*pi*xin_new/25);
        error_in = slope_curve_in-slope_in;
        if error_in <= 0
          break;
        end
        xin_new = xin_new+0.01;
      end

      x_save(1,i) = xin_new;
      y_save(1,i) = yin_new;
    end

    and here calculate xfar_in and yfar_in