Errors in 'MATLAB Function' Block in a Simulink model for a Polynomial Trajectory (Matlab 2019a)

I have a model here to calculate the trajectory for a vehicle obstacle evasion manoeuvre. When I run the Simulink model I get this errors:

'Output 'sum_vector' has variable size but the upper bound is not specified; explicit upper bound must be provided.

Component:MATLAB Function | Category:Coder error'

'Error occurred in 'Ausweich_Wael/MATLAB Function'.

Component:Simulink | Category:Model error'

The code runs fine when I run it as a matlab file it gives out the required plot, but not in Simulink. Worked on it for a long time but couldn't find a solution. Hope somebody can help with this, I really require to finish the model urgently.

This is the Simulink model:

and following is the code in Matlab function:

%   This function calculates the summation 'Σ[g_i(t/t_e)^i]' in the
%   evasion trajectory Polynomial.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%Function inputs: n = Order of the System, t_e = Time taken to complete manoeuvre

function sum_vector = fcn(n, t_e)

t_0 = 0;    %Time at Start of the evasive maneuver [s]
t_s = 0.1;  %Sample time

t = t_0:t_s:t_e;  % t = current time/actual time (from start of manoeuvre to end with an increment     of t_s)
sum_vector=zeros(1,length(t));

switch n                %select the g_i corresponding to Order 'n' of the System
    case 1
       g_i = [3 -2];
    case 2
       g_i = [10 -15 6];
    case 3
       g_i = [35 -84 70 -20];
    case 4
       g_i = [126 -420 540 -315 70];
    case 5    
       g_i = [462 -1980 3465 -3080 1386 -252];
    case 6
       g_i = [1716 -9009 20020 -24024 16380 -6006 924];
    case 7
       g_i = [6435 -40040 108108 -163800 150150 -83160 25740 -3432];
end


for j = 1:length(t)
    sum = 0;
    for i = n+1:(2*n)+1
        sum = sum + g_i(i-n)*((t(j)/t_e)^i);       
    end
    sum_vector(j) = sum;
end

Answer 1

[stozaki]

I think that 'sum_vector' is variable size output. So, it should define as 'coder.varsize('sum_vector',[min max])'.

I have configured to avoid the error, but I am not sure if the algorithm is correct.

Please refer following documentation.

Specify Upper Bounds for Variable-Size Arrays

Declare Variable-Size Inputs and Outputs

Regards,

stozaki