Indez in position 1 exceeds array bounds. Index must not exceed 2.

Hi,

I am currently writing a code but can not figure out why I am getting the following error:

Index in position 1 exceeds array bounds. Index must not exceed 2 (line 33)

r2=stateserror(i+1,1:3);

All I am trying to do is getting it to run the graph

I attached everything I have maybe somebody can help me.

Code

clc;
clear;
R=[-2436.45; -2436.45; 6891.0379];
V=[5.088611; -5.088611; 0];
x0=[R;V];
x0=x0*1000;
dt=1;
tVector=linspace(0,86400,2);
xfinal=rk4_2(x0,dt,tVector);
format longG
disp(xfinal(end,:))

         -5751485.09511207          4721208.32733256           2045927.6438731         -797.745896575586         -3656.44105964464          6139.64338672577

%random noise
sz=size(xfinal(:,1:3));
error= 1/sqrt(3);
random=randn(sz)*error;
stateserror(:,1:3)=random+xfinal(:,1:3);
stateserror(:,4:6)=xfinal(:,4:6);
%% Gibbs and Herrick loop
i=10;
k=1;
tmax=20*60;
while i<=tmax/2
    
    r1=stateserror(1,1:3);
    r2=stateserror(i+1,1:3);
    r3=stateserror(i+2+1,1:3);
    r1magnitude=(r1);
    r2magnitude=(r2);
    r3magnitude=(r3);
    z12=cross(r1,r2);
    z31=cross(r3,r1);
    z23=cross(r2,r3);
    %Angels
    angle12=acosd(dot(r1,r2)/(norm(r1*norm(r2))));
    angle23=acosd(dot(r2,r3)/(norm(r2*norm(r3))));
    angle13=acosd(dot(r1,r3)/(norm(r1*norm(r3))));
    %Gibbs
    N=r1magnitude.*z23+r2magnitude.*z21+r3magnitude.z12;
    Nmag=norm(B);
    D=z12+z23+z31;
    Dmag=norm(D);
    S=(r2magnitude-r2magnitude).*r1+(r3magnitude-r1magnitude).*r2+(r1magnitude-r2magnitude).*r3;
    B=cross(D,r2);
    Lg=sqrt(mu/(Nmag*Dmag));
    v2G=Lg.*B/r2m+Lg.*S;
    v2Gm=norm(v2gG);
    statesm=norm(states(i,4:6));
    Gibb_error=abs((V2Gm-statesm)/statesm)*100;
    %Herrick 
    t31=2*i;
    t32=i;
    t21=i;
    v2HG=-t32*((1/(t21*t31))+(mu/(12*r1magnitude^3)).*r1+(t32-t21)*(1/(t21*t32))+(mu/(12*r2magnitude^3)).*r2+t21*(1/(t32*t31))+(mu/(12*r3magnitude^3)).*r3);
    v2HGm=norm(v2HG);
    HGibb_error=abs((v2HGm-statesm)/statesm)*100;
    angle(k,1)=angle13;
    k=k+1;
    i=i+10;
end

Index in position 1 exceeds array bounds. Index must not exceed 2.

Answer 1

[Walter Roberson]

R=[-2436.45; -2436.45; 6891.0379];
V=[5.088611; -5.088611; 0];
x0=[R;V];
x0=x0*1000;
dt=1;
tVector=linspace(0,86400,2);
xfinal=rk4_2(x0,dt,tVector);
format longG
disp(xfinal(end,:))
%random noise
sz=size(xfinal(:,1:3));
error= 1/sqrt(3);
random=randn(sz)*error;
stateserror(:,1:3)=random+xfinal(:,1:3);
stateserror(:,4:6)=xfinal(:,4:6);

whos stateserror

stateserror is 2 x 6

%% Gibbs and Herrick loop
i=10;

Notice that i starts at 10

k=1;
tmax=20*60;
while i<=tmax/2

    r1=stateserror(1,1:3);

That is okay because you are accessing fixed row 1 (out of 2)

    r2=stateserror(i+1,1:3);

but i is 10 so that is asking for row 11 (out of 2)

    r3=stateserror(i+2+1,1:3);

and that would ask for row 13 (out of 2)

    r1magnitude=(r1);
    r2magnitude=(r2);
    r3magnitude=(r3);
    z12=cross(r1,r2);
    z31=cross(r3,r1);
    z23=cross(r2,r3);
    %Angels
    angle12=acosd(dot(r1,r2)/(norm(r1*norm(r2))));
    angle23=acosd(dot(r2,r3)/(norm(r2*norm(r3))));
    angle13=acosd(dot(r1,r3)/(norm(r1*norm(r3))));
    %Gibbs
    N=r1magnitude.*z23+r2magnitude.*z21+r3magnitude.z12;
    Nmag=norm(B);
    D=z12+z23+z31;
    Dmag=norm(D);
    S=(r2magnitude-r2magnitude).*r1+(r3magnitude-r1magnitude).*r2+(r1magnitude-r2magnitude).*r3;
    B=cross(D,r2);
    Lg=sqrt(mu/(Nmag*Dmag));
    v2G=Lg.*B/r2m+Lg.*S;
    v2Gm=norm(v2gG);
    statesm=norm(states(i,4:6));
    Gibb_error=abs((V2Gm-statesm)/statesm)*100;
    %Herrick 
    t31=2*i;
    t32=i;
    t21=i;
    v2HG=-t32*((1/(t21*t31))+(mu/(12*r1magnitude^3)).*r1+(t32-t21)*(1/(t21*t32))+(mu/(12*r2magnitude^3)).*r2+t21*(1/(t32*t31))+(mu/(12*r3magnitude^3)).*r3);
    v2HGm=norm(v2HG);
    HGibb_error=abs((v2HGm-statesm)/statesm)*100;
    angle(k,1)=angle13;
    k=k+1;
    i=i+10;
end

and nothing in your code grows stateserror, so it has to be pre-created to be 3 more than the largest possible i value you can get (so, 20*60/2 + 3 == 603)

Perhaps you need a longer time vector?

[Oskar Kinat]

Making the time vector longer worked thank you so much!

The only thing I have left is figure out why I am unable to plot the percent error in v2 for each method vs. the angle of separation between vectors.