Thread Subject:

Error using ==> plus Matrix dimensions must agree.

Dear all,

How should I write for v2 if the error like below occurred. Thanks.
u4=(l*T2)/Iz;
v2=(2*m1/m2.*tan(psi).*psidot)/(cos(psi).*cos(psi))+((m1/m2.*u4)/(cos(psi).*cos(psi)));

??? Error using ==> plus
Matrix dimensions must agree.

Error in ==> utama at 48
v2=(2*m1/m2.*tan(psi).*psidot)/(cos(psi).*cos(psi))+((m1/m2.*u4)/(cos(psi).*cos(psi)));

>>

Tell us which variables are scalars and which are arrays.

ImageAnalyst <imageanalyst@mailinator.com> wrote in message <9b92cd08-c0d9-42e6-b602-2cfe72f0ba6e@e40g2000yqn.googlegroups.com>...
> Tell us which variables are scalars and which are arrays.

I paste here the code

clear all;
global mb rho l r b d Jx Jy Jz Jt m1 m2 m3 Ix Iy Iz T1 T2 theta

mb = 21.43; % mass of the vehicle
rho = 1023.0; % fluid density
l = 0.1; % distance
r = 0.1; % radius
b = 0.068; % thrust factor
d = 3.617e-4; % drag factor
Jx = 0.0857; %roll inertia of the vehicle
Jy = 1.1143; % pitch inertia of the vehicle
Jz = 1.1143; % yaw inertia of the vehicle
Jt = 1.1941e-4; % thruster inertia
%
m1 = mb + 0.394*rho*pi*r^3; %total mass in the x-direction
m2 = mb + 5.96*rho*pi*r^3; %total mass in the y-direction
m3 = mb + 5.96*rho*pi*r^3; %total mass in the z-direction
%
Ix = Jx + 0; %total inertia in the x-direction
Iy = Jy + 24.2648*rho*pi*r^5; %total inertia in the y-direction
Iz = Jz + 24.2648*rho*pi*r^5; %total inertia in the z-direction
T1=2;
T2=2;
theta=pi/4;

%define tstart, tend and the number of time steps you want to take, n:
% tstart = 0;
% tend = 15;
% n = 0.01;
% tspan = linspace(tstart,tend,n);
tspan=[0 15];

%define the initial conditions making sure to use the right ordering
xinit = [2;0;pi;0;2;0;2;0];
%Get x,psi.
[t,x] = ode45(@integratingfunction3, tspan, xinit)
%define the output variables:
xx = x(:,1); % value for x
xxdot = x(:,2); %value for xdot
psi = x(:,3); % psi
psidot = x(:,4); %psidot
z = x(:,5); % value for x
zdot = x(:,6); %value for xdot
y = x(:,7); % psi
ydot = x(:,8); %psidot

% input
u1=cos(theta)*cos(psi)*T1/m1;
% u4=(l*T2-Iy*thetadot*phidot+(Jt*Omega+Ix*phidot)*thetadot)/Iz; % asal u4
u4=(l*T2)/Iz; % bila thetadot=0
v1=u1;
v2=(2*m1/m2.*tan(psi).*psidot)/(cos(psi).*cos(psi))+((m1/m2*u4)/(cos(psi).*cos(psi)));
%
% z11=x;
% z12=xdot;
z21=-m1/m3*tan(theta)*sec(psi);
z22=m1/m3*tan(theta).*sec(psi).*tan(psi).*psidot;
z31=m1/m2*tan(psi);
z32=m1/m2*psidot/(cos(psi).*cos(psi));
% z41=y;
% z42=ydot;
%

xddot=v1;
% psiddot=v2;
zddot=z21.*v1;
yddot=z31.*v1;
xinit1 = [2;0;2;0];

figure(1)
plot(x(:,1),'-','LineWidth',1.5) % x
title('position {\it x}','FontSize',18,'FontName','Times')
figure(2)
plot(x(:,3),'-','LineWidth',1.5) % psi
title('psi {\it {\psi}}','FontSize',18,'FontName','Times')
figure(3)
plot(x(:,5),'-','LineWidth',1.5) % z
title('position {\it z}','FontSize',18,'FontName','Times')
_________________
function dxdt = integratingfunction3(t,x)

global mb rho l r b d Jx Jy Jz Jt m1 m2 m3 Ix Iy Iz T1 T2 theta
%Recall that x=[x, xdot, psi, psidot, z, zdot, y, ydot]
%Define new variables: y=x(1), ydot=x(2), etc
p=x(1);%define parameter x as p --> position x
pdot=x(2); % xdot
psi=x(3);
psidot=x(4);
z=x(5);
zdot=x(6);
y=x(7);
ydot=x(8);
%notice that here x is just a 1 by 4 row vector and not a large array like
%in [t,x] = ode45(@fname, tspan, xinit, options)

u4=(l*T2)/Iz; %apabila thetadot=0
%the array dxdt is the same length as x
dxdt = zeros(size(x));
dxdt(1) = pdot;
dxdt(2) = cos(theta)*cos(psi)*T1/m1; % this is xddot =v1
dxdt(3) = psidot;
dxdt(4) = 2*m1/m2*tan(psi)*psidot/(cos(psi)*cos(psi))+ m1/m2*u4/(cos(psi)*cos(psi)); %this is psiddot
dxdt(5) = zdot;
dxdt(6) = -m1/m3*tan(theta)*sec(psi)*dxdt(2); %zddot
dxdt(7) = ydot;
dxdt(8) = m1/m2*tan(psi)*dxdt(2); %yddot

"zainah " <zainahmz@gmail.com> wrote in message <j0ijn4$73u$1@newscl01ah.mathworks.com>...
> I paste here the code
> ........
> v2=(2*m1/m2.*tan(psi).*psidot)/(cos(psi).*cos(psi))+((m1/m2*u4)/(cos(psi).*cos(psi)));
> .......
- - - - - - - - - -
  Instead of exhibiting your voluminous code, you should simply give us what you think are the dimension sizes of all the variables that occur in the v2 computation line. That is, what were the sizes of m1, m2, psi, psidot, and u4, when the v2 computation took place? If there is doubt in the matter, just insert some appropriate 'size' codes before that computation to display these sizes.

  Only then are you likely to receive an answer to your question.

Roger Stafford

You used / instead of ./
Try this line instead:
v2=(2*m1/m2.*tan(psi).*psidot) ./ (cos(psi).*cos(psi))+((m1/m2*u4) ./
(cos(psi).*cos(psi)));