Can't understand why does it say "The expression to the left of the equals sign is not a valid target for an assignment."

1-  function dydt = lagrang(~, y)
2-  syms m1 m2 m3 l1 l2 l3 g
3-  dydt = zeros(6,5,4,3,2,1);
4-  m1=1; m2=1; m3=1; l1=1; l2=1; l3=1; g=9.81;
5-  dydt(1)=y(2);
6-  dydt(3)=y(4);
7-  dydt(5)=y(6);
8-  dydt(1)*(l1^2*(m1+m2+m3))+dydt(3)((m2+m3)*l1*l2*cos(y(1)-y(3)))+dydt(5)*(m3*l1*l3*cos(y(1)-y(5)))=-(m2+m3)*l1*l2*y(4)^2*sin(y(1)-y(3))-m3*l1*l3*y(6)^2*sin(y(1)-y(5))-(m1+m2+m3)*g*l1*cos(y(1));
9-  dydt(1)*((m2+m3)*l1*l2*cos((y(1)-y(3))))+dydt(3)*(l2^2*(m2+m3))+dydt(5)*(m3*l2*l3*cos(y(3)-y(5)))=(m2+m3)*l1*l2*y(2)^2*sin(y(1)-y(3))-m3*l2*l3*y(6)^2*sin(y(3)-y(5))-(m2+m3)*g*l2*cos(y(3));
10- dydt(1)*(m3*l1*l3*cos(y(1)-y(5)))+dydt(2)*m3*l2*l3*cos(y(3)-y(5))+dydt(l3^2*m3)=m3*l1*l3*y(2)^2*sin(y(1)-y(5))+m3*l2*l3*y(4)^2*sin(y(3)-y(5))-m3*g**l3*cos(y(5))


%other part
dydt = [0; pi/2; 0; pi/2; 0; pi/2];
[t1,y1] = ode45(@lagrang, [0 20], dydt);
plot(t1,y1(:,2));

When I run this, I get a problem. Sorry if this is a simple problem to solve, since I am new to this program, it's been one week for me since i started.

Error: File: lagrang.m Line: 8 Column: 98
The expression to the left of the equals sign is not a valid target for an assignment.

Error in odearguments (line 90)
f0 = feval(ode,t0,y0,args{:});   % ODE15I sets args{1} to yp0.

Error in ode45 (line 115)
    odearguments(FcnHandlesUsed, solver_name, ode, tspan, y0, options, varargin);

Error in lagrange_ans (line 2)
[t1,y1] = ode45(@lagrang, [0 20], dydt);

The equations' itself:

Where do i go wrong?

[madhan ravi]

the assigning of equations are wrong

Answer 1

[madhan ravi]

syms theta1(t) theta2(t) theta3(t)
 m1=1; m2=1; m3=1; l1=1; l2=1;
 l3=1; g=9.81; tau1=1; tau2=1; tau3=1;
e1= diff(theta1,2)*(l1^2*(m1+m2+m3)+l1)+...
    diff(theta1,2)*((m2+m3)*(l1*l2*cos(theta1-theta2)))+...
    diff(theta3,2)*(m3*l1*l3*cos(theta1-theta3))==...
    -(m2+m3)*(l1*l2*diff(theta2)^2*sin(theta1-theta2))-...
    m3*l1*l3*diff(theta3)^2*sin(theta1-theta3)-...
    (m1+m2+m3)*g*l1*cos(theta1)+tau1-tau2;
e2= diff(theta1,2)*((m2+m3)*l1*l2*cos(theta1-theta2))+...
    diff(theta2,2)*(l2^2*(m2+m3)+l2)+...
    diff(theta3,2)*(m3*l2*l3*cos(theta2-theta3))==...
    (m2+m3)*l1*l2*diff(theta1)^2*sin(theta1-theta2)-...
    m3*l2*l3*diff(theta3)^2*sin(theta2-theta3)-...
    (m2+m3)*g*l2*cos(theta2)+tau2-tau3;
e3= diff(theta1,2)*(m3*l1*l3*cos(theta1-theta3))+...
    diff(theta2,2)*(m3*l2*l3*cos(theta2-theta3))+...
    diff(theta3,2)*(l3^2*m3+l3)==...
    m3*l1*l3*diff(theta1)^2*sin(theta1-theta3)+...
    m3*l2*l3*diff(theta2)^2*sin(theta2-theta3)-...
    m3*g*l3*cos(theta3)+tau3;
vars = [theta1(t); theta2(t); theta3(t)]
V = odeToVectorField([e1,e2,e3])
M = matlabFunction(V,'vars', {'t','Y'})
interval = [0 20];  %time interval
y0 = [0; pi/2; 0; pi/2; 0; pi/2];   %initial conditions
ySol = ode45(M,interval,y0);
tValues = linspace(interval(1),interval(2),1000);
for i = 1:6 % denotes the number of solutions
yValues = deval(ySol,tValues,i); % i denotes the solution number
plot(tValues,yValues)
hold on
end

Plot of 6 solutions:

[Arda Nova]

When I wanted help, I didn't expect this much help. You opened a new world for me, that I can understand. I am really grateful. Thank you very much.

[madhan ravi]

Anytime :) , glad was of help , if you wish you can vote for the answer too.

[Arda Nova]

Of course, I am sorry, I didn't know. :)

[madhan ravi]

No problem , thank you!

[Arda Nova]

You are welcome. :)

Answer 2

[Steven Lord]

When you use the equals sign to perform assignment, the left-hand side must be a variable or a "piece" of a variable. The following work:

x = 42; % Assigning to a variable

y([2 3 5]) = [1 2 3] % Assigning to a "piece" of a variable

S.z(3) = -999 % Assigning to a piece of a field of a struct array

Let's look at your line 8. I've broken it into pieces using ... so we can read it without wrapping.

dydt(1)*(l1^2*(m1+m2+m3)) ...
+dydt(3)((m2+m3)*l1*l2*cos(y(1)-y(3))) ...
+dydt(5)*(m3*l1*l3*cos(y(1)-y(5))) = ...
-(m2+m3)*l1*l2*y(4)^2*sin(y(1)-y(3)) ...
-m3*l1*l3*y(6)^2*sin(y(1)-y(5))...
-(m1+m2+m3)*g*l1*cos(y(1));

You're trying to assign the result of the computations on the right side of the equals sign to another expression. MATLAB doesn't know what you're trying to assign here. Are you trying to update dydt(1), dydt(3), and dydt(5)? Are you trying to update y(1), y(3), and y(5)?

Because MATLAB can't figure out what you're trying to do, it throws an error. You will need to rewrite your higher order ODE as a system of first-order ODEs as shown in the "Solve Nonstiff Equation" example on the documentation page for ode45. Depending on whether some of those values are constants you may want to use a mass matrix to make rewriting your system a bit easier. See the fem2ode and batonode example files that show how to use a mass matrix.

[Arda Nova]

I am very new to these ideas. I am aware, something is wrong but can't really pinpoint it. I have looked at it before but, couldn't make the samething on my equations. I would work this hard on this, would try to solve it mysellf in time, but the teacher gave the project a bit too late to pass the learning process. I am trying to learn matlab in everyway possible in 1 week, which is not enough to use it properly.

[Steven Lord]

Have you tried the MATLAB Onramp course on this page? That's designed to teach you the basics quickly and I believe it's only two or three hours long (with exercises and videos.)

[Arda Nova]

I didn't but I will. Although it's too late for project, I want to learn the program properly. Thank you.

Answer 3

[KSSV]

Don't use syms

function dydt = lagrang(~, y)
dydt = zeros(6,5,4,3,2,1);
m1=1; m2=1; m3=1; l1=1; l2=1; l3=1; g=9.81;
dydt(1)=y(2);
dydt(3)=y(4);
dydt(5)=y(6);
dydt(1)*(l1^2*(m1+m2+m3))+dydt(3)((m2+m3)*l1*l2*cos(y(1)-y(3)))+dydt(5)*(m3*l1*l3*cos(y(1)-y(5)))=-(m2+m3)*l1*l2*y(4)^2*sin(y(1)-y(3))-m3*l1*l3*y(6)^2*sin(y(1)-y(5))-(m1+m2+m3)*g*l1*cos(y(1));
dydt(1)*((m2+m3)*l1*l2*cos((y(1)-y(3))))+dydt(3)*(l2^2*(m2+m3))+dydt(5)*(m3*l2*l3*cos(y(3)-y(5)))=(m2+m3)*l1*l2*y(2)^2*sin(y(1)-y(3))-m3*l2*l3*y(6)^2*sin(y(3)-y(5))-(m2+m3)*g*l2*cos(y(3));
dydt(1)*(m3*l1*l3*cos(y(1)-y(5)))+dydt(2)*m3*l2*l3*cos(y(3)-y(5))+dydt(l3^2*m3)=m3*l1*l3*y(2)^2*sin(y(1)-y(5))+m3*l2*l3*y(4)^2*sin(y(3)-y(5))-m3*g**l3*cos(y(5))
%other part
dydt = [0; pi/2; 0; pi/2; 0; pi/2];
[t1,y1] = ode45(@lagrang, [0 20], dydt);
plot(t1,y1(:,2));

[Arda Nova]

Wasn't there at first. Added it later. I did delete it now and tried it again. Same result.

[KSSV]

What input you tried? What is y?

[Arda Nova]

In my mind y(1)=theta1 y(2)=thetadot1 y(3)=theta2 y(4)=thetadot2 y(5)=theta3 y(6)=thetadot3. I can't think of a better way to be honest.

[KSSV]

Check your code propely.......there are lot of typo errors.....this line:

dydt(1)*(l1^2*(m1+m2+m3))+dydt(3)((m2+m3)*l1*l2*cos(y(1)-y(3)))+dydt(5)*(m3*l1*l3*cos(y(1)-y(5)))=-(m2+m3)*l1*l2*y(4)^2*sin(y(1)-y(3))-m3*l1*l3*y(6)^2*sin(y(1)-y(5))-(m1+m2+m3)*g*l1*cos(y(1));

ana many othr lines...you have used = for -.....and you have used ** for *. Try with the below:

m1=1; m2=1; m3=1; l1=1; l2=1; l3=1; g=9.81;
dydt(1)=y(2);
dydt(3)=y(4);
dydt(5)=y(6);
dydt(1)*(l1^2*(m1+m2+m3))+dydt(3)*((m2+m3)*l1*l2*cos(y(1)-y(3)))+dydt(5)*(m3*l1*l3*cos(y(1)-y(5)))-(m2+m3)*l1*l2*y(4)^2*sin(y(1)-y(3))-m3*l1*l3*y(6)^2*sin(y(1)-y(5))-(m1+m2+m3)*g*l1*cos(y(1));
dydt(1)*((m2+m3)*l1*l2*cos((y(1)-y(3))))+dydt(3)*(l2^2*(m2+m3))+dydt(5)*(m3*l2*l3*cos(y(3)-y(5)))-(m2+m3)*l1*l2*y(2)^2*sin(y(1)-y(3))-m3*l2*l3*y(6)^2*sin(y(3)-y(5))-(m2+m3)*g*l2*cos(y(3));
dydt(1)*(m3*l1*l3*cos(y(1)-y(5)))+dydt(2)*m3*l2*l3*cos(y(3)-y(5))+dydt(l3^2*m3)-m3*l1*l3*y(2)^2*sin(y(1)-y(5))+m3*l2*l3*y(4)^2*sin(y(3)-y(5))-m3*g*l3*cos(y(5))
%other part
dydt = [0; pi/2; 0; pi/2; 0; pi/2];
[t1,y1] = ode45(@lagrang, [0 20], dydt);
plot(t1,y1(:,2));

dydt = zeros(6,5,4,3,2,1);

The above has no meaning int he code......that should be:

dydt = zeros(6,1);

[Arda Nova]

used = for - ??

used ** for * ??

I couldn't spot that kind of mistakes? I honestly don't know what you say. Maybe missing something I should know.

"dydt = zeros(6,1);" I figured this out myself too, but couldn't put in the code in proper way to be honest.

[Arda Nova]

function dydt = lagrang(~, y)
dydt = zeros(6,5,4,3,2,1);
m1=1; m2=1; m3=1; l1=1; l2=1; l3=1; g=9.81;
dydt(1)=y(2);
dydt(3)=y(4);
dydt(5)=y(6);
dydt(2)*(l1^2*(m1+m2+m3))+dydt(4)((m2+m3)*l1*l2*cos(y(1)-y(3)))+dydt(6)*(m3*l1*l3*cos(y(1)-y(5)))=-(m2+m3)*l1*l2*y(4)^2*sin(y(1)-y(3))-m3*l1*l3*y(6)^2*sin(y(1)-y(5))-(m1+m2+m3)*g*l1*cos(y(1));
dydt(2)*((m2+m3)*l1*l2*cos((y(1)-y(3))))+dydt(4)*(l2^2*(m2+m3))+dydt(6)*(m3*l2*l3*cos(y(3)-y(5)))=(m2+m3)*l1*l2*y(2)^2*sin(y(1)-y(3))-m3*l2*l3*y(6)^2*sin(y(3)-y(5))-(m2+m3)*g*l2*cos(y(3));
dydt(2)*(m3*l1*l3*cos(y(1)-y(5)))+dydt(4)*m3*l2*l3*cos(y(3)-y(5))+dydt(6)(l3^2*m3)=m3*l1*l3*y(2)^2*sin(y(1)-y(5))+m3*l2*l3*y(4)^2*sin(y(3)-y(5))-m3*g**l3*cos(y(5))

I tried this instead, but nothing changed, which was what I expected.