Why is nothing showing up on my plots?

All three plots won't show the iterations from my calculation.

% Simulate the system

for k = 1:(tf - t0)/dt

% Update the reference joint positions

q_r = q + dt*qdr;

% Calculate the Jacobian

J = calc_jacobian(q, L1, L2, L3);

% Update the actual joint positions

q = q_r + dt*J'*qdr;

% Plot the results

if k == 1

plot_results(q, J);

end

end

function plot_results(q, J)

% Plot the joint positions

subplot(3, 1, 1)

hold on

plot(q(1), 'b')

plot(q(2), 'r')

plot(q(3), 'g')

xlabel('Time (s)')

ylabel('Joint angle (deg)')

% Plot the Jacobian determinant

subplot(3, 1, 2)

plot(J'*J)

xlabel('Time (s)')

ylabel('Jacobian determinant')

% Plot the end-effector pose

subplot(3, 1, 3)

plot(q(1), q(2), 'b*')

xlabel('x (m)')

ylabel('y (m)')

axis([-2, 4, -2, 4])

end

[Image Analyst]

You'll need to give us the missing values of t0, dt, tf, q, qdr, etc.

And you're only calling your plot function on the first iteration, not subsequent iterations so of course it doesn't show any other iterations than the first. Why do you think it should?

[Jade Cruz]

Thats what I thought so I originally had it in the for-loop without the if-statement. I was just playing around and thought maybe adding it will change something. I originally had something like this.

for k = 1:(tf - t0)/dt

% Update the reference joint positions

q_r = q + dt*qdr;

% Calculate the Jacobian

J = calc_jacobian(q, L1, L2, L3);

% Update the actual joint positions

q = q_r + dt*J'*qdr;

% Plot the results

plot_results(q, J);

end

Regarding the missing values:

% Define the parameters

L1 = 4;

L2 = 3;

L3 = 2;

dt = 0.1;

t0 = 0;

tf = 5;

q0 = [10, 20, 30]';

qdr = [0.2, -0.3, -0.2]';

% Initialize the variables

q = q0;

J = zeros(3, 3);

I have another function that I used to calculate for J. I will provide if need be,

[Sandeep Mishra]

@Jade Cruz

What output are you getting? Are you getting any error or different plots?

Can you comment calc_jacobian function for better reference?

[Star Strider]

It doesn’t plot anything for me, either (for obvious reasons) —

% Regarding the missing values:
% Define the parameters
L1 = 4;
L2 = 3;
L3 = 2;
dt = 0.1;
t0 = 0;
tf = 5;
q0 = [10, 20, 30]';
qdr = [0.2, -0.3, -0.2]';
% Initialize the variables
q = q0;
J = zeros(3, 3);

for k = 1:(tf - t0)/dt
    % Update the reference joint positions
    q_r = q + dt*qdr;
    % Calculate the Jacobian
    J = calc_jacobian(q, L1, L2, L3);
    % Update the actual joint positions
    q = q_r + dt*J'*qdr;
    % Plot the results
    plot_results(q, J);
end

Unrecognized function or variable 'calc_jacobian'.

.

[Jade Cruz]

function J = calc_jacobian(q,L1,L2,L3)

% Calculate the Jacobian

J = [-L1*sin(q(1)) - L2*sin(q(1) + q(2)) - L3*sin(q(1) + q(2) + q(3));

L1*cos(q(1)) + L2*cos(q(1) + q(2)) + L3*cos(q(1) + q(2) + q(3));

1];

end

Here's the missing function. The plots would appear but nothing would show on it. Just blank plots.

[Sandeep Mishra]

@Jade Cruz

Just to clarify, since (tf - t0)/dt = 50, so do you want 50 plot figures? which contains 3 subplots?

[Jade Cruz]

@Sandeep Mishra Would that be possible? That is what I was trying for

[Sandeep Mishra]

I have answered that for example case with 3 different figures.

If you still face any error, do comment

[Jade Cruz]

@Sandeep Mishra Thank you for your help! I have posted a comment below your answer. I appreciate the continuous support.

Answer 1

[Steven Lord]

When you call plot with just a color in the line specification, what you get is a plot with no markers.

h = plot(1, 'b');

h.Marker

ans = 'none'

What does a line "connecting" one point look like? Nothing. If instead you'd specified a marker in addition to a color, you'd get a line "connecting" one point and a marker on that point.

figure
h = plot(1, 'bx');

If you want to create a plot where you add points to a line iteratively, I'd use the animatedline function to create the line then call addpoints on the animated line. You won't be able to see the iterative process in Answers, but if you run the code below in an interactive session of MATLAB you'll see the plot appear, point by point.

x = 0:360;
y = sind(x);
figure
axis([0 360 -1 1])
h = animatedline('LineStyle', '-', 'Color', 'b');
for k = 1:numel(x)
    addpoints(h, x(k), y(k))
    pause(1/32)
end

[Jade Cruz]

Thank you so much. This was what I was thinking. I will try this out!

Answer 2

[Sandeep Mishra]

Hello Jade,

I understood from our discussion in the comments that you are trying to plot multiple plots (in your case = 50)

In MATLAB, you can use figure(n) to create a figure with number property is equal to n.

You can refer the below example code (plot with 3 figures)

clear;

% Define the parameters
L1 = 4;
L2 = 3;
L3 = 2;
dt = 0.1;
t0 = 0;
tf = 5;
q0 = [10, 20, 30]';
qdr = [0.2, -0.3, -0.2]';
% Initialize the variables
q = q0;
J = zeros(3, 3);

% for k = 1:(tf - t0)/dt

% Example 
for k = 1:3
    % Update the reference joint positions
    q_r = q + dt*qdr;
    % Calculate the Jacobian
    J = calc_jacobian(q, L1, L2, L3)
    % Update the actual joint positions
    q = q_r + dt*J'*qdr
    % Plot the results
    figure(k)
    plot_results(q, J);
end

J = 3×1

5.7498 -4.7984 1.0000

q = 3×1

10.2589 20.2089 30.2189

J = 3×1

7.0790 -2.0237 1.0000

q = 3×1

10.4612 20.3612 30.3812

J = 3×1

7.1168 0.3353 1.0000

q = 3×1

10.5935 20.4435 30.4735

function J = calc_jacobian(q,L1,L2,L3)
% Calculate the Jacobian
J = [-L1*sin(q(1)) - L2*sin(q(1) + q(2)) - L3*sin(q(1) + q(2) + q(3)); 
    L1*cos(q(1)) + L2*cos(q(1) + q(2)) + L3*cos(q(1) + q(2) + q(3));    
    1];
end

function plot_results(q, J)
% Plot the joint positions
subplot(3, 1, 1)
hold on
plot(q(1), 'b')
plot(q(2), 'r')
plot(q(3), 'g')
xlabel('Time (s)')
ylabel('Joint angle (deg)')
% Plot the Jacobian determinant
subplot(3, 1, 2)
plot(J'*J)
xlabel('Time (s)')
ylabel('Jacobian determinant')
% Plot the end-effector pose
subplot(3, 1, 3)
plot(q(1), q(2), 'b*')
xlabel('x (m)')
ylabel('y (m)')
axis([-2, 4, -2, 4])
end

You can refer to the below documentation to learn more about "figure" in MATLAB.

https://in.mathworks.com/help/matlab/ref/figure.html#d124e460113

[Jade Cruz]

That is perfect but I do notice that all the figures do not show any sort of information or plot. Would the problem be my equations and not my code?

[Sandeep Mishra]

I hope @Steven Lord cleared your doubt @Jade Cruz ?

Answer 3

[Star Strider]

It now fills the matrices corrrectly and plots something, however I cannot determine what you want it to plot —

% Regarding the missing values:
% Define the parameters
L1 = 4;
L2 = 3;
L3 = 2;
dt = 0.1;
t0 = 0;
tf = 5;
q0 = [10, 20, 30]';
qdr = [0.2, -0.3, -0.2]';
% Initialize the variables
q(:,1) = q0;
J = zeros(3, 3);

kv = 1:(tf - t0)/dt;                                        % Create Vector

for k = 1:numel(kv)
    % Update the reference joint positions
    q_r(:,k+1) = q(:,k) + dt*qdr;                           % Subscript Result (Create Matrix Of Column Vectors)
    % Calculate the Jacobian
    J(:,k+1) = calc_jacobian(q(:,k), L1, L2, L3);           % Subscript Result (Create Matrix Of Column Vectors)
    % Update the actual joint positions
    q(:,k+1) = q_r(:,k) + dt*J(:,k)'*qdr;                   % Subscript Result (Create Matrix Of Column Vectors)
end
    % Plot the results
figure
plot(q, J)
grid
xlabel('q')
ylabel('J')
title('Original Matrices')

figure
plot(q.', J.')
grid
xlabel('q')
ylabel('J')
title('Transposed Matrices')

function J = calc_jacobian(q,L1,L2,L3)
% Calculate the Jacobian
J = [-L1*sin(q(1)) - L2*sin(q(1) + q(2)) - L3*sin(q(1) + q(2) + q(3)); 
    L1*cos(q(1)) + L2*cos(q(1) + q(2)) + L3*cos(q(1) + q(2) + q(3));    
    1];
end

I defer to you to figure out how to plot the results.

.