stewart-simscape/docs/identification.html

922 lines
43 KiB
HTML
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
<head>
<!-- 2021-01-08 ven. 15:52 -->
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
<title>Identification of the Stewart Platform using Simscape</title>
<meta name="generator" content="Org mode" />
<meta name="author" content="Dehaeze Thomas" />
<link rel="stylesheet" type="text/css" href="https://research.tdehaeze.xyz/css/style.css"/>
<script type="text/javascript" src="https://research.tdehaeze.xyz/js/script.js"></script>
<script>
MathJax = {
svg: {
scale: 1,
fontCache: "global"
},
tex: {
tags: "ams",
multlineWidth: "%MULTLINEWIDTH",
tagSide: "right",
macros: {bm: ["\\boldsymbol{#1}",1],},
tagIndent: ".8em"
}
};
</script>
<script id="MathJax-script" async
src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-svg.js"></script>
</head>
<body>
<div id="org-div-home-and-up">
<a accesskey="h" href="./index.html"> UP </a>
|
<a accesskey="H" href="./index.html"> HOME </a>
</div><div id="content">
<h1 class="title">Identification of the Stewart Platform using Simscape</h1>
<div id="table-of-contents">
<h2>Table of Contents</h2>
<div id="text-table-of-contents">
<ul>
<li><a href="#orge8b6206">1. Modal Analysis of the Stewart Platform</a>
<ul>
<li><a href="#org40f9c57">1.1. Initialize the Stewart Platform</a></li>
<li><a href="#orgd9529ee">1.2. Identification</a></li>
<li><a href="#orgbdba4a6">1.3. Coordinate transformation</a></li>
<li><a href="#org11e3698">1.4. Analysis</a></li>
<li><a href="#org1db5fc4">1.5. Visualizing the modes</a></li>
</ul>
</li>
<li><a href="#orgfeed9a3">2. Transmissibility Analysis</a>
<ul>
<li><a href="#org5ba3096">2.1. Initialize the Stewart platform</a></li>
<li><a href="#org279dcc8">2.2. Transmissibility</a></li>
</ul>
</li>
<li><a href="#org3ad92e9">3. Compliance Analysis</a>
<ul>
<li><a href="#orgc957431">3.1. Initialize the Stewart platform</a></li>
<li><a href="#org26cb46a">3.2. Compliance</a></li>
</ul>
</li>
<li><a href="#org51e266f">4. Functions</a>
<ul>
<li><a href="#org25ca725">4.1. Compute the Transmissibility</a>
<ul>
<li><a href="#orgafb57d0">Function description</a></li>
<li><a href="#orga00af61">Optional Parameters</a></li>
<li><a href="#org17a8811">Identification of the Transmissibility Matrix</a></li>
<li><a href="#orgbc9a383">Computation of the Frobenius norm</a></li>
</ul>
</li>
<li><a href="#orgb6e05b3">4.2. Compute the Compliance</a>
<ul>
<li><a href="#org210c0ca">Function description</a></li>
<li><a href="#org24feeb1">Optional Parameters</a></li>
<li><a href="#org2c35042">Identification of the Compliance Matrix</a></li>
<li><a href="#orgb002200">Computation of the Frobenius norm</a></li>
</ul>
</li>
</ul>
</li>
</ul>
</div>
</div>
<p>
In this document, we discuss the various methods to identify the behavior of the Stewart platform.
</p>
<ul class="org-ul">
<li><a href="#orgd142bb4">1</a></li>
<li><a href="#org5213401">2</a></li>
<li><a href="#org39baa25">3</a></li>
</ul>
<div id="outline-container-orge8b6206" class="outline-2">
<h2 id="orge8b6206"><span class="section-number-2">1</span> Modal Analysis of the Stewart Platform</h2>
<div class="outline-text-2" id="text-1">
<p>
<a id="orgd142bb4"></a>
</p>
</div>
<div id="outline-container-org40f9c57" class="outline-3">
<h3 id="org40f9c57"><span class="section-number-3">1.1</span> Initialize the Stewart Platform</h3>
<div class="outline-text-3" id="text-1-1">
<div class="org-src-container">
<pre class="src src-matlab">stewart = initializeStewartPlatform();
stewart = initializeFramesPositions(stewart);
stewart = generateGeneralConfiguration(stewart);
stewart = computeJointsPose(stewart);
stewart = initializeStrutDynamics(stewart);
stewart = initializeJointDynamics(stewart, <span class="org-string">'type_F'</span>, <span class="org-string">'universal_p'</span>, <span class="org-string">'type_M'</span>, <span class="org-string">'spherical_p'</span>);
stewart = initializeCylindricalPlatforms(stewart);
stewart = initializeCylindricalStruts(stewart);
stewart = computeJacobian(stewart);
stewart = initializeStewartPose(stewart);
stewart = initializeInertialSensor(stewart);
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">ground = initializeGround(<span class="org-string">'type'</span>, <span class="org-string">'none'</span>);
payload = initializePayload(<span class="org-string">'type'</span>, <span class="org-string">'none'</span>);
controller = initializeController(<span class="org-string">'type'</span>, <span class="org-string">'open-loop'</span>);
</pre>
</div>
</div>
</div>
<div id="outline-container-orgd9529ee" class="outline-3">
<h3 id="orgd9529ee"><span class="section-number-3">1.2</span> Identification</h3>
<div class="outline-text-3" id="text-1-2">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-matlab-cellbreak"><span class="org-comment">%% Options for Linearized</span></span>
options = linearizeOptions;
options.SampleTime = 0;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Name of the Simulink File</span></span>
mdl = <span class="org-string">'stewart_platform_model'</span>;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Input/Output definition</span></span>
clear io; io_i = 1;
io(io_i) = linio([mdl, <span class="org-string">'/Controller'</span>], 1, <span class="org-string">'openinput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Actuator Force Inputs [N]</span>
io(io_i) = linio([mdl, <span class="org-string">'/Relative Motion Sensor'</span>], 1, <span class="org-string">'openoutput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Position/Orientation of {B} w.r.t. {A}</span>
io(io_i) = linio([mdl, <span class="org-string">'/Relative Motion Sensor'</span>], 2, <span class="org-string">'openoutput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Velocity of {B} w.r.t. {A}</span>
<span class="org-matlab-cellbreak"><span class="org-comment">%% Run the linearization</span></span>
G = linearize(mdl, io);
<span class="org-comment">% G.InputName = {'tau1', 'tau2', 'tau3', 'tau4', 'tau5', 'tau6'};</span>
<span class="org-comment">% G.OutputName = {'Xdx', 'Xdy', 'Xdz', 'Xrx', 'Xry', 'Xrz', 'Vdx', 'Vdy', 'Vdz', 'Vrx', 'Vry', 'Vrz'};</span>
</pre>
</div>
<p>
Let&rsquo;s check the size of <code>G</code>:
</p>
<div class="org-src-container">
<pre class="src src-matlab">size(G)
</pre>
</div>
<pre class="example">
size(G)
State-space model with 12 outputs, 6 inputs, and 18 states.
'org_babel_eoe'
ans =
'org_babel_eoe'
</pre>
<p>
We expect to have only 12 states (corresponding to the 6dof of the mobile platform).
</p>
<div class="org-src-container">
<pre class="src src-matlab">Gm = minreal(G);
</pre>
</div>
<pre class="example">
Gm = minreal(G);
6 states removed.
</pre>
<p>
And indeed, we obtain 12 states.
</p>
</div>
</div>
<div id="outline-container-orgbdba4a6" class="outline-3">
<h3 id="orgbdba4a6"><span class="section-number-3">1.3</span> Coordinate transformation</h3>
<div class="outline-text-3" id="text-1-3">
<p>
We can perform the following transformation using the <code>ss2ss</code> command.
</p>
<div class="org-src-container">
<pre class="src src-matlab">Gt = ss2ss(Gm, Gm.C);
</pre>
</div>
<p>
Then, the <code>C</code> matrix of <code>Gt</code> is the unity matrix which means that the states of the state space model are equal to the measurements \(\bm{Y}\).
</p>
<p>
The measurements are the 6 displacement and 6 velocities of mobile platform with respect to \(\{B\}\).
</p>
<p>
We could perform the transformation by hand:
</p>
<div class="org-src-container">
<pre class="src src-matlab">At = Gm.C<span class="org-type">*</span>Gm.A<span class="org-type">*</span>pinv(Gm.C);
Bt = Gm.C<span class="org-type">*</span>Gm.B;
Ct = eye(12);
Dt = zeros(12, 6);
Gt = ss(At, Bt, Ct, Dt);
</pre>
</div>
</div>
</div>
<div id="outline-container-org11e3698" class="outline-3">
<h3 id="org11e3698"><span class="section-number-3">1.4</span> Analysis</h3>
<div class="outline-text-3" id="text-1-4">
<div class="org-src-container">
<pre class="src src-matlab">[V,D] = eig(Gt.A);
</pre>
</div>
<table border="2" cellspacing="0" cellpadding="6" rules="groups" frame="hsides">
<colgroup>
<col class="org-right" />
<col class="org-right" />
<col class="org-right" />
</colgroup>
<thead>
<tr>
<th scope="col" class="org-right">Mode Number</th>
<th scope="col" class="org-right">Resonance Frequency [Hz]</th>
<th scope="col" class="org-right">Damping Ratio [%]</th>
</tr>
</thead>
<tbody>
<tr>
<td class="org-right">1.0</td>
<td class="org-right">780.6</td>
<td class="org-right">0.4</td>
</tr>
<tr>
<td class="org-right">2.0</td>
<td class="org-right">780.6</td>
<td class="org-right">0.3</td>
</tr>
<tr>
<td class="org-right">3.0</td>
<td class="org-right">903.9</td>
<td class="org-right">0.3</td>
</tr>
<tr>
<td class="org-right">4.0</td>
<td class="org-right">1061.4</td>
<td class="org-right">0.3</td>
</tr>
<tr>
<td class="org-right">5.0</td>
<td class="org-right">1061.4</td>
<td class="org-right">0.2</td>
</tr>
<tr>
<td class="org-right">6.0</td>
<td class="org-right">1269.6</td>
<td class="org-right">0.2</td>
</tr>
</tbody>
</table>
</div>
</div>
<div id="outline-container-org1db5fc4" class="outline-3">
<h3 id="org1db5fc4"><span class="section-number-3">1.5</span> Visualizing the modes</h3>
<div class="outline-text-3" id="text-1-5">
<p>
To visualize the i&rsquo;th mode, we may excite the system using the inputs \(U_i\) such that \(B U_i\) is co-linear to \(\xi_i\) (the mode we want to excite).
</p>
<p>
\[ U(t) = e^{\alpha t} ( ) \]
</p>
<p>
Let&rsquo;s first sort the modes and just take the modes corresponding to a eigenvalue with a positive imaginary part.
</p>
<div class="org-src-container">
<pre class="src src-matlab">ws = imag(diag(D));
[ws,I] = sort(ws)
ws = ws(7<span class="org-type">:</span>end); I = I(7<span class="org-type">:</span>end);
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">for</span> <span class="org-variable-name"><span class="org-constant">i</span></span> = <span class="org-constant">1:length(ws)</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">i_mode = I(<span class="org-constant">i</span>); <span class="org-comment">% the argument is the i'th mode</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">lambda_i = D(i_mode, i_mode);
xi_i = V(<span class="org-type">:</span>,i_mode);
a_i = real(lambda_i);
b_i = imag(lambda_i);
</pre>
</div>
<p>
Let do 10 periods of the mode.
</p>
<div class="org-src-container">
<pre class="src src-matlab">t = linspace(0, 10<span class="org-type">/</span>(imag(lambda_i)<span class="org-type">/</span>2<span class="org-type">/</span><span class="org-constant">pi</span>), 1000);
U_i = pinv(Gt.B) <span class="org-type">*</span> real(xi_i <span class="org-type">*</span> lambda_i <span class="org-type">*</span> (cos(b_i <span class="org-type">*</span> t) <span class="org-type">+</span> 1<span class="org-constant">i</span><span class="org-type">*</span>sin(b_i <span class="org-type">*</span> t)));
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">U = timeseries(U_i, t);
</pre>
</div>
<p>
Simulation:
</p>
<div class="org-src-container">
<pre class="src src-matlab">load(<span class="org-string">'mat/conf_simscape.mat'</span>);
<span class="org-matlab-simulink-keyword">set_param</span>(<span class="org-variable-name">conf_simscape</span>, <span class="org-string">'StopTime'</span>, num2str(t(<span class="org-variable-name">end</span>)));
<span class="org-matlab-simulink-keyword">sim</span>(mdl);
</pre>
</div>
<p>
Save the movie of the mode shape.
</p>
<div class="org-src-container">
<pre class="src src-matlab">smwritevideo(mdl, sprintf(<span class="org-string">'figs/mode%i'</span>, <span class="org-constant">i</span>), ...
<span class="org-string">'PlaybackSpeedRatio'</span>, 1<span class="org-type">/</span>(b_i<span class="org-type">/</span>2<span class="org-type">/</span><span class="org-constant">pi</span>), ...
<span class="org-string">'FrameRate'</span>, 30, ...
<span class="org-string">'FrameSize'</span>, [800, 400]);
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">end</span>
</pre>
</div>
<div id="orgd5bd1cd" class="figure">
<p><img src="figs/mode1.gif" alt="mode1.gif" />
</p>
<p><span class="figure-number">Figure 1: </span>Identified mode - 1</p>
</div>
<div id="org5c59f9a" class="figure">
<p><img src="figs/mode3.gif" alt="mode3.gif" />
</p>
<p><span class="figure-number">Figure 2: </span>Identified mode - 3</p>
</div>
<div id="org0f2e8c4" class="figure">
<p><img src="figs/mode5.gif" alt="mode5.gif" />
</p>
<p><span class="figure-number">Figure 3: </span>Identified mode - 5</p>
</div>
</div>
</div>
</div>
<div id="outline-container-orgfeed9a3" class="outline-2">
<h2 id="orgfeed9a3"><span class="section-number-2">2</span> Transmissibility Analysis</h2>
<div class="outline-text-2" id="text-2">
<p>
<a id="org5213401"></a>
</p>
</div>
<div id="outline-container-org5ba3096" class="outline-3">
<h3 id="org5ba3096"><span class="section-number-3">2.1</span> Initialize the Stewart platform</h3>
<div class="outline-text-3" id="text-2-1">
<div class="org-src-container">
<pre class="src src-matlab">stewart = initializeStewartPlatform();
stewart = initializeFramesPositions(stewart, <span class="org-string">'H'</span>, 90e<span class="org-type">-</span>3, <span class="org-string">'MO_B'</span>, 45e<span class="org-type">-</span>3);
stewart = generateGeneralConfiguration(stewart);
stewart = computeJointsPose(stewart);
stewart = initializeStrutDynamics(stewart);
stewart = initializeJointDynamics(stewart, <span class="org-string">'type_F'</span>, <span class="org-string">'universal_p'</span>, <span class="org-string">'type_M'</span>, <span class="org-string">'spherical_p'</span>);
stewart = initializeCylindricalPlatforms(stewart);
stewart = initializeCylindricalStruts(stewart);
stewart = computeJacobian(stewart);
stewart = initializeStewartPose(stewart);
stewart = initializeInertialSensor(stewart, <span class="org-string">'type'</span>, <span class="org-string">'accelerometer'</span>, <span class="org-string">'freq'</span>, 5e3);
</pre>
</div>
<p>
We set the rotation point of the ground to be at the same point at frames \(\{A\}\) and \(\{B\}\).
</p>
<div class="org-src-container">
<pre class="src src-matlab">ground = initializeGround(<span class="org-string">'type'</span>, <span class="org-string">'rigid'</span>, <span class="org-string">'rot_point'</span>, stewart.platform_F.FO_A);
payload = initializePayload(<span class="org-string">'type'</span>, <span class="org-string">'rigid'</span>);
controller = initializeController(<span class="org-string">'type'</span>, <span class="org-string">'open-loop'</span>);
</pre>
</div>
</div>
</div>
<div id="outline-container-org279dcc8" class="outline-3">
<h3 id="org279dcc8"><span class="section-number-3">2.2</span> Transmissibility</h3>
<div class="outline-text-3" id="text-2-2">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-matlab-cellbreak"><span class="org-comment">%% Options for Linearized</span></span>
options = linearizeOptions;
options.SampleTime = 0;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Name of the Simulink File</span></span>
mdl = <span class="org-string">'stewart_platform_model'</span>;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Input/Output definition</span></span>
clear io; io_i = 1;
io(io_i) = linio([mdl, <span class="org-string">'/Disturbances/D_w'</span>], 1, <span class="org-string">'openinput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Base Motion [m, rad]</span>
io(io_i) = linio([mdl, <span class="org-string">'/Absolute Motion Sensor'</span>], 1, <span class="org-string">'openoutput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Absolute Motion [m, rad]</span>
<span class="org-matlab-cellbreak"><span class="org-comment">%% Run the linearization</span></span>
T = linearize(mdl, io, options);
T.InputName = {<span class="org-string">'Wdx'</span>, <span class="org-string">'Wdy'</span>, <span class="org-string">'Wdz'</span>, <span class="org-string">'Wrx'</span>, <span class="org-string">'Wry'</span>, <span class="org-string">'Wrz'</span>};
T.OutputName = {<span class="org-string">'Edx'</span>, <span class="org-string">'Edy'</span>, <span class="org-string">'Edz'</span>, <span class="org-string">'Erx'</span>, <span class="org-string">'Ery'</span>, <span class="org-string">'Erz'</span>};
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">freqs = logspace(1, 4, 1000);
<span class="org-type">figure</span>;
<span class="org-keyword">for</span> <span class="org-variable-name">ix</span> = <span class="org-constant">1:6</span>
<span class="org-keyword">for</span> <span class="org-variable-name">iy</span> = <span class="org-constant">1:6</span>
subplot(6, 6, (ix<span class="org-type">-</span>1)<span class="org-type">*</span>6 <span class="org-type">+</span> iy);
hold on;
plot(freqs, abs(squeeze(freqresp(T(ix, iy), freqs, <span class="org-string">'Hz'</span>))), <span class="org-string">'k-'</span>);
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
ylim([1e<span class="org-type">-</span>5, 10]);
xlim([freqs(1), freqs(end)]);
<span class="org-keyword">if</span> ix <span class="org-type">&lt;</span> 6
xticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">if</span> iy <span class="org-type">&gt;</span> 1
yticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
</pre>
</div>
<p>
From (<a href="#citeproc_bib_item_1">Preumont et al. 2007</a>), one can use the Frobenius norm of the transmissibility matrix to obtain a scalar indicator of the transmissibility performance of the system:
</p>
\begin{align*}
\| \bm{T}(\omega) \| &= \sqrt{\text{Trace}[\bm{T}(\omega) \bm{T}(\omega)^H]}\\
&= \sqrt{\Sigma_{i=1}^6 \Sigma_{j=1}^6 |T_{ij}|^2}
\end{align*}
<div class="org-src-container">
<pre class="src src-matlab">freqs = logspace(1, 4, 1000);
T_norm = zeros(length(freqs), 1);
<span class="org-keyword">for</span> <span class="org-variable-name"><span class="org-constant">i</span></span> = <span class="org-constant">1:length(freqs)</span>
T_norm(<span class="org-constant">i</span>) = sqrt(trace(freqresp(T, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">*</span>freqresp(T, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">'</span>));
<span class="org-keyword">end</span>
</pre>
</div>
<p>
And we normalize by a factor \(\sqrt{6}\) to obtain a performance metric comparable to the transmissibility of a one-axis isolator:
\[ \Gamma(\omega) = \|\bm{T}(\omega)\| / \sqrt{6} \]
</p>
<div class="org-src-container">
<pre class="src src-matlab">Gamma = T_norm<span class="org-type">/</span>sqrt(6);
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-type">figure</span>;
plot(freqs, Gamma)
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
</pre>
</div>
</div>
</div>
</div>
<div id="outline-container-org3ad92e9" class="outline-2">
<h2 id="org3ad92e9"><span class="section-number-2">3</span> Compliance Analysis</h2>
<div class="outline-text-2" id="text-3">
<p>
<a id="org39baa25"></a>
</p>
</div>
<div id="outline-container-orgc957431" class="outline-3">
<h3 id="orgc957431"><span class="section-number-3">3.1</span> Initialize the Stewart platform</h3>
<div class="outline-text-3" id="text-3-1">
<div class="org-src-container">
<pre class="src src-matlab">stewart = initializeStewartPlatform();
stewart = initializeFramesPositions(stewart, <span class="org-string">'H'</span>, 90e<span class="org-type">-</span>3, <span class="org-string">'MO_B'</span>, 45e<span class="org-type">-</span>3);
stewart = generateGeneralConfiguration(stewart);
stewart = computeJointsPose(stewart);
stewart = initializeStrutDynamics(stewart);
stewart = initializeJointDynamics(stewart, <span class="org-string">'type_F'</span>, <span class="org-string">'universal_p'</span>, <span class="org-string">'type_M'</span>, <span class="org-string">'spherical_p'</span>);
stewart = initializeCylindricalPlatforms(stewart);
stewart = initializeCylindricalStruts(stewart);
stewart = computeJacobian(stewart);
stewart = initializeStewartPose(stewart);
stewart = initializeInertialSensor(stewart, <span class="org-string">'type'</span>, <span class="org-string">'accelerometer'</span>, <span class="org-string">'freq'</span>, 5e3);
</pre>
</div>
<p>
We set the rotation point of the ground to be at the same point at frames \(\{A\}\) and \(\{B\}\).
</p>
<div class="org-src-container">
<pre class="src src-matlab">ground = initializeGround(<span class="org-string">'type'</span>, <span class="org-string">'none'</span>);
payload = initializePayload(<span class="org-string">'type'</span>, <span class="org-string">'rigid'</span>);
controller = initializeController(<span class="org-string">'type'</span>, <span class="org-string">'open-loop'</span>);
</pre>
</div>
</div>
</div>
<div id="outline-container-org26cb46a" class="outline-3">
<h3 id="org26cb46a"><span class="section-number-3">3.2</span> Compliance</h3>
<div class="outline-text-3" id="text-3-2">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-matlab-cellbreak"><span class="org-comment">%% Options for Linearized</span></span>
options = linearizeOptions;
options.SampleTime = 0;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Name of the Simulink File</span></span>
mdl = <span class="org-string">'stewart_platform_model'</span>;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Input/Output definition</span></span>
clear io; io_i = 1;
io(io_i) = linio([mdl, <span class="org-string">'/Disturbances/F_ext'</span>], 1, <span class="org-string">'openinput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Base Motion [m, rad]</span>
io(io_i) = linio([mdl, <span class="org-string">'/Absolute Motion Sensor'</span>], 1, <span class="org-string">'openoutput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Absolute Motion [m, rad]</span>
<span class="org-matlab-cellbreak"><span class="org-comment">%% Run the linearization</span></span>
C = linearize(mdl, io, options);
C.InputName = {<span class="org-string">'Fdx'</span>, <span class="org-string">'Fdy'</span>, <span class="org-string">'Fdz'</span>, <span class="org-string">'Mdx'</span>, <span class="org-string">'Mdy'</span>, <span class="org-string">'Mdz'</span>};
C.OutputName = {<span class="org-string">'Edx'</span>, <span class="org-string">'Edy'</span>, <span class="org-string">'Edz'</span>, <span class="org-string">'Erx'</span>, <span class="org-string">'Ery'</span>, <span class="org-string">'Erz'</span>};
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">freqs = logspace(1, 4, 1000);
<span class="org-type">figure</span>;
<span class="org-keyword">for</span> <span class="org-variable-name">ix</span> = <span class="org-constant">1:6</span>
<span class="org-keyword">for</span> <span class="org-variable-name">iy</span> = <span class="org-constant">1:6</span>
subplot(6, 6, (ix<span class="org-type">-</span>1)<span class="org-type">*</span>6 <span class="org-type">+</span> iy);
hold on;
plot(freqs, abs(squeeze(freqresp(C(ix, iy), freqs, <span class="org-string">'Hz'</span>))), <span class="org-string">'k-'</span>);
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
ylim([1e<span class="org-type">-</span>10, 1e<span class="org-type">-</span>3]);
xlim([freqs(1), freqs(end)]);
<span class="org-keyword">if</span> ix <span class="org-type">&lt;</span> 6
xticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">if</span> iy <span class="org-type">&gt;</span> 1
yticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
</pre>
</div>
<p>
We can try to use the Frobenius norm to obtain a scalar value representing the 6-dof compliance of the Stewart platform.
</p>
<div class="org-src-container">
<pre class="src src-matlab">freqs = logspace(1, 4, 1000);
C_norm = zeros(length(freqs), 1);
<span class="org-keyword">for</span> <span class="org-variable-name"><span class="org-constant">i</span></span> = <span class="org-constant">1:length(freqs)</span>
C_norm(<span class="org-constant">i</span>) = sqrt(trace(freqresp(C, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">*</span>freqresp(C, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">'</span>));
<span class="org-keyword">end</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-type">figure</span>;
plot(freqs, C_norm)
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
</pre>
</div>
</div>
</div>
</div>
<div id="outline-container-org51e266f" class="outline-2">
<h2 id="org51e266f"><span class="section-number-2">4</span> Functions</h2>
<div class="outline-text-2" id="text-4">
</div>
<div id="outline-container-org25ca725" class="outline-3">
<h3 id="org25ca725"><span class="section-number-3">4.1</span> Compute the Transmissibility</h3>
<div class="outline-text-3" id="text-4-1">
<p>
<a id="org78f2be2"></a>
</p>
</div>
<div id="outline-container-orgafb57d0" class="outline-4">
<h4 id="orgafb57d0">Function description</h4>
<div class="outline-text-4" id="text-orgafb57d0">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">function</span> <span class="org-variable-name">[T, T_norm, freqs]</span> = <span class="org-function-name">computeTransmissibility</span>(<span class="org-variable-name">args</span>)
<span class="org-comment">% computeTransmissibility -</span>
<span class="org-comment">%</span>
<span class="org-comment">% Syntax: [T, T_norm, freqs] = computeTransmissibility(args)</span>
<span class="org-comment">%</span>
<span class="org-comment">% Inputs:</span>
<span class="org-comment">% - args - Structure with the following fields:</span>
<span class="org-comment">% - plots [true/false] - Should plot the transmissilibty matrix and its Frobenius norm</span>
<span class="org-comment">% - freqs [] - Frequency vector to estimate the Frobenius norm</span>
<span class="org-comment">%</span>
<span class="org-comment">% Outputs:</span>
<span class="org-comment">% - T [6x6 ss] - Transmissibility matrix</span>
<span class="org-comment">% - T_norm [length(freqs)x1] - Frobenius norm of the Transmissibility matrix</span>
<span class="org-comment">% - freqs [length(freqs)x1] - Frequency vector in [Hz]</span>
</pre>
</div>
</div>
</div>
<div id="outline-container-orga00af61" class="outline-4">
<h4 id="orga00af61">Optional Parameters</h4>
<div class="outline-text-4" id="text-orga00af61">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">arguments</span>
<span class="org-variable-name">args</span>.plots logical {mustBeNumericOrLogical} = <span class="org-constant">false</span>
<span class="org-variable-name">args</span>.freqs double {mustBeNumeric, mustBeNonnegative} = logspace(1,4,1000)
<span class="org-keyword">end</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">freqs = args.freqs;
</pre>
</div>
</div>
</div>
<div id="outline-container-org17a8811" class="outline-4">
<h4 id="org17a8811">Identification of the Transmissibility Matrix</h4>
<div class="outline-text-4" id="text-org17a8811">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-matlab-cellbreak"><span class="org-comment">%% Options for Linearized</span></span>
options = linearizeOptions;
options.SampleTime = 0;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Name of the Simulink File</span></span>
mdl = <span class="org-string">'stewart_platform_model'</span>;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Input/Output definition</span></span>
clear io; io_i = 1;
io(io_i) = linio([mdl, <span class="org-string">'/Disturbances/D_w'</span>], 1, <span class="org-string">'openinput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Base Motion [m, rad]</span>
io(io_i) = linio([mdl, <span class="org-string">'/Absolute Motion Sensor'</span>], 1, <span class="org-string">'output'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Absolute Motion [m, rad]</span>
<span class="org-matlab-cellbreak"><span class="org-comment">%% Run the linearization</span></span>
T = linearize(mdl, io, options);
T.InputName = {<span class="org-string">'Wdx'</span>, <span class="org-string">'Wdy'</span>, <span class="org-string">'Wdz'</span>, <span class="org-string">'Wrx'</span>, <span class="org-string">'Wry'</span>, <span class="org-string">'Wrz'</span>};
T.OutputName = {<span class="org-string">'Edx'</span>, <span class="org-string">'Edy'</span>, <span class="org-string">'Edz'</span>, <span class="org-string">'Erx'</span>, <span class="org-string">'Ery'</span>, <span class="org-string">'Erz'</span>};
</pre>
</div>
<p>
If wanted, the 6x6 transmissibility matrix is plotted.
</p>
<div class="org-src-container">
<pre class="src src-matlab">p_handle = zeros(6<span class="org-type">*</span>6,1);
<span class="org-keyword">if</span> args.plots
fig = <span class="org-type">figure</span>;
<span class="org-keyword">for</span> <span class="org-variable-name">ix</span> = <span class="org-constant">1:6</span>
<span class="org-keyword">for</span> <span class="org-variable-name">iy</span> = <span class="org-constant">1:6</span>
p_handle((ix<span class="org-type">-</span>1)<span class="org-type">*</span>6 <span class="org-type">+</span> iy) = subplot(6, 6, (ix<span class="org-type">-</span>1)<span class="org-type">*</span>6 <span class="org-type">+</span> iy);
hold on;
plot(freqs, abs(squeeze(freqresp(T(ix, iy), freqs, <span class="org-string">'Hz'</span>))), <span class="org-string">'k-'</span>);
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
<span class="org-keyword">if</span> ix <span class="org-type">&lt;</span> 6
xticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">if</span> iy <span class="org-type">&gt;</span> 1
yticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
linkaxes(p_handle, <span class="org-string">'xy'</span>)
xlim([freqs(1), freqs(end)]);
ylim([1e<span class="org-type">-</span>5, 1e2]);
han = <span class="org-type">axes</span>(fig, <span class="org-string">'visible'</span>, <span class="org-string">'off'</span>);
han.XLabel.Visible = <span class="org-string">'on'</span>;
han.YLabel.Visible = <span class="org-string">'on'</span>;
xlabel(han, <span class="org-string">'Frequency [Hz]'</span>);
ylabel(han, <span class="org-string">'Transmissibility [m/m]'</span>);
<span class="org-keyword">end</span>
</pre>
</div>
</div>
</div>
<div id="outline-container-orgbc9a383" class="outline-4">
<h4 id="orgbc9a383">Computation of the Frobenius norm</h4>
<div class="outline-text-4" id="text-orgbc9a383">
<div class="org-src-container">
<pre class="src src-matlab">T_norm = zeros(length(freqs), 1);
<span class="org-keyword">for</span> <span class="org-variable-name"><span class="org-constant">i</span></span> = <span class="org-constant">1:length(freqs)</span>
T_norm(<span class="org-constant">i</span>) = sqrt(trace(freqresp(T, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">*</span>freqresp(T, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">'</span>));
<span class="org-keyword">end</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">T_norm = T_norm<span class="org-type">/</span>sqrt(6);
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">if</span> args.plots
<span class="org-type">figure</span>;
plot(freqs, T_norm)
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
xlabel(<span class="org-string">'Frequency [Hz]'</span>);
ylabel(<span class="org-string">'Transmissibility - Frobenius Norm'</span>);
<span class="org-keyword">end</span>
</pre>
</div>
</div>
</div>
</div>
<div id="outline-container-orgb6e05b3" class="outline-3">
<h3 id="orgb6e05b3"><span class="section-number-3">4.2</span> Compute the Compliance</h3>
<div class="outline-text-3" id="text-4-2">
<p>
<a id="org13d7e8a"></a>
</p>
</div>
<div id="outline-container-org210c0ca" class="outline-4">
<h4 id="org210c0ca">Function description</h4>
<div class="outline-text-4" id="text-org210c0ca">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">function</span> <span class="org-variable-name">[C, C_norm, freqs]</span> = <span class="org-function-name">computeCompliance</span>(<span class="org-variable-name">args</span>)
<span class="org-comment">% computeCompliance -</span>
<span class="org-comment">%</span>
<span class="org-comment">% Syntax: [C, C_norm, freqs] = computeCompliance(args)</span>
<span class="org-comment">%</span>
<span class="org-comment">% Inputs:</span>
<span class="org-comment">% - args - Structure with the following fields:</span>
<span class="org-comment">% - plots [true/false] - Should plot the transmissilibty matrix and its Frobenius norm</span>
<span class="org-comment">% - freqs [] - Frequency vector to estimate the Frobenius norm</span>
<span class="org-comment">%</span>
<span class="org-comment">% Outputs:</span>
<span class="org-comment">% - C [6x6 ss] - Compliance matrix</span>
<span class="org-comment">% - C_norm [length(freqs)x1] - Frobenius norm of the Compliance matrix</span>
<span class="org-comment">% - freqs [length(freqs)x1] - Frequency vector in [Hz]</span>
</pre>
</div>
</div>
</div>
<div id="outline-container-org24feeb1" class="outline-4">
<h4 id="org24feeb1">Optional Parameters</h4>
<div class="outline-text-4" id="text-org24feeb1">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">arguments</span>
<span class="org-variable-name">args</span>.plots logical {mustBeNumericOrLogical} = <span class="org-constant">false</span>
<span class="org-variable-name">args</span>.freqs double {mustBeNumeric, mustBeNonnegative} = logspace(1,4,1000)
<span class="org-keyword">end</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab">freqs = args.freqs;
</pre>
</div>
</div>
</div>
<div id="outline-container-org2c35042" class="outline-4">
<h4 id="org2c35042">Identification of the Compliance Matrix</h4>
<div class="outline-text-4" id="text-org2c35042">
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-matlab-cellbreak"><span class="org-comment">%% Options for Linearized</span></span>
options = linearizeOptions;
options.SampleTime = 0;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Name of the Simulink File</span></span>
mdl = <span class="org-string">'stewart_platform_model'</span>;
<span class="org-matlab-cellbreak"><span class="org-comment">%% Input/Output definition</span></span>
clear io; io_i = 1;
io(io_i) = linio([mdl, <span class="org-string">'/Disturbances/F_ext'</span>], 1, <span class="org-string">'openinput'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% External forces [N, N*m]</span>
io(io_i) = linio([mdl, <span class="org-string">'/Absolute Motion Sensor'</span>], 1, <span class="org-string">'output'</span>); io_i = io_i <span class="org-type">+</span> 1; <span class="org-comment">% Absolute Motion [m, rad]</span>
<span class="org-matlab-cellbreak"><span class="org-comment">%% Run the linearization</span></span>
C = linearize(mdl, io, options);
C.InputName = {<span class="org-string">'Fdx'</span>, <span class="org-string">'Fdy'</span>, <span class="org-string">'Fdz'</span>, <span class="org-string">'Mdx'</span>, <span class="org-string">'Mdy'</span>, <span class="org-string">'Mdz'</span>};
C.OutputName = {<span class="org-string">'Edx'</span>, <span class="org-string">'Edy'</span>, <span class="org-string">'Edz'</span>, <span class="org-string">'Erx'</span>, <span class="org-string">'Ery'</span>, <span class="org-string">'Erz'</span>};
</pre>
</div>
<p>
If wanted, the 6x6 transmissibility matrix is plotted.
</p>
<div class="org-src-container">
<pre class="src src-matlab">p_handle = zeros(6<span class="org-type">*</span>6,1);
<span class="org-keyword">if</span> args.plots
fig = <span class="org-type">figure</span>;
<span class="org-keyword">for</span> <span class="org-variable-name">ix</span> = <span class="org-constant">1:6</span>
<span class="org-keyword">for</span> <span class="org-variable-name">iy</span> = <span class="org-constant">1:6</span>
p_handle((ix<span class="org-type">-</span>1)<span class="org-type">*</span>6 <span class="org-type">+</span> iy) = subplot(6, 6, (ix<span class="org-type">-</span>1)<span class="org-type">*</span>6 <span class="org-type">+</span> iy);
hold on;
plot(freqs, abs(squeeze(freqresp(C(ix, iy), freqs, <span class="org-string">'Hz'</span>))), <span class="org-string">'k-'</span>);
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
<span class="org-keyword">if</span> ix <span class="org-type">&lt;</span> 6
xticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">if</span> iy <span class="org-type">&gt;</span> 1
yticklabels({});
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
<span class="org-keyword">end</span>
linkaxes(p_handle, <span class="org-string">'xy'</span>)
xlim([freqs(1), freqs(end)]);
han = <span class="org-type">axes</span>(fig, <span class="org-string">'visible'</span>, <span class="org-string">'off'</span>);
han.XLabel.Visible = <span class="org-string">'on'</span>;
han.YLabel.Visible = <span class="org-string">'on'</span>;
xlabel(han, <span class="org-string">'Frequency [Hz]'</span>);
ylabel(han, <span class="org-string">'Compliance [m/N, rad/(N*m)]'</span>);
<span class="org-keyword">end</span>
</pre>
</div>
</div>
</div>
<div id="outline-container-orgb002200" class="outline-4">
<h4 id="orgb002200">Computation of the Frobenius norm</h4>
<div class="outline-text-4" id="text-orgb002200">
<div class="org-src-container">
<pre class="src src-matlab">freqs = args.freqs;
C_norm = zeros(length(freqs), 1);
<span class="org-keyword">for</span> <span class="org-variable-name"><span class="org-constant">i</span></span> = <span class="org-constant">1:length(freqs)</span>
C_norm(<span class="org-constant">i</span>) = sqrt(trace(freqresp(C, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">*</span>freqresp(C, freqs(<span class="org-constant">i</span>), <span class="org-string">'Hz'</span>)<span class="org-type">'</span>));
<span class="org-keyword">end</span>
</pre>
</div>
<div class="org-src-container">
<pre class="src src-matlab"><span class="org-keyword">if</span> args.plots
<span class="org-type">figure</span>;
plot(freqs, C_norm)
<span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'XScale'</span>, <span class="org-string">'log'</span>); <span class="org-type">set</span>(<span class="org-variable-name">gca</span>, <span class="org-string">'YScale'</span>, <span class="org-string">'log'</span>);
xlabel(<span class="org-string">'Frequency [Hz]'</span>);
ylabel(<span class="org-string">'Compliance - Frobenius Norm'</span>);
<span class="org-keyword">end</span>
</pre>
</div>
<style>.csl-entry{text-indent: -1.5em; margin-left: 1.5em;}</style><h2 class='citeproc-org-bib-h2'>Bibliography</h2>
<div class="csl-bib-body">
<div class="csl-entry"><a name="citeproc_bib_item_1"></a>Preumont, A., M. Horodinca, I. Romanescu, B. de Marneffe, M. Avraam, A. Deraemaeker, F. Bossens, and A. Abu Hanieh. 2007. “A Six-Axis Single-Stage Active Vibration Isolator Based on Stewart Platform.” <i>Journal of Sound and Vibration</i> 300 (3-5):64461. <a href="https://doi.org/10.1016/j.jsv.2006.07.050">https://doi.org/10.1016/j.jsv.2006.07.050</a>.</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div id="postamble" class="status">
<p class="author">Author: Dehaeze Thomas</p>
<p class="date">Created: 2021-01-08 ven. 15:52</p>
</div>
</body>
</html>