メインコンテンツ
結果:
Simulink has been an essential tool for modeling and simulating dynamic systems in MATLAB. With the continuous advancements in AI, automation, and real-time simulation, I’m curious about what the future holds for Simulink.
What improvements or new features do you think Simulink will have in the coming years? Will AI-driven modeling, cloud-based simulation, or improved hardware integration shape the next generation of Simulink?
The pdist function allows the user to specify the function coding the similarity between rows of matrices (called DISTFUN in the documentation).
With the increasing diffusion of large datasets, techniques to sparsify graphs are increasingly being explored and used in practical applications. It is easy to code one own's DISTFUN such that it returns a sparse vector. Unfortunately, pdist (and pdist2) will return a dense vector in the output, which for very large graphs will cause an out of memory error. The offending code is
(lines 434 etc.)
% Make the return have whichever numeric type the distance function
% returns, or logical.
if islogical(Y)
Y = false(1,n*(n-1)./2);
else % isnumeric
Y = zeros(1,n*(n-1)./2, class(Y));
end
To have pdist return a sparse vector, the only modification that is required is
if islogical(Y)
Y = false(1,n*(n-1)./2);
elseif issparse(Y)
Y = sparse(1,n*(n-1)./2);
else % isnumeric
Y = zeros(1,n*(n-1)./2, class(Y));
end
It is a bit more work to modify squareform to produce a sparse matrix, given a sparse vector produced by the modified pdist. Squareform includes several checks on the inputs, but the core functionality for sparse vectors would be given by something like
% given a sparse vector d returned by pdist, compute a sparse squareform
[~,j,v] = find(d);
[m,n] = pdist_ind2sub(j, nobs);
W = sparse(m, n, v, nobs, nobs);
W = W + W';
Here, pdist_ind2sub is a function that given a set of indices into a vector produced by pdist, returns the corresponding subscripts in a (triangular) matrix. Computing this requires information about the number of observations given to pdist, i.e. what was n in the preceding code. I could not figure out a way to use the function adjacency to accomplish this.
MATLAB FEX(MATLAB File Exchange) should support Markdown syntax for writing. In recent years, many open-source community documentation platforms, such as GitHub, have generally supported Markdown. MATLAB is also gradually improving its support for Markdown syntax. However, when directly uploading files to the MATLAB FEX community and preparing to write an overview, the outdated document format buttons are still present. Even when directly uploading a Markdown document, it cannot be rendered. We hope the community can support Markdown syntax!
BTW，I know that open-source Markdown writing on GitHub and linking to MATLAB FEX is feasible, but this is a workaround. It would be even better if direct native support were available.
I have already created a 8-DOF electric vehicle model in MATLAB and during the verification method, I noticed that the roll angle does not really align with the data from CarMaker. I would like to seek for the oppinion from this community on how to improve the graph that I have obtained. I have also attatched the parameters of this vehicle as well as the graph that I have obtained. The blue plots indicates the data from a real world based electric model from CarMaker and yellow is from the 8-DOF model in matlab.
these are the vehicle parameters
data.g = 9.81; % [m/s^2] acceleration of gravity
data.f_res = 1.1e-2; % [-] rolling friction parameter
data.k_res = 6.5e-7; % [s^2/m^2] rolling friction coefficient
data.rho = 1.205; % [kg/m^3] air density
data.area = 2.156; % [m^2] cross section
data.cx = 0.30; % [-] drag coefficient
data.mass = 2200.10; % [kg] total vehicle mass
data.Jx = 552.75; % [kg*m^2] roll-axis inertia
data.Jz = 3002.5; % [kg*m^2] yaw-axis inertia
data.Jw = 2; % [kg*m^2] spin-axis inertia of wheel
data.radius = 0.3401; % [m] wheel radius
% Reduced Pacejka tyre model data
data.tyre_par(1) = 82.8868; % Pacejka coeff.
data.tyre_par(2) = 1.2070; % Pacejka coeff.
data.tyre_par(3) = 1.1351; % Pacejka coeff.
data.tyre_par(4) = 14.4035; % Pacejka coeff.
data.tyre_par(5) = 1.1932; % Pacejka coeff.
data.tyre_par(6) = -0.0001; % Pacejka coeff.
data.tyre_par(7) = 2.1219; % Pacejka coeff.
% Pacejka 5.2 tyre model data
tyre = ImportTyreData('.', 'Tyre_VSM.tir');
tyre = rmfield(tyre, 'file');
data.tyre_par_full = tyre;
data.wbase_f = 1.4727; % [m] front wheelbase
data.wbase_r = 1.4553; % [m] rear wheelbase
data.wbase = data.wbase_f + data.wbase_r; % [m] wheelbase
data.track = 1.655; % [m] track
data.h_cg = 0.631; % [m] centre of gravity height from ground
data.h_roll = 0.091; % [m] roll centre height from ground
data.k_roll_f = 8.67e4; % [Nm] roll stiffness at front
data.k_roll_r = 7.80e4; % [Nm] roll stiffness at rear
data.c_roll_f = 1.2e6; % [Nm/s] roll damping at front
data.c_roll_r = 6e5; % [Nm/s] roll damping at rear
data.k_act_roll = 0.9; % [-] active anti-roll coefficient
data.em_curve = [ % electric motor torque-speed curve
0, 900, 1000, 1100, 1200, 1300 % [RPM]
1500, 1500, 1400, 1000, 500, 0 % [Nm]
];
data.torque_bk_lb = -5e3; % [Nm] minimum brake torque
You've probably heard about the DeepSeek AI models by now. Did you know you can run them on your own machine (assuming its powerful enough) and interact with them on MATLAB?
In my latest blog post, I install and run one of the smaller models and start playing with it using MATLAB.
Larger models wouldn't be any different to use assuming you have a big enough machine...and for the largest models you'll need a HUGE machine!
Even tiny models, like the 1.5 billion parameter one I demonstrate in the blog post, can be used to demonstrate and teach things about LLM-based technologies.
Have a play. Let me know what you think.
My intention is to generate th code as mentioned below
void computeArea() {
rectangle.area = rectangle.length * rectangle.width;
}
ı m tryna prepare a battery simulation with simulink but when ı push right click on "battery-table based" , and then ı go simscape button, and ı just only see "log simulation data" . Have you any reccomend for this problem? It s probably an easy solution, but ı can't
This is the question I am trying to solve and I also uploading the code i have written and error I am getting, Please help me the error please.
Currently, according to the official documentation, "DisplayName" only supports character vectors or single scalar string as input. For example, when plotting three variables simultaneously, if I use a single scalar string as input, the legend labels will all be the same. To have different labels, I need to specify them separately using the legend function with label1, label2, label3.
Here's an example illustrating the issue:
x = (1:10)';
y1 = x;
y2 = x.^2;
y3 = x.^3;
% Plotting with a string scalar for DisplayName
figure;
plot(x, [y1,y2,y3], DisplayName="y = x");
legend;
% To have different labels, I need to use the legend function separately
figure;
plot(x, [y1,y2,y3], DisplayName=["y = x","y = x^2","y=x^3"]);
% legend("y = x","y = x^2","y=x^3");
Hi!
I’m facing a problem in which I need an “intermediate” level of detail for my simulation.
In my case, I’d like to simulate a radar altimeter flying at low altitude over some terrain or sea. Over this surface, suppose to have an obstacle (tree/house/ship/…). I know everything about my altimeter (pulsed radar, frenquency, pulse duration, beam width, …). A possible outcome of such a simulation could be the assessment of the impact of different gain patterns on the received pulse.
What I have always found on the internet are either too simple solution (like solving the radar equations) or too complex (Method of Moments, or similar approaches).
Regarding the radar equation, I have always wandered how it can deal with the echoes coming from the outer regions of the beamwidth of the altimeter antenna (the equation only has the boresight gain as input parameter).
On the other hand, in my opinion, approaches like MoM are really too complicated and beyond my scope.
I had a look and tried to implement some of the Matlab functions that already exist (e.g., the ones on the FMCW Radar Altimeter Simulation example), but I don’t think they meet my needs.
So I decided to try to write my own code, providing the shape of the terrain/sea surface, the shape for the obstacles (for now, just simple shapes)… I guess I’d have to sample the domain, evaluating the echoes for all these elements… however, even in this case there are a lot of parameters that I don’t know how to handle properly, for example: - is it reasonable to discretize terrain or sea instead of assuming some model for the backscatter? - how should the domain be discretized? - how can I guarantee the conservation of power, considering the effects of the radiation pattern of the antenna and the aforementioned discretization of the domain?
Thanks in advance for your support.
Best regards, Alessandro
In one of my MATLAB projects, I want to add a button to an existing axes toolbar. The function for doing this is axtoolbarbtn:
axtoolbarbtn(tb,style,Name=Value)
However, I have found that the existing interfaces and behavior make it quite awkward to accomplish this task.
Here are my observations.
Adding a Button to the Default Axes Toolbar Is Unsupported
plot(1:10)
ax = gca;
tb = ax.Toolbar
Calling axtoolbarbtn on ax results in an error:
>> axtoolbarbtn(tb,"state")
Error using axtoolbarbtn (line 77)
Modifying the default axes toolbar is not supported.
Default Axes Toolbar Can't Be Distinguished from an Empty Toolbar
The Children property of the default axes toolbar is empty. Thus, it appears programmatically to have no buttons, just like an empty toolbar created by axtoolbar.
cla
plot(1:10)
ax = gca;
tb = ax.Toolbar;
tb.Children
ans = 0x0 empty GraphicsPlaceholder array.
tb2 = axtoolbar(ax);
tb2.Children
ans = 0x0 empty GraphicsPlaceholder array.
A Workaround
An empty axes toolbar seems to have no use except to initalize a toolbar before immediately adding buttons to it. Therefore, it seems reasonable to assume that an axes toolbar that appears to be empty is really the default toolbar. While we can't add buttons to the default axes toolbar, we can create a new toolbar that has all the same buttons as the default one, using axtoolbar("default"). And then we can add buttons to the new toolbar.
That observation leads to this workaround:
tb = ax.Toolbar;
if isempty(tb.Children)
% Assume tb is the default axes toolbar. Recreate
% it with the default buttons so that we can add a new
% button.
tb = axtoolbar(ax,"default");
end
btn = axtoolbarbtn(tb);
% Then set up the button as desired (icon, callback,
% etc.) by setting its properties.
As workarounds go, it's not horrible. It just seems a shame to have to delete and then recreate a toolbar just to be able to add a button to it.
The worst part about the workaround is that it is so not obvious. It took me a long time of experimentation to figure it out, including briefly giving it up as seemingly impossible.
The documentation for axtoolbarbtn avoids the issue. The most obvious example to write for axtoolbarbtn would be the first thing every user of it will try: add a toolbar button to the toolbar that gets created automatically in every call to plot. The doc page doesn't include that example, of course, because it wouldn't work.
My Request
I like the axes toolbar concept and the axes interactivity that it promotes, and I think the programming interface design is mostly effective. My request to MathWorks is to modify this interface to smooth out the behavior discontinuity of the default axes toolbar, with an eye towards satisfying (and documenting) the general use case that I've described here.
One possible function design solution is to make the default axes toolbar look and behave like the toolbar created by axtoolbar("default"), so that it has Children and so it is modifiable.
I am curious as to how my goal can be accomplished in Matlab.
The present APP called "Matching Network Designer" works quite well, but it is limited to a single section of a "PI", a "TEE", or an "L" topology circuit.
This limits the bandwidth capability of the APP when the intended use is to create an amplifier design intended for wider bandwidth projects.
I am requesting that a "Broadband Matching Network Designer" APP be developed by you, the MathWorks support team.
One suggestion from me is to be able to cascade a second section (or "pole") to the first.
Then the resulting topology would be capable of achieving that wider bandwidth of the microwave amplifier project where it would be later used with the transistor output and input matching networks.
Instead of limiting the APP to a single frequency, the entire s parameter file would be used as an input.
The APP would convert the polar s parameters to rectangular scaler complex impedances that you already use.
At that point, having started out with the first initial center frequency, the other frequencies both greater than and less than the center would come into use by an optimization of the circuit elements.
I'm hoping that you will be able to take on this project.
I can include an attachment of such a Matching Network Designer APP that you presently have if you like.
That network is centered at 10 GHz.
Kimberly Renee Alvarez.
310-367-5768
私の場合、前の会社が音楽認識アプリの会社で、アルゴリズム開発でFFTが使われていたことがきっかけでした。でも、MATLABのすごさが分かったのは、機械学習のオンライン講座で、Andrew Ngが、線型代数を使うと、数式と非常に近い構文のコードで問題が処理できることを学んだ時でした。
Three former MathWorks employees, Steve Wilcockson, David Bergstein, and Gareth Thomas, joined the ArrayCast pod cast to discuss their work on array based languages. At the end of the episode, Steve says,
> It's a little known fact about MATLAB. There's this thing, Gareth has talked about the community. One of the things MATLAB did very, very early was built the MATLAB community, the so-called MATLAB File Exchange, which came about in the early 2000s. And it was where people would share code sets, M files, et cetera. This was long before GitHub came around. This was well ahead of its time. And I think there are other places too, where MATLAB has delivered cultural benefits over and above the kind of core programming and mathematical capabilities too. So, you know, MATLAB Central, File Exchange, very much saw the future.
Listen here: The ArrayCast, Episode 79, May 10, 2024.
Dears,
I am running a MS-DSGE model using RISE toolbox. I want to add a fiscal shock and examine its effect on output, price...
%fiscal shock
shock_type = {'eps_G'};
%here is my variable list of a cell array of character variables and not a struct.
var_list={'log_y','C','pi_ann','B_nominal','B','sp','i_ann','r_real_ann','P'};
% EXOGENOUS SWITCHING
myirfs1=irf(m1,'irf_periods',24,'irf_shock_sign',1);
myirfs1 = struct()
myirfs1.eps_CP = struct();
myirfs1.eps_G = struct();
myirfs1.eps_T = struct();
myirfs1.eps_a = struct();
myirfs1.eps_nu = struct();
myirfs1.eps_z = struct();
var_aux = {'log_y','C','pi_ann','B_nominal','B','sp','i_ann','r_real_ann','P'};
var_aux3 = {'eps_G_log_y','eps_G_C','eps_G_pi_ann','eps_G_B_nominal','eps_G_B','eps_G_sp','eps_G_i_ann','eps_G_r_real_ann','eps_G_P'};
fieldnames(myirfs1)
myirfs1.eps_G.var = var_aux3 % assign the data array to the struct variable
irf_fisc = struct();
for i = 1:numel(var_aux)
irf_fisc.var_aux{i} = [0,myirfs1.eps_G.var{i}]';
end
irf_fisc.var_aux(1)
irf_fisc
% what is the write syntax to assign value (simulated data) to the struct?
myirfs1.eps_G.logy = data(:,1)/10; %Is the suggested code. but where is the data variable located? should I create it data = randn(TMax, N); or it is already simulated?
Dears,
I need your help. hocan I access the subfields within eps_G, where eps_G is a structure.
whos myirfs1
Name Size Bytes Class Attributes
myirfs1 1x1 374094 struct
%% disp(fieldnames(myirfs1))
>> disp(fieldnames(myirfs1))
{'eps_CP'}
{'eps_G' }
{'eps_T' }
{'eps_a' }
{'eps_nu'}
{'eps_z' }
% choose 1 or 2 below
shock_type = {'eps_G','eps_nu'};
var_aux = {'log_y','C','pi_ann','B_nominal','B','sp','i_ann','r_real_ann','P'};
var_aux2 = {'log_y_eps_nu','C_eps_nu','pi_ann_eps_nu','B_nominal_eps_nu','B_eps_nu','sp_eps_nu','i_ann_eps_nu','r_real_ann_eps_nu','P_eps_nu'};
var_aux3 = {'eps_G_log_y','eps_G_C','eps_G_pi_ann','eps_G_B_nominal','eps_G_B','eps_G_sp','eps_G_i_ann','eps_G_r_real_ann','eps_G_P'};
%Irfs of monetary and fiscal policy
irf_mon = struct();
irf_fisc = struct();
%% disp(fieldnames(myirfs1))
>> disp(fieldnames(myirfs1))
{'eps_CP'}
{'eps_G' }
{'eps_T' }
{'eps_a' }
{'eps_nu'}
{'eps_z' }
% when i run the following code it is unrecognized. can you suggest me what to do?
for i = 1:numel(var_aux)
irf_mon.(var_aux{i}) = [0,myirfs1(1).(var_aux3{i})]';
irf_fisc.(var_aux{i}) = [0,myirfs1(1).(var_aux3{i})]';
end
Unrecognized field name "log_y_eps_G".
This topic is for discussing highlights to the current R2025a Pre-release.
So you've downloaded the R2025a pre-release, tried Dark mode and are wondering what else is new. A lot! A lot is new!
One thing I am particularly happy about is the fact that Apple Accelerate is now the default BLAS on Apple Silicon machines. Check it out by doing
>> version -blas
ans =
'Apple Accelerate BLAS (ILP64)'
If you compare this to R2024b that is using OpenBLAS you'll see some dramatic speed-ups in some areas. For example, I saw up to 3.7x speed-up for matrix-matrix multiplication on my M2 Mabook Pro and 2x faster LU factorisation.
Details regarding my experiments are in this blog post Life in the fast lane: Making MATLAB even faster on Apple Silicon with Apple Accelerate » The MATLAB Blog - MATLAB & Simulink . Back then you had to to some trickery to switch to Apple Accelerate, now its the default.
I just published a blog post called "The Story of TIMEIT." I've been thinking about writing something like this ever since Mike Croucher's tic/toc blog post last spring.
There were a lot of opinions about TIMEIT expressed in the comments of that blog post, including some of mine.
My blog post today gives a more full history of the function, its design goals, and how it works. I thought it might prompt more discussion, so I'm creating this thread as a place for it.
If you are an interested user of TIMEIT, feel free to weigh in here with your thoughts. Perhaps the thread will influence MathWorks regarding what to do with TIMEIT, or with related performance measurement capabilities.
また、以下のリストから Web サイトを選択することもできます。
南北アメリカ
- América Latina (Español)
- Canada (English)
- United States (English)
ヨーロッパ
- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)
- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom(English)
アジア太平洋地域
- Australia (English)
- India (English)
- New Zealand (English)
- 中国
- 日本Japanese (日本語)
- 한국Korean (한국어)