MATLAB: Am I having difficulties uploading data types, other than double, from the xPC Target application

dataintegerSimulink Real-Timetargettypeuploadxpcxpcscope

Why am I having difficulties uploading data types, other than double, from my xPC Target application?

I have a simple Simulink model which produces an integer array in the MATLAB workspace. It uses a type conversion to convert data to 16 bit signed integer. The yout data can be plotted and confirmed to be integer.

When the identical model is compiled for an xPC target, executed on that target and the output log uploaded, the resulting data appears to be invalid. The host interprets it as floating point data smaller than REALMIN.

How can I log integer data on an xPC target and upload and interpret it correctly it after a run?

Best Answer

This enhancement has been incorporated in Release 14 Service Pack 3 (R14SP3). For previous product releases, read below for any possible workarounds:

You were not able to log your data in the right format because xPC Target currently does not support this through the MATLAB Command Prompt. You can verify this by opening the RTW page from your Simulink Model. Click on Tools/Real-Time Workshop/Options to get to the RTW page. Then choose the 'xPC Target Code Generation Options' category . As you can see, the Signal Logging Buffer Size is in Doubles.

As a workaround please use the XPCSCOPE tool to export your data. The XPCSCOPE is useful for Signal Tracing from xPC Target. A tutorial on how to use this tool is available in the xPC Target User's Guide:

web([docroot '/toolbox/xpc/ch_basic_tutorial8.html'])

A summary of the steps to follow are:

1. Type the following at the MATLAB Command Prompt:

     xpcscope

2. Click File/New Scope

3. Open your model and build it onto the target

4. From the xPC Scope window click on Add/Remove Signals

5. Choose a signal to add and click on the Add Signal button

6. From the MATLAB Command prompt type:

      tg.start

7. Now click on the start button inside the xPC Scope window

8. Wait for the scope to collect the number of samples you entered, or in the Scope window, click the Stop button.

9. In the xPC Target Scope window, and from the Plot menu, click Variable Name for Export. You can change those variables and then click on Apply and then click Close

10. The Export button is then useful to export your data after a run or durring a run of your target application

Related Solutions

MATLAB: How to get a complete record of a signal from the xPC Target using xPC Target 2.0.1 (R13SP1)

You can use two xPC host scopes to collect the data. Have the first scope trigger the second scope and the second scope trigger the first scope. Then use a script in MATLAB to collect the data from multiple scopes and save the data into a single MATLAB variable while your simulation is running on the xPC Target.

See following steps for a complete example illustrating the above-described solution:

1) Download the files linked at the bottom of this solution and place them in a directory on your MATLAB path

2) Make sure that you are properly connected to your xPC Target box by running the XPCTEST command

3) Open the downloaded model by typing the following at at the MATLAB command prompt

test_logging

4) Build the test_logging model by selecting Simulation -> Simulation Parameters from the menu and selecting the RTW pane and clicking on the Build button

5) When the build completes successfully, type the following at the MATLAB command prompt to run the downloaded script

collect_data

MATLAB: Is there an example available that demonstrates how to gain access to global signals using the C-API in Real-Time Workshop

This enhancement has been incorporated in Release 2008b (R2008b) in the following demo 'Target-Based Data Interface: C-API Data Definition File' which can be opened by typing 'rtwdemo_capi' on MATLAB Command Prompt.

For previous product releases, read below for any possible workarounds:

The attached example application logs values of all global signals in a model to a text file. This code is intended as a starting point for gaining access to signal addresses by using the C-API in Real-Time Workshop. You can extend the code to perform signal logging and monitoring. The application:

Uses the "rtmGetDataMapInfo" macro to gain access to the model mapping information (MMI) substructure of the real-time model structure. In the following macro call, rtM is the pointer to the real-time model structure in model.c or model.cpp.

     rtwCAPI_ModelMappingInfo* mmi = &(rtmGetDataMapInfo(rtM).mmi);

Uses custom function "capi_StartBlockIOLogging" to initialize signal logging, custom function "capi_UpdateBlockIOLogging" to log a signal value at each time step, and custom function "capi_StopBlockIOLogging" to terminate logging and write the values to a text file.

The custom functions are integrated with model code by using the TLC file capitestsigfile.tlc. The TLC file places the following function calls in model.c:

     * Call to "capi_UpdateBlockIOLogging()" in the "model_output" function
     * Call to "capi_StartBlockIOLogging()" in the "model_initialize" function
     * Call to "capi_StopBlockIOLogging()" in the "model_terminate" function

To use the example, download the attached files and place them in your current working directory. The attached files include

     * capitestsigfile.tlc
     * capi_BIOSignalLog.h
     * capi_BIOSignalLog.c
     * example_capi.mdl

To configure and build the example, using Real-Time Workshop 6.1 (R14SP1) do the following:

1. Open the model "example_capi.mdl."

2. In the model window, click Simulation > Configuration Parameters.

2. Choose a target file. If you want to use a grt.tlc target file instead of the default ert.tlc target file:

a. In the left pane of the Configuration Parameters dialog box, click Real-Time Workshop.

b. Click Browse to open the System Target File Browser.

c. Select a grt.tlc target.

3. Integrate the custom application code with the generated code, by including the following string in the TLC options field of the Build process section of the Real-Time Workshop pane :

     -acapiSigTestFile="capitestsigfile.tlc"

4. Click Real-Time Workshop > Interface and select "C-API" for the Interface field under Data exchange. For the purpose of this example, you can limit the data exchange to “Signals in C API” only.

5. If you are using an ert.tlc file, select the following options on the Real-Time Workshop Interface pane:

    * Support complex numbers
    * MAT-file logging

6. Return to the Real-Time Workshop pane and clear the "Generate code only" check box. The Build button appears.

7. Click Apply.

8. Click Build. Real-Time Workshop generates an executable file.

To run the executable, type "!example_capi" at the MATLAB prompt. Real-Time Workshop creates the text file example_capi_ BlockIOLog.txt in your working directory. Examine this text file using an editor, such as the MATLAB Editor.

For more information about using the C-API to interface with signals, see the documentation accessed by typing the following at the MATLAB 7.0.1 (R14SP1) prompt:

web([docroot,'/toolbox/rtw/rtw_ug/f6010dfi2.html'])

Best Answer

Related Solutions

MATLAB: How to get a complete record of a signal from the xPC Target using xPC Target 2.0.1 (R13SP1)

MATLAB: Is there an example available that demonstrates how to gain access to global signals using the C-API in Real-Time Workshop

Related Question