 |
| |
|
|
| |
|
TECHNICAL UPDATES |
| |
|
Products Updates |
| |
|
Link for Analog Devices VisualDSP++®
For Embedded Software Programmers developing with
VisualDSP++ who want to automate their verification tasks
and reuse existing MATLAB/Simulink models as test benches
for embedded code (hand written or automatically generated)
running on Analog Devices DSPs, Link for Analog Devices VisualDSP++®
offers the ability to automate verification of VisualDSP++
projects from MATLAB and Simulink. Unlike other manual solutions,
Link for Analog Devices VisualDSP++ enables the embedded software
engineer to use MATLAB scripts and Simulink models to automate
testing and verification of embedded code on Analog Devices
DSPs supported by VisualDSP++ platform.
For Algorithm
Developers and Systems Engineers working in Simulink who want
to use automatic code generation for rapid prototyping on
Analog Devices DSPs, Link for Analog Devices VisualDSP++®
offers the ability to generate projects for VisualDSP++ and
to perform processor-in-the-loop simulations from Simulink
models using code generated by Real-Time Workshop Embedded
Coder. Unlike manually coding projects in VisualDSP++, Link
for Analog Devices VisualDSP++ automates the task of generating
C code from Simulink models and creating VisualDSP++ projects
with the generated C code that can be compiled, executed,
and tested on Analog Devices DSP supported by VisualDSP++
platform.
For more
information about Link for Analog Devices VisualDSP++®,
please visit the following URL:
http://www.mathworks.com/products/visualdsp/
Simulink® Design Verifier
For embedded
system developers using Simulink or Stateflow who find it
difficult to develop complete tests for executable specifications
of their components, Simulink Design Verifier is an automatic
test generator and
property prover that significantly reduce the effort required
to verify your design.
Unlike
time-consuming, manual test development, Simulink Design Verifier
uses formal methods to automatically generate tests that guarantee
complete structural and functional test coverage.
Unlike
other test generators on the market, Simulink Design Verifier
takes advantage of Simulink's production code technology to
provide an integrated tool chain for producing fully tested
and verified embedded code.
For more
information about Simulink Design Verifier, please visit the
following URL:
http://www.mathworks.com/products/sldesignverifier/ |
| |
| |
| Tips
and Techniques |
| |
|
Multithreaded
Computation in MATLAB
If you are currently using a multiprocessor or multicore
system, you can now enable multithreaded computation to improve
MATLAB computation performance. By default, multithreaded
computation is not enabled; to enable it, use MATLAB preferences.
The preference automatically detects the number of CPUs on
your system and recommends the number of threads based on
that.
Enabling
Multithreaded Computation
To enable multithreaded computation, follow these steps:
- Select
File > Preferences > General > Multithreading.
The General Multithreading Preferences panel opens.
- On
the General Multithreading Preferences panel, select Enable
multithreaded computation. Specify the Maximum number of
computational threads.
Accepting
the Automatic option is recommended-MATLAB automatically sets
the value to the actual number of computational cores on your
system.
Note that
if your system uses hyperthreading (where one processor is
logically configured as two), MATLAB sets the value to 1.
If you choose Manual, enter the maximum number of threads
you want to set; use a positive integer not greater than 16.
Selecting
a number other than the recommended value might increase performance
for some computations, but might decrease performance for
others.
Debugging
MATLAB M-files from the MATLAB Command Prompt
What
Debugging Tools Are Available at the MATLAB Command Prompt?
This section describes how to make use of functions to
debug programs from the MATLAB command prompt. There are altogether
seven topics and in the e-newsletter issue, we will look at
the fifth topic. The rest of the topics will be covered in
subsequent e-newsletter issues.
| Topics |
| 1. |
Setting,
Clearing, and Querying Breakpoints |
| 2. |
Moving
from Workspace to Workspace |
| 3. |
Executing
Your Code Using the DBSTEP Function |
| 4. |
Displaying
Status Messages Periodically |
| 5. |
Using
the TRY/CATCH Block to Capture Errors |
| 6. |
Using
the ERROR Function with the LASTERR and RETHROW Functions |
| 7. |
The
WHICH Function |
Topic 5: Using the TRY/CATCH Block to Capture Errors
There's one scenario that you don't want the failure of one
line of your code to cause your entire program to fail even
though you know it will fail under certain circumstances.
The way to allow this line of code to execute but capture
the error if it occurs is to use a TRY/CATCH block. The line
of code to be executed will be the TRY section of the block,
while the error handling code makes up the CATCH block.
For example,
in the sample code above, MATLAB will attempt to open a file,
read the first few characters, and close the file. The TRY/CATCH
block handles the case where the file you specify does not
exist or there is a problem opening it.
Call it
with the name of a text file (test.txt) as the input argument,
first 5 characters was returned correctly as the answer. Then
call it with a string that is not the filename of a text file
(test1.txt) on your system as input, if the TRY/CATCH block
did not exist, the attempt to open the nonexistent file would
return an error -- with TRY/CATCH, you can handle that scenario.
Another scenario that could cause problems is if the call
to FSCANF function in the example file failed and terminated
execution of the example. Without the TRY/CATCH block, the
file you opened would remain open, potentially causing it
to become corrupted or even causing your operating system
to run out of file handles. By using the TRY/CATCH block,
you can clean up open files and other resources you have used
even if the code you are using causes an error during its
execution.
|
| |
| |
|
EVENTS & TRAINING |
| |
|
Learn and do more with MATLAB & Simulink
'Test and Measurement using MATLAB' training workshop
Workshop
Highlights:
This one day hands-on workshop is designed to cover real-time
interfaces of MATLAB for data acquisition and instrument control,
including:
- Acquiring
measured data from a data acquisition device
- Communicating
with an external instrument via its GPIB
- Using
the Data Acquisition Toolbox to perform computations and
analyses of acquired data
- Using
synchronized analog I/O and digital I/O
In-house
or customized training is also available on request, please
contact Activemedia at 6742-8173 for details. Other relevant
training courses provided by Activemedia include:
- Comprehensive
MATLAB
- Advanced
MATLAB Programming Techniques
- Building
GUIs with MATLAB
- Applying
Image Processing Techniques with MATLAB and SIMULINK
- Applying
Neural Network with MATLAB |
| |
| |
|
Visit
www.activemedia.com.sg
or Contact us at: |
|
Singapore:
(65) 6742 8173
enquiry@activemedia.com.sg |
Malaysia:
(60) 3 7880 8522
enquiry@activemedia.com.my |
Thailand:
(66) 2 612 9390-1
info@activemedia.in.th |
|
| |
| |
|
Customer Applications |
| |
|
Land
Rover Vehicles Achieve EPA Certification with MathWorks Tools
for Embedded Code Generation and add2 Target Hardware
|
|
To enable four-wheel drive vehicles to pass emissions
tests requiring two-wheel drive operation |
|
|
Use MathWorks tools to design, simulate, generate, and
implement a MICROGen-based system to emulate the vehicle's
transfer case controller |
|
|
• Emissions certification achieved
• Development time cut by 75%
• Reusable solution implemented |
|
| |
|
How
can you force a vehicle built to operate in four-wheel
drive to operate in two-wheel drive? Guidelines set
by emissions certification agencies can require four-wheel
drive vehicle manufacturers to answer this question.
In many emissions testing procedures, the vehicle must
run on a dynamometer, which requires two wheels to remain
motionless while the remaining two drive wheels spin.
For a vehicle that always runs in four-wheel drive,
this mode of operation is impossible.
|
Testing a Land Rover
vehicle in a laboratory environment. |
|
|
When
faced with this problem for Land Rover vehicles, Jaguar
and Land Rover used MathWorks tools and MICROGen target
hardware from add2 Limited to emulate the controller
for the transfer case in the transmission. This solution
enabled Jaguar and Land Rover to operate the vehicle
in two-wheel drive and achieve emissions certification
from the Environmental Protection Agency (EPA) and other
agencies without a problem. |
| |
|
Challenge
To
test the vehicle on a two-wheel dynamometer, engineers
removed one of the vehicle's prop shafts. This change,
however, caused problems in the vehicle's control systems
as they sensed two wheels spinning and two wheels motionless-a
situation that typically occurs only during a real-world
catastrophic failure. In particular, the vehicle's transfer
case controller is a fully electronic system that would
not function properly in this mode of operation.
Because
they were departing from the intended production specification,
Jaguar and Land Rover needed to replace the transfer
case controller to stop the control systems from entering
a different operating mode and essentially destroying
the clutch. They also needed to emulate the signals
from the transfer case controller to the CAN bus so
that the rest of the car believed the controller was
still operating normally.
Jaguar
and Land Rover initially considered contacting the transfer
case controller manufacturer to create a one-off version
of the controller. This costly approach would have required
a minimum of 12 weeks, forcing Jaguar and Land Rover
to postpone scheduled emissions tests. |
| |
|
Solution
A team of three engineers from add2, Jaguar, and Land
Rover used Simulink®, Real-Time Workshop® Embedded
Coder, and Embedded Target for Motorola® MPC555
to design, simulate, test, and implement a MICROGen-based
system. With this system, Jaguar and Land Rover now
operate Land Rover vehicles in two-wheel drive for emissions
testing, saving weeks in development time.
The team used Simulink to create a high-level model
of the transfer case controller, including input from
and output to a CAN bus. They then ran simulations in
Simulink to verify that all output signals were responding
correctly to input events.
Using
Real-Time Workshop Embedded Coder and Embedded Target
for Motorola MPC555, the group then automatically generated
code from their Simulink model and deployed it to the
MICROGen target hardware, which is a generic target
with integrated signal conditioning for rapid control
prototyping, communications gateway, and smart data-logging
applications. The engineers disconnected the vehicle's
original transfer case controller and installed the
new MICROGen emulator for on-vehicle testing.
During
testing, the engineers would note any faults that were
raised by other control systems in the vehicle. They
then used Simulink to modify the models, before regenerating
the code and running additional tests. Updates to the
MICROGen code were done in the vehicle, enabling some
design iterations to be completed in fewer than ten
minutes.
Once
the team ran the car reliably in two-wheel drive, they
conducted extensive tests to ensure no more faults occurred.
The vehicle then underwent emissions certification testing.
Jaguar and Land Rover have since applied this approach
for a number of certifying authorities and continue
to expand its use on Land Rover models. |
| |
|
Results
-
Emissions certification achieved. MathWorks
and add2 products enabled Jaguar and Land Rover
to depart from the vehicle's intended production
specification. All vehicle systems see the valid
CAN signals as expected. As a result, Jaguar and
Land Rover can confidently present the emissions
test results to certification authorities around
the world. This approach has been used in successful
certification tests for Japan and for the EPA in
the U.S.
-
Development time cut by 75%. The transfer case
controller manufacturer needed 12 weeks to produce
a first version, which would likely require additional
time to make it work properly. MathWorks tools and
MICROGen enabled the team to complete the whole
project in fewer than three weeks, enabling them
to meet their scheduled test dates.
-
Reusable solution implemented. Jaguar and Land
Rover can rapidly reuse their solution on new vehicles.
The first system was developed for the Land Rover
LR3, and has been subsequently applied in much less
time to the Range Rover Sport and Range Rover. For
their investment in a solution for one car, the
team has three and potentially many more applications.
|
|
Products
Used
• Embedded
Target for Motorola® MPC555
• Real-Time
Workshop®
• Real-Time
Workshop® Embedded Coder
• Simulink® |
|
| |
|
|
|
