The flexibility to explore concepts is essential to innovation.
Yet an idea that cannot be implemented in an actual system remains just thatan idea. In a typical development process, many engineers, scien- tists, and researchers rely on the flexibility of the MATLAB language to explore designs, but for system implementation, they must translate their high-level MATLAB algorithms into the lower-level language of C.
E mbedded MATLAB code enables a
MATLAB to C workflow that brings explo- During the translation to C, typical modi- fications include: equivalent and that no manual coding errors were introduced during development. ration and implementation together. Em- Optimizing to accommodate processor In addition, it is often necessary to write bedded MATLAB is a subset of the MATLAB speed and memory limitations C-code test harnesses to construct and technical computing language that lets you Introducing appropriate numerical repre- download tables and other data structures generate C code directly from MATLAB algo- sentations (floating or fixed point) on the processor. Writing and maintaining rithms. With this capability, teams can Incorporating real-time data manage- maintain a single design source and use one ment, such as buffering, streaming, and language in one development environment pipelining The Embedded MATLAB subset from concept to implementation. Manually translating MATLAB code to C includes more than 270 MATLAB opera- poses several challenges. First, the design and tors and functions, as well as functions from Closing the Gap Between Concept implementation teams use different tools and Signal Processing Toolbox, Aerospace and Implementation development environments, making it dif- Toolbox, and Fixed-Point Toolbox. It To build an embedded system, engineers ficult to communicate complex algorithmic includes MATLAB language features such typically follow a familiar paradigm: de- concepts. Second, the C code often diverges as matrices and arrays, real and complex velop in MATLAB, translate to C, and from the MATLAB concept code. This means numbers, structures, flow control, and compile for the target hardware, verifying that additional steps must be taken to verify subscripting. at each step. that the concept code and the C code remain
w w w. m a t h w o r k s .c o m Reprinted from TheMathWorks News&Notes | 2 0 0 9 1
TheMathWorks News&Notes
MATLAB code for a Kalman filter with three states (inset) and equivalent C code generated with the Embedded MATLAB subset. The 17 lines of MATLAB code translate to 144 lines of C code, including comments.
2 Reprinted from TheMathWorks News&Notes | 2 0 0 9 w w w. m a t h w o r k s .c o m
They are used in a wide range of applica- tions, including control, signal and image processing, radar and sonar, and financial modeling. The simple tracking algorithm is devel- oped in MATLAB in a file named kalmanf.m (Figure 1). This algorithm can be automatically translated to C using emlc, a command-line tool in Real-Time Workshop that generates C code from Embedded MATLAB code. Typical syntax for translation is >> emlc -s cfg -eg {A,B,C,Q,R,u,t,yv}... kalmanf.m The configuration parameter (follow- ing the -s delimiter) is used to specify code generation properties and fine-tune the Figure 1. Embedded MATLAB code for a Kalman filter. compiler. These options can be specified using the GUI shown in Figure 2. The example option (following the -eg delimiter) sets the data types and dimen- sions of function variables by specifying an example at the function interface. The generated C code is created in a file named kalmanf.c. In MATLAB, the algorithm is expressed in 17 compact lines of code based on matrix operations. The equivalent C code uses for loops for scalar computations of the matrix operations and comprises 144 lines. Because data types for all variables of the algorithm were introduced at compile time, the designer has the flexibility to develop multiple versions of the algorithm destined for different processors. The generated C code is readable, and it contains the same Figure 2. The Real-Time Workshop GUI for configuring code generation properties from Embedded comments as the original MATLAB code, MATLAB code. inserted at corresponding lines of the al- gorithm. This means that you only need to maintain a single copy in MATLAB for fur- these custom applications is cumbersome, a typical tracking algorithm implemented ther design iterations. The C code is gener- takes time, and consumes engineering re- in MATLAB. ated automatically, reflecting any changes. sources. With Embedded MATLAB code, The plot in Figure 3 shows the results the embedded algorithms and data struc- Generating C Code from of the generated code. These results were tures that you implemented in MATLAB can Embedded MATLAB Code verified as equivalent to the MATLAB be automatically translated to C. Kalman filters are recursive filters that simulation code. To illustrate the connection between estimate the state of a linear dynamic sys- concept and implementation, let us look at tem from a series of noisy measurements.
w w w. m a t h w o r k s .c o m Reprinted from TheMathWorks News&Notes | 2 0 0 9 3
Incorporating MATLAB Code into Simulink and Stateflow Models For many teams, the job is complete once the generated C code has been verified against the elaborated MATLAB code and the com- piled algorithms are deployed to the target. If the algorithm will be part of a larger, more complex system, you can extend the workflow by placing precompiled algorithmic func- tion blocks developed using the Embedded MATLAB subset in your Simulink models or Stateflow charts (Figure 4). You can then per- form system-level simulations and generate C or HDL code for implementation. Embedded MATLAB serves as a com- mon thread between MATLAB, Simulink, and Stateflow. In this way, it streamlines the workflow between algorithm developers and system modelers and brings exploration and implementation together. Figure 3. Output from the MATLAB and C-code versions of the Kalman filter.
Figure 4. Simulink radar system model with the Embedded MATLAB function block referencing the Kalman tracking filter.
Resources
Webinar: Algorithm Design and
Code Generation with Embedded MATLAB www.mathworks.com/nn9/wbnr30822
2009 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be 4 trademarks or registered trademarks of their respective holders. 91786v00 11/09
Python Advanced Programming: The Guide to Learn Python Programming. Reference with Exercises and Samples About Dynamical Programming, Multithreading, Multiprocessing, Debugging, Testing and More
Python Advanced Programming: The Guide to Learn Python Programming. Reference with Exercises and Samples About Dynamical Programming, Multithreading, Multiprocessing, Debugging, Testing and More
