Feedback

Release 2019a Highlights

Release 2019a offers hundreds of new and updated features and functions along with 10 new products. 

You’ll find new capabilities for your projects – no matter what you’re working on.

You can now train deep neural networks to solve controls, autonomous systems, and robotics problems using reinforcement learning algorithms in MATLAB and Simulink. 

You can also train deep learning networks on 3D image data, and combine LSTM networks and CNNs for video classification and gesture recognition.

You can automatically generate MATLAB code for the layergraphs you create and CUDA code for the ONNX and Keras networks you import.

19a also includes new and updated capabilities for systems engineering. System Composer is a new product for designing and analyzing system and software architectures so that they’re ready for implementation in Simulink.  

You can now create standalone Stateflow charts in MATLAB. This means you can design complex state machines without having to write deeply nested if-else or switch-case statements.

If you are designing automotive systems, you can take advantage of new products and features – including using data from high-definition maps that have been specifically designed for automated driving applications, analyzing energy and power losses at the component and system level and, with the new AUTOSAR Blockset, you can model and simulate AUTOSAR Classic and Adaptive software. You can also automatically generate C/C++ production code and export AUTOSAR XML files using Embedded Coder.

There are plenty of other updates to MATLAB and Simulink, including projects in MATLAB, which help you organize, manage, and share your work.

And the new Simulink Onramp, a hands-on tutorial that takes you step-by-step through the basics of Simulink. 

There’s plenty more in this latest release. To find out what else is new, go to mathworks.com, where you can download 19a and get started. 

Product Focus