Most engineers use MATLAB as an analysis and simulation tool, regardless of discipline. Functions, algorithms, and calculations can be performed quickly without the tedium and strict code rules of a programming language. Provided tool kits provide pre-made analysis and simulation tools in a broad range of areas from machine learning to signal analysis to stress and strain analysis. MATLAB provides a broad and versatile set of tools and capabilities widely used across the range of engineering disciplines.
MathWorks expands the capabilities of MATLAB and other tools in the MATLAB suite with the release R2017a. Engineers can complete their analysis faster and visualize results better with improved performance and graphics.
Users can more easily annotate figures with the improved Live Editor. Users can add and edit titles, labels, legends and other annotations through a convenient user interface without having to write code in their m-file to annotate the graph. Users can also transfer data to other applications with live script outputs.
Analysts can perform a broader range of array analysis with improved capabilities for processing and analyzing arrays with sort, selection and statistical functions. Users can better visualize results using new heatmap chart functions that improve data visualization capabilities.
Users can access this release of MATLAB online through a web browser for better access from multiple locations and collaboration among geographically diverse teams.
The regression learner trains regression models to predict data.
Improved linear regression analysis allows users analyze relationships between response and predictor data. Users can analyze a data set to fit curves and develop equations with new vector auto regression models that support multivariate analysis for support analysis of multi-dimensional data.
Convolutional Neural Network functions can be colored and visualized.
Engineers and scientists using MATLAB to train and analyze neural networks will gain performance and capability. This release can autonomously parallelize and distribute processes across multiple CPU and GPU cores as well as clusters and cloud computing systems with flexible process scheduling improve performance and reduce run time.
Features extracted with convolutional functions such as wavelets can be visualized with multiple 2D arrays, one for each filter, and the set of filters can be graphically visualized as a 3D plot using color and intensity to show the third dimension.
Lane and vehicle detection with the new Automated Driving System Toolbox™
Engineers developing autonomous vehicles can use the new Automated Driving System Toolbox™ to design, test, and evaluate driving systems for self-driving cars and other autonomous vehicles. The system supports fusion and integration of input data from multiple sensors and provides tools to simulate sensor inputs, develop control responses, and simulate the effect of the control response. The toolkit includes algorithms such as lane marker detection, vehicle detection with machine learning, and image to vehicle coordinate transforms.
Revisions of other tools in the MATLAB suit also improve performance and capability. Simulink features include tools to easily update project files, streaming for large input signals and data sets, and simplified wiring and interconnection tools. Improved CAD import tools simplify importing models from other CAD tools for simulation and integration with MATLAB models. Improved verification and code analysis tools simplify code checking and verification and validation.
The latest release of the MATLAB suite offers a significant increase in performance and capability. Learn more about the latest release at the MathWorks Matlab web site.