Learn to develop virtual sensor models using advanced AI techniques such as feedforward neural networks, LSTMs, and decision trees.

Virtual sensor modeling is a robust approach to mimicking the behavior of physical sensors. Creating a virtual sensor, or soft sensor, is particularly useful when direct measurement is challenging or when physical sensors add unnecessary cost and complexity. For instance, in developing a BMS for electric vehicles or laptops, accurately determining the Battery State of Charge (SOC) is crucial but often not directly measurable. AI techniques, including deep learning and machine learning, can enhance or even replace traditional methods such as Kalman filters. However, when you integrate AI models into these applications, they must integrate with other system components, such as power limitation, fault detection, and cell balancing algorithms. Developing such complex systems requires integrating, implementing, and testing various components while reducing the need for costly and time-intensive hardware prototypes. Model-based design is a proven strategy to achieve this.

Through the example of Battery Management System (BMS) State of Charge (SOC) estimation, you’ll learn how to seamlessly integrate AI models into Model-Based Design. This integration enables design testing through simulation and allows implementation on target hardware via automatic code generation. We’ll also discuss balancing AI trade-offs, from model accuracy to deployment efficiency.

Agenda:

• AI for MBD Overview
• Designing and training machine learning components with Statistics and Machine Learning Toolbox
• Designing and training deep learning components with Deep Learning Toolbox
• Importing trained TensorFlow models into MATLAB
• Integrating machine learning and deep learning models into Simulink for system-level simulation
• Generating library-free C code and performing PIL tests
• Summary

Featured products

All products mentioned in this webinar are developed by MathWorks.

• MATLAB
• Simulink
• Deep Learning Toolbox
• Statistics and Machine Learning Toolbox
• Embedded Coder
• Simulink Coder

How to Access:

To watch the webinar, simply fill out the form on this page. Once registered, you’ll gain immediate access and can watch it directly on the page. Additionally, you’ll receive an email with a link to access the on-demand session at your convenience.