Webinar Details AI-Powered Virtual Sensor Model Development Workflow
A free webinar will showcase a comprehensive workflow for creating AI-based virtual sensor models. This process covers the entire lifecycle, from initial design and training to validation, verification, compression, and deployment onto embedded processors. The workflow is designed to operate within a single, integrated environment. Key features include the ability to integrate AI models directly into Simulink for system-level simulation and testing, as well as the application of formal verification methods to ensure predictable neural network behavior. Participants will learn how to reduce the memory footprint and increase the execution speed of AI models through compression techniques. The webinar will also demonstrate the generation of library-free C code from these AI models, followed by Processor-in-the-Loop (PIL) testing. Attendees will gain insights into profiling code performance and evaluating trade-offs in design and model selection. The entire process, including the design and training of AI-based virtual sensors, is facilitated using MATLAB.
AI-driven virtual sensor modeling offers a pathway to enhance system efficiency and reduce reliance on physical hardware. By integrating AI development within platforms like Simulink and MATLAB, developers can streamline the process from concept to deployment. The emphasis on formal verification and code optimization for embedded systems addresses critical challenges in ensuring reliability and performance in real-world applications. This approach highlights a broader trend towards sophisticated AI integration in engineering, enabling more complex simulations and predictive capabilities. The trade-offs between model complexity, computational resources, and verification rigor will be crucial considerations for widespread adoption.
AI-generated to prompt reflection — not editorial opinion, not advice, not a statement of fact. How this works.