Bringing AI to Real-Time Insights, using FPGA, GPU and CPU
Artificial Intelligence (AI) has made machine learning techniques unavoidable in many systems to gain insights to predict future behavior. To collect these insights in real-time, these algorithms should be deployed on embedded platforms like CPUs, GPUs or FPGAs.
In this webinar we will be looking at a Battery Management System where neural networks can be used for battery State of Charge (SOC) estimation. An accurate determination of the State of Charge (SOC) in a battery indicates to the user how long they can continue to use the battery-powered device before a recharge is needed.
The development of these systems remains complex by the underlying theory and the mapping to programming languages like C, CUDA and VHDL. MATLAB® and Simulink® offer an integrated framework dedicated to the design and evaluation of your machine learning algorithms. Thanks to its code generation capability adopted by many manufacturers, it simplifies the deployment on embedded targets (CPU, GPU, FPGA).
Join this webinar to discover the process to follow to achieve this deployment and the tools at your disposal to converge towards an efficient and verified implementation.
- Using neural networks for battery State of Charge (SOC) estimation
- Creating and training an artificial neural network
- Integrating neural networks with Simulink
- Converting to fixed-point and optimizing on resources
- Generating C, CUDA and VHDL code