Edge AI: build smart products and machines with Embedded Machine Learning
Edge machine learning (ML) runs AI models locally on devices like sensors or machines, enabling real-time decisions, lower latency, improved privacy, and less reliance on connectivity. Ideal for smart, autonomous systems in tough or remote environments.
For whom?
- Product builders and machine builders that want to embed intelligence in their products and machines.
- R&D engineers , R&D managers, product managers, innovation managers.
Key takeaways
- Development guidelines for Embedded ML tailored to SMEs
- Live demos and discussions on different edge platforms (microcontroller, NVIDIA Jetson nano, Google Coral, Raspberry Pi), edge libraries and tools (TensorFlow Lite, Edge Impulse)
- How to quickly build a proof-of-concept
- Insights in required effort, skills and make-or-buy decisions
- Inspiring industry cases
Programme
Time | Subject | Speaker |
| 12:30 | Welcome and sandwiches | |
| 13:00 | Drivers for Edge AI, value and impact When does edge machine learning make sense? What is the impact on your product and product development? What can we learn from existing products with Edge AI? | Pieter Beyl - Sirris |
| 13:30 | Development guidelines for Embedded ML tailored to SMEs We cover all key steps—from problem identification, defining objectives, data collection, data analysis and preparation, model selection, training, model optimisation, deployment, and finally monitoring and updating the solution. | Miguel Lejeune - Sirris |
| 14:10 | Live demo Edge Impulse: Quickly build a proof-of-concept for Anomaly Detection and Classification on an STM32 Microcontroller This demo covers the various steps of the Edge Impulse workflow to illustrate how to quickly build and deploy an end-to-end embedded ML solution for anomaly detection and classification. | Frank Van den Broek - Sirris |
| 14:30 | Coffee break & networking | |
| 14:45 | Live demonstration: training an embedded ML model and deployment on a microcontroller In this demonstration we use open-source libraries to cover different aspects involved in training an embedded neural network such as data augmentation, quantization and pruning, and the deployment of the model on a microcontroller. | Frank Van den Broek - Sirris |
| 16:10 | Powerful edge AI with NVIDIA Jetson Nano When to use the NVIDIA Jetson Nano, libraries, pitfalls and test results for a segmentation model. | Brecht De Cock - Sirris |
| 16:30 | Closing remarks | Frank Van den Broek - Sirris |
| 16:40 - 17:00 | Networking drink |
Flemish companies can take advantage of a subsidy from VLAIO for this course
This programme is part of Industriepartnershap in which 13 Flemish innovation partners offer an integrated service to stimulate growth and innovation in the Flemish industry in the 6 following themes: AI, Cybersecurity, Circularity, Digitisation, Climate & Energy and Industry 4.0. They do so under the leadership of Agoria and Sirris and with the support of Agentschap Innoveren & Ondernemen.
Contact person
Read more about this expertise
For all practical information, information on invoicing, location etc of this event, please contact our event manager
adres events@sirris.be
Date
Location
Price
Language
English
Status
Similar activities
Training in foundry technology: gain in-depth, practical knowledge - Module 2
Wear Less, Last Longer: how tribological coatings and surface treatments extend the lifetime of tools and components
Sirris Digital Innovation Day: Multimodal Sensing & AI for Industrial Innovation
