In the frontline of AI solutions
Machine learning algorithms are often applied also on embedded devices. This is because the amount of data rapidly accelerates and computing in the cloud does not always scale for the real time performance required. AFRY can assist you in the development and integration of machine learning algorithms on embedded targets.
Edge computing, for example computing close to the data input sources, is a hot topic of modern connected IoT infrastructures. Edge computing in combination with embedded AI/machine learning solutions opens for new applications and services such as faster and smarter autonomous systems.
Semiconductor manufacturers supports the edge computing trend by implementing hardware solutions for AI applications in their silicon's also for embedded use. The toolchains for implementing machine learning solutions on embedded targets are tailored for this to support execution on hardware with limited resources.
We assist you in selecting the right hardware, models, methods and toolchains to enable the use of machine learning technology in your products that can put you ahead of your competitors. We also work on algorithm software implementations and on machine learning related to autonomous systems.
AFRY has an internal global AI competence network that enables our engineers to stay in the frontline of machine learning R&D. You can make use of this knowledge base to get a competitive advantage.
What we offer
- System analysis – Is a machine learning algorithm the best solution for your embedded system?
- Selection and tailoring of machine learning algorithms
- Dataset selection
- Model training
- Algorithm implementation/integration on target
- Testing in real environment