A micro-intelligence engine learns in a closed system
The Fraunhofer Institute for Microelectronic Circuits and Systems IMS has developed an artificial intelligence (AI) engine for embedded systems and microcontrollers. It does not require any connection to the cloud.29 Jun. 2019 Thomas H. Grimm
Researchers at the IMS refer to the system as sensor-oriented AI. Contrary to the general cloud computing trend, the system is intended to facilitate machine learning on microcontrollers. The devices can be equipped with additional functions without the need for any hardware modifications. As no sensitive data leaves the system, data protection is much simpler. The data volume to be transferred can be reduced to a minimum.
Artificial Intelligence for Embedded Systems , AIfES for short, is a fully configurable neural network on microcontrollers. It focuses on what is referred to as microintelligence in order to improve the interaction between humans and technology. Further fields of application include measurement technology, Industry 4.0 and hardware accelerators. A machine learning library can be used to implement self-learning small electronic devices that do not require a cloud connection or support from more powerful computers. The library was developed in the programming language C and can be used for other platforms such as PC, Raspberry Pi and Android. The AI’s neural network also works for deep learning. The source code was specially optimized for embedded systems by reducing it to the minimum.
Interested in news about exhibitors, top offers and trends in the industry?
Your web browser is outdated. Update your browser for more security, speed and optimal presentation of this page.Update Browser