Skip to main content

Intelligent Systems

An embedded system consists of a combination of hard- and software components, either fixed in function, described by hardware description languages and executed on a FPGA/ASCI, or programmable on an application specific micro- or digital signal processor or on a full featured generic application processor running an full featured operating system.

Embedded Systems have become an enabler for the rapid growing field of intelligent and connected Internet-of-Things (IoT) ecosystems. Due to the increasing computational power and the communication capabilities of embedded systems, the functional demands and requirements expected from an embedded system have been increased over the past years.

This calls for a new generation of intelligent embedded systems with modern state-of-the-art algorithms and inter-system connection based on modern IoT protocols, which serve the demands in smart interconnected industrial control systems, power electronic systems and machines.

Also, the algorithms executed on an embedded system have evolved towards more intelligent algorithms taking advantage of machine learning and artificial intelligence tools. Some modern embedded architectures come with build in hardware AI accelerators to accelerate artificial intelligence and machine learning applications.

Through our research and development projects, but also due to our teachings in embedded architectures we have a long experience in applying full featured System-On-Modules (SOM) with modern ARM processors running full embedded Linus operating systems as well as state-of-the-art System-On-Chip (SoC) devices, such as the Zynq-7000 family, which integrates the software programmability of a ARM processor with the hardware programmability of an FPGA.

Research scope

Investigation and research of methods and architectural designs for distributed (decentralized) decision making in embedded edge nodes

  • Secure and reliable fully decentralized IoT protocols.
  • Mathematical optimization and machine learning based algorithms for embedded edge device decision making.
  • Embedded architectures based on embedded Linux and System-On-Chip (SoC, e.g. Xilinx Zynq) hardware.
  • Intelligent battery and charge management systems.

We have a strong focus on applications on intelligent embedded systems for IoT applications coupled with modern artificial intelligence and machine learning algorithms and modern optimization algorithms for real-time decision making.

In our research we aim on developing and show-casting solutions based cutting edge modern embedded system and IoT technologies, coupled with modern machine learning and artificial intelligence algorithms.

Specifically, we focus on development of intelligent embedded systems, which are key enabler in the field of electrification and modern power electronic systems such as battery and charge management systems.

In our research we focus is on modern charge  controller and battery management systems for electric vehicles and electrified systems and related IoT protocols such as MQTT, SMQTT, CoAP as well as proprietary protocols based on classical networking and modern IoT mesh networking protocols such as the Thread protocol. We are especially interested in the efficiency and reliability of modern IoT protocols when used in meshed or star network topologies. For example, intelligent charge controllers for electric vehicle charging infrastructures, which are interconnected via an IoT middleware and coordinate charging of electric vehicles in a distributed decentralised load management system.

Research group 


Find selected projects here:


Find latest publications here:

  • A decentralised electric vehicle charge management system based on WIFI mesh networking and the Thread protocol. Lass T., Brehm R., Qian K, 2020. In preparation for IEEE International Conference on Computer Communications 2020.
  • Experimental evaluation of a method for simulation based learning for a multi-agent system acting in a physical environment. Qian, K., Brehm, R. W. & Duggen, L., 2019, Proceedings of the 11th International Conference on Agents and Artificial Intelligence. Rocha, A., Steels, L. & van den Herik, J. (red.). SCITEPRESS Digital Library, Bind 1: ICAART. s. 103-109, Read more
  • A framework for a dynamic inter-connection of collaborating agents with multi-layered application abstraction based on a software-bus system. Brehm, R., Redder, M., Flaegel, G., Menz, J. & Bruce-Boye, C., 1. aug. 2018, Intelligent Decision Technologies 2018: Proceedings of the 10th KES International Conference on Intelligent Decision Technologies (KES-IDT 2018, Read more

Associate Prof. Robert Brehm Tel. +45 65501612 /

Send email

Last Updated 29.01.2021