PhD scholarship in Intelligent Maintenance of IoT Infrastructures

Updated: about 2 months ago
Job Type: FullTime
Deadline: 15 May 2021

Are you interested in conducting cutting edge research in the intersection of the Internet of Things (IoT) and Artificial Intelligence (AI) focusing on resource-limited edge IoT devices? Are you a hands-on person, interested in low-level programming, applied machine learning, and deploying IoT devices in the real world? The Embedded Systems Engineering (ESE) section of DTU Compute offers a PhD scholarship funded by the newly established Digital Research Centre Denmark (DIREC) . The position is available from June 1, 2021, or later according to mutual agreement. 

With the emergence of IoT and the Industry 4.0 revolution, multiple industries automate and enhance the efficiency of their operations using networks of embedded devices, such as industrial sensors, actuators, and robots. Such IoT deployments are expected to operate robustly for extended periods of time. To that end, upon the deployment of an IoT network, maintenance operations need to be scheduled and performed. Examples of maintenance operations include recharging batteries and replacing faulty components, amongst others. IoT deployments often support critical operations that must not be interrupted. Hence, the scheduling of maintenance operations must also take into account application-layer constraints (for example, when some downtime is acceptable without risking delays in fulfilling customer contracts). How can we leverage machine learning and AI to manage and maintain IoT infrastructures intelligently and efficiently? 


This PhD project is focused on developing novel algorithms and systems to monitor, plan and support the maintenance of IoT deployments in real-time. To that end, the project will investigate and propose novel real-time AI-based predictive maintenance algorithms. Real-time predictions will be used as input to a maintenance planning framework that generates optimised maintenance schedules.

Some tentative tasks within the project include:

  • Resource-efficient maintenance planning in a resource-constrained industrial IoT environment based on on-device embedded machine learning.
  • Development of AI-based predictive maintenance algorithms based on data available from the project’s partners or other public datasets.
  • Investigation of on-device predictive maintenance in severely resource-constrained platforms, such as industrial sensors that are based on tiny microcontrollers.
  • Real-time maintenance planning in an unconstrained industrial IoT environment based on a traditional centralised learning infrastructure. The trade-offs between on-device and centralised machine learning shall be investigated.
  • Deployment and monitoring of a small IoT testbed at DTU Compute for the experimental evaluation of the proposed intelligent maintenance framework.


You must have a two-year master's degree (120 ECTS points) or a similar degree with an academic level equivalent to a two-year master's degree. Applicants with a 300 ECTS (i.e., 5 years) education at bachelor and master level are also encouraged to apply.

You must have a master’s degree in computer science or computer engineering or equivalent. You must have a strong background in applied machine learning and low-level systems programming in C. Knowledge of constrained programming is desirable. Some experience with low-power wireless networking protocols is also desirable. Experience in writing and publishing scientific papers is an advantage. You must be fluent in English, both speaking and writing, and possess excellent communication skills.

Application procedure

To apply, please read the full job advertisement at DTU Job and Career .

Application deadline:15 May 2021

View or Apply

Similar Positions