Research

 

Research Focus

I conduct research within the area of ubiquitous computing with a specific focus on energy informatics, occupancy sensing and mobile sensing. I conduct my research with an experimental foundation at the intersection of ubiquitous computing, machine learning and system research with applications within energy informatics, Internet of things and smart buildings.

Energy Informatics

Within the emerging discipline of Energy Informatics people are researching, developing and applying information and communication technologies, energy engineering and computer science to address energy challenges. Read more about the many different research areas within energy informatics and our projects here. In my research I in particular focus on the application of energy informatics for smart buildings.

Selected Publication: Bo Nørregaard Jørgensen, Mikkel Baun Kjærgaard, Sanja Lazarova-Molnar, Hamid Reza Shaker, Christian T. Veje: Challenge: Advancing Energy Informatics to Enable Assessable Improvements of Energy Performance in Buildings. e-Energy 2015: 77-82, ACM. Go to document

Software Example: Brick

Occupancy Sensing

Occupancy behavior in a building context refers to all actions of the occupant (including presence) that affect building operation and control. Occupancy sensing is systems that measure, estimate, model and predict occupancy behavior based on inputs from building control, Internet of Things and mobile sensing infrastructures. My research has among others contributed with: 

  • Methods for fusion of data from multiple sensor modalities for recognition of occupancy presence and activities.
  • System support for occupancy sensing including meta data about sensors.
  • Methods for privacy-aware publication and sharing of occupancy data.

Selected Publication: Fisayo Caleb Sangoboye, Mikkel Baun Kjærgaard: PLCount: A Probabilistic Fusion Algorithm for Accurately Estimating Occupancy from 3D Camera Counts. BuildSys 2016: 147-156, ACM. Go to document

Software Example: OccuRE

Mobile Sensing

The advent of rich sensors in common smartphones has created new possibilities for using mobile sensing to gather data about the world, e.g., reflecting users’ actual rather than reported behavior. My research has among others contributed with: 

  • Methods for improving energy efficiency of mobile sensing in order not to drain the batteries of mobile devices.
  • Methods for scaling activity recognition with mobile sensing data to the level of crowds of people.
  • Methods for handling sensor heterogeneity across mobile devices.

Selected Publication: M. B. Kjærgaard, J. Langdal, T. Godsk, and T. Toftkjær, "EnTracked: energy-efficient robust position tracking for mobile devices," in Proceedings of the 7th International Conference on Mobile Systems, Applications, and Services (MobiSys 2009), 2009, pp. 221-234, ACM. Go to document

Software Example: EnTracked

Pervasive Positioning

Information about where an object is in the world enables a wealth of applications. Pervasive positioning denotes the system goal to provide positioning anytime, anywhere of anything. My research has among others contributed with: 

  • Methods for accurate indoor positioning with heterogenous measurements.
  • Methods for fusing data from different sensor modalities to improve combined indoor / outdoor positioning.
  • System support for developing position-based software solutions.

Selected Publication: M. B. Kjærgaard and C. V. Munk, "Hyperbolic Location Fingerprinting: A Calibration-Free Solution for Handling Differences in Signal Strength," in Proceedings of the Sixth Annual IEEE International Conference on Pervasive Computing and Communications (PerCom 2008), 2008, pp. 110-116, IEEE. Go to document