Abstract

Energy-Efficient Location Tracking with Mobile Phones, Mattias Linnap

Many recent context-aware systems use mobile phones with a GPS sensor to track a person’s location. However, because of high power demands of the GPS sensor and wireless communication, these systems update a person’s position infrequently, or rely on the user to trigger position updates at meaningful places. An ubiquitous location tracker systems should be able to continuously track and update a person’s movement without interactive hints from the user, be accurate enough for the context-aware applications, and consume only a small fraction of mobile phone’s power. A model-based sensing approach will let the context-aware application set required accuracy bounds and minimise the energy used to fulfil them. I will present results of an experiment that compares model-based location tracking ideas in the literature, and the need and directions for further research.

Towards accurate dead-reckoning using optical sensors, Agata Brajdic

The goal of this talk is to present findings about using optical sensors as a basis for indoor object localization and tracking. An optical sensor provides information about object’s movements in two directions. In order to calculate both object’s displacement and orientation, (at least) two sensors need to be employed. Important is also the choice of the lens that is placed in front of the sensor’s CCD . I will present and discuss results obtained with experimental setup based on ADNS -2610 optical sensors, which forms the hardware ground for my PhD work on localization and tracking of multiple objects. The talk will be structured as an overview of my first year PhD report.