Project Group: Robolan

The initial planning of a large-scale WLAN, such as the eduroam network at our university, is a complicated task. Due to the multipath effects (e.g. reflection, refraction, diffraction, scattering) that affect RF signals as well as the complex structure of many buildings, even high-quality deployment planning tools often fail to properly predict an optimal set of positions for WLAN access points. As a result, ensuring that a particular deployment achieves the desired quality usually involves manual on site measurements with iterative refinements. In addition, since most buildings are subject to changes (e.g., removal of furniture, addition of more RF senders), the performance of deployments often degrades over time. Thus, in practice, maintaining a high-quality usually requires repeated measurements and fine-tuning.

The goal of this project group is to develop an autonomous robot platform to automate WLAN quality measurements. Besides from basic navigation functions (e.g., simultaneous localization and mapping, path planning, autonomous exploration) the platform shall encompass components to capture, store and retrieve measurements and map information as well as an application that provides a spatial visualization for network planners. The validation of the platform will be done experimentally using an existing robot in one of the buildings of the university.

From a theoretical perspective, the project group will cover basic concepts of autonomous mobile robots as well as the development of a small distributed system. The practical part will touch on a broad spectrum of hard- and software including robot control using ROS, computer vision and application development using web technologies. Students taking this course must be fluent in at least one object-oriented programming language (e.g., Python, C++ or Java) and should be able to apply their knowledge to other languages quickly. In addition, students should be able to work with Linux-based operating systems via a shell (e.g. bash, zsh). Note that after the initial introduction, we will be working most of the time in one of our labs. If you have other obligations (e.g. a daytime job), make sure to reserve enough time for the course or choose another project group that is more suited for working remotely.