Projects

SwarmRob – A Toolkit for Reproducibility and Sharing of Experimental Artifacts in Robotics Research



Due to the complexity of robotics, the reproducibility of results and experiments is one of the fundamental problems in robotics research. While the problem has been identified by the community, the approaches that address the problem appropriately are limited. The proposed toolkit tries to deal with the problem of reproducibility and sharing of experimental artifacts in robotics research by a holistic approach based on operating-system-level virtualization. The experimental artifacts of an experiment are isolated in „containers“ that can be distributed to other researchers. As a result, a novel experimental workflow to describe, execute and distribute experimental software-artifacts to heterogeneous robots emerged, that supports researchers in executing and reproducing experimental evaluations.

Seerose – Service Robots in Smart Homes



The research project deals with the integration of service robots into an intelligent home environment to cooperatively accomplish tasks. This combination will extend the sensory variety of the smart home and the service robots and will allow more sophisticated applications. One focus of the project is the development of an adequate middleware architecture to enable communication between different agents in the system, e.g. robots, sensors and actuators. Furthermore, the robots will be programmable by employing user-friendly learning from demonstration techniques. The whole project is embedded into a health care scenario to support elderly people in their everday lives.

Further information can be found at: http://www.iot-minden.de/course/seerose/


Fall and Activity Recognition in Smart Homes


The project focuses on the development of a fall detection and activity recognition system and the integration of the system into a smart home environment. It is based on a small-sized customized wearable device that is attached to the user’s waist. Activity recognition is performed locally on the wearable with its restricted computational capabilities in real time. A special focus is set on the porting of appropriate machine learning algorithms onto the chip. The smart home integration of the system allows urgent support for fallen people as well as a fast response of the intelligent environment depending on the current user’s activity.

Further information can be found at: http://www.iot-minden.de/course/smart-fall/