The Fulbright Germany‘s Berlin Seminar offered a variety of panels, workshops, excursions and seminars – one of which was given by Tim about bees, robots and how Fulbright helped continuing his career in California. Thanks for the invite, Fulbright Commission!
Our manuscript on our automatic dance detection and decoding system was published in PLoS One!
Prof. Kirk Visscher (University of California, Riverside) will give a talk on monday! House hunting by committee in honey bee swarms: The blind intelligence of distributed decision-making A swarm of honey bees must locate a new nest cavity rapidly, but… Continue Reading
A few days ago we uploaded a manuscript on our new vision system for the decoding of honeybee dances to arxiv. Promptly, MIT Technology Review publishes an article on it: https://www.technologyreview.com/s/608796/machine-vision-decodes-honeybee-waggle-dances/
Honeybees are extraordinary navigators. We study their behavior in a joint project with Prof. Randolf Menzel who uses a harmonic radar system to track the flight of individual bees. In one project branch we analyze these flight paths and ask… Continue Reading
The “Research to Market Challenge” is a competition for research-based product and business ideas from Freie Universität Berlin, Charité Berlin and cooperating institutes. Tim was awarded the second place in the category “Digital” for his bee-inspired peer-to-peer charging concept for… Continue Reading
In our research project “NeuroCopter” we study the honeybee brain while the animal is navigating. We attach living honeybees to a quadcopter, fly over known or unknown terrain, and observe the bee’s behavior and brain activity. This is our first… Continue Reading
In our collaboration with James Nieh (UC San Diego) we investigated how bees that follow either the waggle or the tremble dance are attracted to the signalling bee. We tracked the movements of followers and dancers in video recordings and analyzed… Continue Reading
Solving problems in Machine Learning very often relates to the question how to represent the data. While in the past most problems were approached with manually crafted features, we observe a move towards learning algorithms that derive robust and informative features. The 5th International… Continue Reading