Author: robolab

The national television made an AI-special, including a visit to theUniversity of Amsterdam. The Dutch Nao Team was at that training the new team-members, so the Intelligent Robotics Lab was fully operational.

The Dutch Nao Team was invited to play a number of demonstration games against the current World Champion: Nao Team HWTK from Leipzig. They will play at the Herbstmesse in Dornbirn, so they can expect a lot of audience.    

The night flight of a swarm of drones was on the cover of the July edition of Science Robotics. In the article the co-authors of Gutzi Eiben showed that the behavior, learned by evolution in a simulation of 1000 drones, could be reproduced with 30 real drones.

The TechLabs of the two Amsterdam universities work closely together. The Intelligent Robotics Lab has an extensive list of equipment which can be borrowed. This month our Nao robot Barathon was borrowed by the Computational Lexicology research-group, to demonstrate their Natural Language Processing at a keynote at the 21st International Conference on Text, Speech and…

At the RoboCup 2018 in Montreal, Canada, the Joint Rescue Forces have won the Infrastructure competition with their Machine Learning to the Rescue workshop.

As described in our team description paper, Luis Gustavo Nardin, Sebastian Castro and I will develop a workshop on how different machine learning techniques could be applied in the RoboCup Rescue Simulation League. We will do that during the RoboCup German Open 2018 in Magdenburg. For more details on the Machine Learning capabilities of Matlab,…

The Dutch Nao Team is qualified as one of the  24 teams worldwide to compete in the RoboCup Soccer Standard  Platform League. The list of competitors can be found at the RoboCup website.

Today the Dutch Nao Team published their qualification material: the Team Qualification Document and a Qualification video:  

The Dutch Nao Team visited for the 3rd time the Robotic Hamburg Open Workshop, organized by team HULKS from Technischen Universität Hamburg.

To perform rapid prototyping with machine learning libraries, which need more computing power than regular laptop’s and robot’s can provide, a new Machine Learning Machine is introduced to the lab. This machine not only has a good CPU, but also a good GPU. More details on this workstation can be found at github.