Starkit

← Back to teams list

Team website
Qualification video
Team short paper
Hardware specifications

Software description

walking

Please give a brief summary of your walking algorithm (max. 1000 characters).

We took Rhoban Walk Engine without any changes. Code available at: https://github.com/Rhoban/kid_size_public/blob/master/Motion/engines

vision

Please give a brief summary of your vision algorithm, i.e. how your robots detect balls, field borders, field lines, goalposts and other robots (max. 1000 characters).

Now we use two-step algorithm for detection ball. Firstly we looking for ROI and then classify this ROI with Neural Network. We also carry out experiments with object detection networks like yolo-tiny v3. It shows quite good results. Moreover we want to use CNN for opponent detection. Now we have simple line detection algorithm. It consist of hough transform and then filtering with size of lines. We want to increase quality and stability of this algorithm.

localization

Please give a brief summary of how your robots localize themselves on the field (max. 1000 characters).

For localisation we use Particle Filter. We extract posts, lines, angles and intersections as observation for Particle Filter.

behavior

Please give a brief summary of your behavioral architecture and the decision processes of your robots (max. 1000 characters).

The behaviour of the robot is designed using finite state machines. Changing states are based on various information such as game status, time spent in current state or information from the vision and localisation modules. Also we use BehaviourViewer from Rhoban team to test new behaviours in simulation.

contributions

List your previous participation (including rank) and contribution to the RoboCup community (release of open-source software, datasets, tools etc.)

– Robocup Asia-Pacific 2019 – 1st place in Humanoid KidSize – Robocup 2019 – 1st place in SPL Challenge Shield – Robocup German Open 2019 – 3rd place in Humanoid KidSize – Robocup Russia Open 2019 – 1st place in Humanoid KidSize Our team was an active organizer of Robocup Asia-Pacific 2019 Moscow. We made and provided the field, gates, Wi-Fi router, and also we were the TC/OC of the competitions in the Humanoid league of RCAP 2019 Moscow. In general, our activity is aimed at popularizing RoboCup in Russia. Moreover, we plan to organize a tournament in the Humanoid soccer league in Russia on the basis of our institute in the spring of 2020. We gave a lot of interviews to newspapers and magazines. Our interviews were shown on Russian television. An article in the popular Russian magazine “Popular Mechanics” was dedicated to our team and robosoccer. An article in a scientific journal is listed in the Publications section of this application.

publications

Please list RoboCup-related papers your team published in 2019.

THE ROBOSOCCER AS A MODERN EDUCATIONAL PLATFORM IN THE FIELD OF ARTIFICIAL INTELLIGENCE. R. Gorbachev, S. Semendyaev, I. Khokhlov, V. Litvinenko, I. Ryakin. IEEE Proceedings of international conference on artificial intelligence: applications and innovations (IC-AIAI 2019), p. 55-57, 2019 (DOI 10.1109/IC-AIAI.2019.00019)