czech english

Robotour 2017 - Workshop

team presentations + PAIR'17 conference

On Sunday there is always a workshop of all teams after the contest. They share their findings what worked and what did not, what is in reality inside of their smart robots, how they liked the new rules etc. This year we were recording so although it is not the top quality you can see and hear what was presented.

Cogito (1st place)

Jiri Isa from Cogito team, the winner of the Robotour 2017 - outdoor delivery challenge, speaks about Testing, reliability and redundancy as the key factors leading to successful robot.

JECC Fesl (2nd place)

Udo Schmelmer from JECC Fesl team speaks about their waterproof robot, construction details, computational power etc.

Istrobotics (3rd place)

Radoslav Kovac from team Istrobotics, 3rd place in Robotour 2017 and winner from last year, speaks about their modified monster truck robot (original top speed 48km per hour), team history, 3D printed parts, new LED indicators, … You will also learn what were the root causes of some failures we have seen during the competition.

Short Circuits Prague (4th-5th place)

Pavel Jiroutek from Short Circuits Prague team speaks about their family team, comeback to Robotour, robot hardware (modified RC car), … Pavel presents his development cycle with simulator, replaying log files and access to robot webserver providing detailed info even during the real run. There is a small discussion about usability of ROS and experiences of various teams at the end of his presentation.

Planning in Artificial Intelligence and Robotics (PAIR'17)

Majer, Halodova, Krajnik: A precise teach and repeat visual navigation system based on the convergence theorem

We present a simple teach-and-repeat visual navigation method robust to appearance changes induced by varying illumination and naturally-occurring environment changes. The method is computationally efficient, it does not require camera calibration and it can learn and autonomously traverse arbitrarily-shaped paths.
During the teaching phase, where the robot is driven by a human operator, the robot stores its velocities and image features visible from its on-board camera. During autonomous navigation, the method does not perform explicit robot localisation in the 2d/3d space but it simply replays the velocities that it learned during a teaching phase, while correcting its heading relatively to the path based on its camera data.
The experiments performed indicate that the proposed navigation system corrects position errors of the robot as it moves along the path. Therefore, the robot can repeatedly drive along the desired path, which was previously taught by the human operator. The presented system, which is based on a method that won the Robotour challenge in 2009 and 2008, is provided as open source at
The research is supported by the Czech Science Foundation projects 17-27006Y and 15-09600Y.
If you would like to somehow support this contest or you have some comment/query, please use our contact form.