Projektpraktikum Advanced Lab Humanoid RoboCup

Lecturer (assistant)
  • Gordon Cheng [L]
  • Quentin Leboutet
Number0000003831
Type
Duration4 SWS
TermWintersemester 2018/19
Language of instructionEnglish
Position within curriculaSee TUMonline
DatesSee TUMonline

Dates

Admission information

Objectives

"At the end of this course, students are able to: - Use the robot operating system (ROS) with the NAO robot. - Create and implement algorithms from different areas including control, vision, planning, and learning, in the context of robot soccer. - Evaluate these algorithms on the robots. - Set up a team of robots which can play soccer against another team. Non-technical skills are also learned, students are able to: - Organize and manage a large engineering project. - Work in small groups and communicate the results achieved per group. "

Description

"Students should form a team for a soccer competition between robots. The student team designs, implements, and tests advanced robot control algorithms enabling the robots to play a soccer game. The final goal is to participate in the RoboCup competition which is an international initiative that fosters research in robotics and artificial intelligence. The RoboCup Standard Platform League (SPL) is the target league, in which the competitions are carried out with the NAO robots. The course is split into several phases: - Kick-off meeting: The students meet with a supervisor in order to set the schedule and to form groups. The splitting into groups should happen according to the students' knowledge. The students designate a team leader (preferably a master student) who stays in touch with one of the supervisors of this course. - Initialization phase: The students set up and configure the equipment (PCs, robots, software). - Development and test phase: The students design and implement the algorithms. This development goes hand in hand with the validation/test phase. - Final phase: Final test scenarios on the robot soccer field evaluate the performance of the robots. The scenarios should be compliant with typical scenarios occurring in the Standard Platform League. Qualification for RoboCup competitions: This practical project is also aimed at selecting those students who are motivated enough to continue to work on the soccer task throughout the year. This practical course should contribute to the long-term goal of establishing a RoboCup student team, who is able to qualify for and participate in the official RoboCup competitions."

Prerequisites

"Students should already passed the Projektpraktikum Introduction Lab Humanoid RoboCup (EINEU022). On the practical level, students must have very good programming skills in Pythone C++ and good knowledge in dumanoid robocup soccor ffield such as robot vision, robot learning, path planning, and locomotion. "

Teaching and learning methods

"Following teaching methods are used: - Introductory lectures - Application-specific tutorials (encompassing control, vision, planning, and learning) - Independent student work (including work in the laboratory with the robots)"

Examination

"The examination consists of the practical part (60%), a written report (20%), and a presentation (20%). In the practical part, students have to implement basic algorithms and demonstrate them in the context of a robot soccer scenario. The written report should reflect the ability to analyse and understand scientific and technical problems related to robot soccer, and the ability to apply knowledge from different areas of robotics, in order to provide solutions in form of the implemented algorithms. The presentation evaluates the ability to summarize the major facts and accomplishments."

Recommended literature

"L. L. Forero, J. M. Yanez, J. Ruiz-del-Solar: ""Integration of the ROS Framework in Soccer Robotics: The NAO Case"", RoboCup 2013: Robot World Cup XVII, Lecture Notes in Computer Science, Vol. 8371, pp. 664-671, 2014. Kober, Bagnell and Peters, Reinforcement learning in robotics: A survey, IJRR, 2013. "

Links