Vision Centric Challenge Mosaic

Transcription

1 Vision Centric Challenge Mosaic Computer vision gives robots the ability to see. Using cameras as the main sensory modality of robotics has the following advantages: Low cost compared to expensive laser scanners, Richness of information, and Low power consumption compared to laser scanners. In order to promote research & development on computer vision and autonomous mobile robotics, we challenge college students as well as talented high school students with the following Vision based Robot Competitions during the Robofest 2016 competition season. Team Age Definitions College Enrolled: maximum 2 members per team. Could be graduate or under-graduate level. Challenge Synopsis Due to the camera s limited field of vision, the robot with limited height can see only a part of a numerical digit or alphabet letter pattern on a mosaic comprised of 15 pieces of colored paper on the floor arranged in 5 rows and 3 columns. The robot must move to read all paper colors necessary to identify the digit or letter represented. For example, the figure below represents the number 2 and letter A. The robot must report (display) the recognized digit or letter after spinning approximately 720 degrees twice on the field. See Appendix A for all the digit patterns and Appendix B for all the letter patterns to be used for the challenges. Challenge Field Setup Description Two different color papers will be taped on a mat, or carpet. The exact colors of papers and mat are unknown until the day of competition. The mat or carpet will be placed on an unknown floor color. Letter size (27.9cms x 21.5cms) papers, and not A4 papers papers, will be used to create the mosaic. The gap between columns will be between 15cms and 20cms. The gap between rows will be between 5cms and 10cms. Starting location could be any: A, B or C Camera angle will be released on Jan 14, Lighting conditions on the course are unknown and dynamic.

2 Competition Rules There will be 3 rounds. For each round, each robot has a maximum of 4 minutes to report the character. All the robots will be impounded (quarantined) before starting each round. The Judge will start the robot. Teams are NOT allowed to touch the robot after impounding. The team must provide verbal or written instruction to the Judge indicating how to start the robot. The winner will be decided by the number of successful rounds. To complete a successful round, the robot must first spin twice (~720 ) on the field and then display the identified character to the Judge. If multiple teams tie for the number of successful rounds, the teams will rerun with more difficult patterns until a winner is decided. See Appendix C for examples on tie. The robot will be considered off-course and disqualified from that round if the robot leaves the mat field completely. Team members cannot have any interaction with the robot. For example, giving sound or visual signals to the robot is not allowed. Robot Requirements Must be autonomous. (No remote control by human driver or remote computer is allowed) Any robot platform with only one camera is allowed to enter the competition. Camera angle limitation may be enforced on January 14. You may use any other sensors such as a digital compass. Any programming language can be used. Width: must be less than 61cms. Length: less than 92cms. Height: maximum 61cms. Weight: no limit. Competition Date and Location March 5, 2016 at the American University of Beirut. Attendance and team registration on the day starts at 8AM and ends at 9AM Questions regarding rules: farah@etclb.com Question regarding registration, and event details: info@etclb.com Challenge updates will be released on our facebook page: facebook.com/nerdlebanon and on our wikspace for registered teams only. National Champion is announced on March 5 during the #NERD16 closing ceremony and qualifies to the RoboFest world championship at Lawrence Technological University, Southfield, Michigan.

3 Appendix A Digits (0~9)

4 Appendix B Letters (A, B, C, D, E, F, H, J, K, L, M, P, R, S, T, U, W, X, Y)

5 Appendix C Some examples of possible variations (3, 4, A, C)