תכנון ופיתוח מערכת ראייה, עיבוד תמונה ולוקליזציה לקבוצת רובוטים המשחקים כדורגל בליגת KSL בתחרות רובוקאפ

שלחו לחבר
שנה
2017
אחראי אקדמי

Design and develop vision, image processing and localization for robots that play soccer

The students will design image processing system for a soccer playing robot team including ball identification, goal identification, geometric features on the field, other robots on the field, the team goal-keeper, distance to various objects, using of edge detection filters, phase and angle conversion filters, color identification and calibration, object movements and more. In addition, the students will build a GUI for robots position identification.
The students will design a localization system for a soccer playing robot team. It includes robot localization under different condition, solving the "kidnapped robot" problem, using image processing system output for localization, ball localization, goals localization, localization of opponent robots, using particle filter and other Bayesian filters, for robust operation, using the goalkeeper as a reference point when available, decision making and states analyzing.

דרישות:

עיבוד ספרתי של תמונות ומערכות משובצות מחשב מומלץ גם למידת מכונה

מקורות:

  1. Gonzalez, R.C. and Woods, R.E., Digital Image Processing, Pearson Prentice-Hall, New Jersy, 2008.
  2. N. Ergin O¨ zkucur and H. Levent Akın, LocalizationWith Non-Unique Landmark Observations, In: Proc. of RoboCup 2010.
  3. Anderson, P., Yusmanthia, Y., Hengst, B., and Sowmya, A., Robot Localization Using Natural Landmarks, Proc. of RoboCup 2012, pp. 118-129, 2012
  4. Sven Behnke, Jürgen Müller, Michael Schreiber: Toni: A Soccer Playing Humanoid Robot. In: RoboCup 2005: 59-704.
  5. Laowattana D., et al.: Team KMUTT, Team description paper. Robocup 2009, Humanoid League (2009)
  6. Lu H., Huang D., Xiong D., Li X., and Zheng Z. A Robust Place Recognition Algorithm Based on Omnidirectional Vision for Mobile Robots. Proc. of RoboCup 2012, pp. 286-297, 2012.