Robust posture measurement and identification of soccer robots

To achieve robust posture measurement and identification (ID) of a group of fast-moving mobile robots in a robot soccer system in real time, a novel image-processing algorithm is proposed. The algorithm consists of an illumination-adaptive filter, a fast look-up table, a run-length-encoding- based image reconstruction and a robot ID design. Camera calibration and elimination of parallax errors are also considered. Experimental results verify the speed, reliability and accuracy of the proposed algorithm under different illumination conditions and rapid movement of mobile robots.

[1]  Takeo Kanade,et al.  Geometric camera calibration using systems of linear equations , 1988, Proceedings. 1988 IEEE International Conference on Robotics and Automation.

[2]  Reinaldo A. C. Bianchi,et al.  Implementing Computer Vision Algorithms in Hardware: An FPGA/VHDL-Based Vision System for a Mobile Robot , 2001, RoboCup.

[3]  Dana H. Ballard,et al.  Generalizing the Hough transform to detect arbitrary shapes , 1981, Pattern Recognit..

[4]  Peter I. Corke,et al.  A tutorial on visual servo control , 1996, IEEE Trans. Robotics Autom..