Micro Robot Soccer Tournament Research Articles

Control of a soccer robot

The paper describes some results of development of robot soccer team RoBohemia from Brno University of Technology. The robot soccer team belongs to MIROSOT (MIcro RObot SOccer Tournament) robot soccer category. The paper introduces mathematical model and simulation scheme of the robot player simultaneously with some construction details. Both dynamic and kinematic models are considered. The paper describes also a control scheme and problems connected with control of the robot player. Control scheme must guarantee non-oscillatory and fast movement of the robot during a match.

Visual Servo Control Algorithm for Soccer Robots Considering Time-Delay

ABSTRACTAn efficient visual servo control algorithm which uses primary/secondary controller and the Smith predictor based on an accurate model of robot soccer system including time-delay is suggested for micro-robot soccer tournament (MIROSOT). In this paper, we analyze and design control system of soccer robot in the frame of visual servoing. We think that our approach encourages more systematic approaches to soccer robot system. MIROSOT system is a typical direct visual servo system with a fixed camera and mobile robots with internal sensors such as encoders. In real games, encoders often offer wrong position data due to slips and collisions between robots or between a robot and a wall. However, vision provides positions and orientations of robots in certain accuracy regardless of slips or collisions. In spite of this advantage inevitable time-delay arising from vision and communication subsystems seriously causes undesired effects in control of robots. In this paper we present a visual servo controller...

Vision Systems and Software Design for Soccer Playing Mobile Robots

ABSTRACTThis article describes the design of the software framework and the vision system used by the UC Davis MIcro-RObot SOccer Tournament (MIROSOT) team. A fault tolerant and highly modular software framework is described. A software-based vision system capable of extracting location, orientation, and identity of objects is presented and justified. Our robot color scheme is illustrated and compared with another common system.