Towards an intelligent vision system for mobile robots in RoboCup environment

Abstract

One of the main challenges in RoboCup where a team of robots play soccer against another such team, is to maintain a high level of speed and accuracy in decision making and performing actions by the robot players. Although we might be able to use complicated hardware and software on the robots to achieve the desired accuracy, but such systems might not be applicable in real-time RoboCup environment due to their need for high processing time. To reduce the processing time we developed some basic ideas on the robot's front and omni-directional vision systems. These ideas are inspired by a number of features in the human vision system towards enhancing naive vision systems that work intelligently. These ideas included efficient need-based vision, reducing the number of objects to be detected to a few objects of interest in each frame with the minimum needed accuracy, introduction of static and dynamic regions of interest, proposing first, those areas that are most probable to find our objects of interest, the usage of some domain specific knowledge that is used in detecting and tracking a unique safe point on the ball, and also introducing fast and accurate methods for separating the area inside the soccer field and its outside region in order to reduce the search space to only the area inside the soccer field. We have implemented these methods on RoboCup environment and satisfactory results were obtained. In addition, the algorithm that separates the soccer field from outside region was applied on many images obtained from some real soccer fields and showed a very good performance.