| Recent years saw an increase in research interests and kinds of competitions in robotics soccer. It is no doubt that the soccer robot is a new interdisciplinary research area, referring to diverse fields like, robotics, intelligent control, artificial life, etc. Soccer robot system is a typical multi-agent system and multi-robot cooperative autonomous system. Furthermore, it provides .a standard test-bed and theoretic research model of those fields.Robocup small-size league system consists of four components: robot body subsystem, wireless communication subsystem, vision subsystem and expert decision subsystem.According to the factors, opponent ability, game importance, present score for example, we can use various strategies in the robot soccer games. Therefore we induce and ameliorate the existing strategies and set up a "strategy store" for the decision-making system to select. Moreover, we build up the robot-motion model, ball-motion model and robot-pushball model based on our own hardware system and analyse the basic behaviors and actions of the robot body.Path planning problem is a research mainly about how to make out an optimum route for the robot to avoid obstacles and arrive at the goal from the start safely and uncollidedly in a workplace where obstacles exist. Because the game position changes quickly, the planning result made out in the former period may be altered accordingly in the next time since the environment has been changed. So the decision-making algorithm must have better real-time characteristic. In this paper, the predictive control and artificial potential field algorithm is integrated to control the robot movement. Two typical actions, which are obstacle avoiding and robot shooting movements, are analyzed and designed using predictive artificial potential field algorithm. The simulation result on the MiroSot 3vs3 platform shows that this algorithm meets the reality characteristic demand well and is adapted to the dynamic position path planning ultimately.
|
PDF Full Text Request