改进的人工势场法在机器人足球路径规划中的应用

Application of improved artificial potential field method in the path planning of robot soccer

传统的人工势场法在静态的路径规划中有较好的效果,对于足球机器人这类对抗性与实时性比较强的动态环境的路径规划往往存在一定的缺陷。由于传统的人工势场法是单纯的考虑势能与距离间的关系,所以在实际的路径规划中有些不必要的避障也考虑在内,无形中就增加了计算的繁琐度。在改进的人工势场法中我们引进了机器人与障碍物的相对位置与相对角度以及机器人与障碍物的距离跟机器人与目标点的距离的联系,根据障碍物所处的相对位置以及与机器人的相对角度来确定障碍物对机器人的威胁程度,从而来确定斥力的大小。在传统的算法中机器人所受的引力随距离减小而减小,斥力随距离减小而增加。所以当障碍物出现在目标点附近且在机器人威胁区域内时,往往会因为斥力相对引力足够大而使机器人无法到达目标点。所以在考虑斥力时我们有必要把机器人与目标点的距离也考虑在内。最后本文用仿真方法证明该方法的可行性。

Traditional artificial potential field method has a good effects on static path planning, but for soccer robot that has path planning of confrontational and practical dynamic environment, it also has some weekness. As traditional artificial potential field method only consider the relationship between potential energy and distance, in practical path planning ,it considers some unnecessary obstacles and adds calculations. In improved artificial potential field method ,we introduce the relationship between robot and obstacles of relative position and angle, and the relationship between the distance of robot and obstacles and the distance of robot and target. We confirm repulsive force size on the base of the relative position of obstacles and the relative angle of robot to make sure the threat level for robots. In traditional algorithm, the gravity on robot becomes smaller when distance reduces, the repulsion becomes bigger when distance increases. When obstacles is near by target and the threatened area of the robot, robots won＇t get the target owning to the repulsion is relatively bigger than gravity. So we should consider the distance between robot and target when regarding repulsion. In the end ,we prove the viability by the means of simulation.