基于路径预测的机器人足球路径规划  

Robot soccer path planning based on path prediction

作　　者：刘付民[1] 

机构地区：[1]河南理工大学计算机科学与技术学院,河南焦作454000

出　　处：《计算机工程与设计》2012年第1期341-345,共5页Computer Engineering and Design

基　　金：河南省科技计划基金项目(092300410216)

摘　　要：针对常用的机器人路径规划算法过于复杂并且在每个运动周期都计算路径的问题,提出了一种结合路径预测的路径最优算法。充分利用预测结果减少每周期的路径规划时间;用微量调整动态控制机器人左右轮速度,并充分利用折线路径的短距离优势,为避障机器人创建一条最短路径;以基于周期性预测在同个时间轴上的相交作为碰撞信号,来减少每个周期的重复性计算时间。实验结果表明,该方法能大大提高机器人路径规划的速度,降低不同周期上路径规划结果不一致导致的运动震荡。The commonly used intelligent methods are artificial potential field method, visual vertex graph method and genetic method. However, these methods are more complex and need calculation for each movement cycle path in the robot. An optimal path algorithm that combines the path planning is presented, which takes full use of the path planning to reduce the cycle time and increase the stability of the robot movement. It is able to create the shortest path for the obstacle-avoidance robot with micro adjustment to control the speed of two wheels dynamically and make fully use of the advantage of the short distance of the broken line. Based on periodic forecast, the intersection on a timeline as a collision signal can reduce time-computing repeatability of each cycle, thus, the speed of the robot path planning is enhanced greatly and the sports concussion resulted by different path plan- ning results on different cycles is reduced.

关 键 词：路径预测 动态检测 路径规划 机器人 人工智能 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]