机器人足球进攻策略的仿真研究  

Simulation Research on the Offensive Strategy of Robot Soccer

作　　者：李文彬[1] 史豪斌[1] 李伟华[1] 齐勇[1] 

机构地区：[1]西北工业大学计算机学院,陕西西安710129

出　　处：《计算机仿真》2011年第2期187-191,共5页Computer Simulation

基　　金：国家部委基金项目(51315080404);航空科技创新基金项目(08E53003)

摘　　要：研究解决高速比赛中机器人由于不能及时准确预测对方防守而使己方进攻缺乏目的性与针对性的问题,为准确优化设计进攻目标策略,提出了一种随机森林预测模型的进攻路径选择算法。算法利用防守适应值分析模型评估对方机器人的威胁性,针对对方防守弱点选择有效的进攻路径,采用切线法控制机器人击球,使球的运动与预先设计的进攻路径匹配,进行了仿真实验,结果表明提高了策略的实用性和可行性,同时系统策略也增加了机器人的自主性和智能性,使球队的整体能力得到了改善。During high-speed matches,the offense of robot scorers is easy to be deficient in purposiveness and pertinence as it is hard to make timely and accurate predictions about hostile defense.In order to solve the problem,a selection algorithm of offensive path based on random forests prediction model was proposed in this paper.The algorithm evaluated the threat of hostile robots via the analysis model of defense fitness value,and selected an effective offensive path according to the hostile defensive weakness.Then tangent control method was implemented to control the action that robot kicked the ball,so that the movement of ball was matched with the offensive path designed in advance and the practicality of the strategy was enhanced.Simulation experimental results showed that the autonomy and intelligence of robots were strengthened and the overall ability of team was improved with the proposed offensive strategy of robot soccer.

关 键 词：机器人足球 随机森林 进攻策略 防守适应值 切线控制 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]