Abstract: This paper makes FIRA SimuroSot 5v5 game simulation platform as the background, for the problem that how to maximize the threat of attack in strategy, proposes a selection method of the best offensive player and offensive point based on prediction models. It builds the robots and the ball motion models, according to the established prediction models to determine the effective attack time, which can calculate the offensive players and offensive points. It carries out a detailed threat of attack-degree evaluation to each offensive player and offensive point and selects the best offensive player and the best offensive point. Simulation tests verify the validity of the method.

Key words: prediction model, effective attack time, threat of attack-degree, best offensive player, best offensive point

摘要： 以FIRA SimuroSot5v5仿真平台为背景，针对策略中如何最大化进攻威胁的问题，提出一种基于预测模型的最佳进攻球员和进攻点选择方法。对机器人和球的运动规律进行建模，根据建立的预测模型确定机器人的有效进攻时间，计算可进攻球员和可进攻点，并对每个可进攻球员及进攻点进行进攻威胁度评估，得到最佳进攻球员和最佳进攻点。仿真实验验证了该方法的有效性。

关键词: 预测模型, 有效进攻时间, 进攻威胁度, 最佳进攻球员, 最佳进攻点

CLC Number: 

• TP18