基于决策树的多防御方目标攻防决策

针对多方对抗博弈问题的复杂性和多样性,提出了利用决策树思想对多防御方的目标攻防问题的策略进行决策树分类,再利用决策树分类模型对多防御方的目标攻防问题的攻防策略进行快速决策。首先,对多防御方目标攻防问题利用决策树ID3算法以攻防双方的策略作为类别属性进行分类,得到多防御方目标攻防问题的决策树模型;然后,对多方对抗中的攻击方,基于距离的威胁权重函数得到攻击方的轨迹;最后,得到防御方的协同合作策略,并进行了仿真验证。

Multi-Defender Target Strike and Defense Decision Based on Decision Tree

In view of the complexity and diversity of the multi-agent confrontation game problem, a decision tree classification of the strategy of multi-defender target strike and defense problems is proposed by using the idea of decision tree,and then the decision tree classification model is used to classify the offensive and defensive strategies of target strike and defense problems of multi-defender.Firstly,the decision tree ID3 algorithm is used to classify the multi-defender target strike and defense problem with the strategies of both offensive and defensive parties as the category attribute, and the decision tree model of the multi-defender target strike and defense problem is obtained. For the attacker in the multi-party confrontation, the trajectory of the attacker is obtained according to the distance-based threat weight function. Finally, the cooperative strategy of the defender has been obtained, and simulation verification was carried out.