Air-to-surface Autonomous Attack Decision-making Of UCAV Based On FCM

In order to adapt to the increasingly complex battlefield environment,enhance unmanned combat aerial vehicles'(UCAV)combat capability under supervision and control mode,the problem of UCAV's autonomous attack decision-making is put forward.Considering that(1)in the depth front of battlefield,UCAV grasps the latest information;(2)the communication between UCAV and control station is limited;(3)other practical problems such as the operator's heavy burden,improving UCAV's ability of autonomous attack decision-making becomes the key to enhance the combat effectiveness.Using the refined heuristic knowledge to generate the favorable action plan autonomously,then the operator confirms the finally instruction,which is a feasible way.The main contributions are as follows:1)A level of automation(LOA)scheme for unmanned aerial vehicles(UAV)is proposed,which consists of the name of levels,the interactive mode and interactive content between man and machine.Based on the principles for LOA establishing from foreign references,combined with the actual conditions,define the LOA of UCAV in this thesis.2)The UCAV's autonomous decision making framework with three layers is built,which includes situation awareness layer,decision inference layer and decision interactive layer.Three layers are respectively responsible for the completion of situation awareness and information processing,the organization and implementation of decision reasoning process,the feedback of the decision results and the response of the operator's control instructions.Hierarchical UCAV's autonomous attack decision-making model provides a framework for the construction of concrete models.3)Designed a hierarchical limited intervention mechanism of manned aerial vehicles(MAV),which includes four kinds of intervention from strong to weak.This kind of intervention mechanism reflects the hierarchy of decision needs and the limitation of operator intervention.It not only makes full use of UCAV's autonomous decision-making capability,but also reduces the working pressure of the commanders.4)Based on fuzzy cognitive map(FCM)theory and combined with task enactment,established hierarchical autonomous attack decision-making model framework of UCAV based on agent-based fuzzy cognitive map(ABFCM).The framework decomposes the contents of UCAV's autonomous attack decision-making into agents of different levels and effects.On the basis of respectively constructing causal inference model of every agent,this paper organizes the whole decision inference behavior legitimately and orderly.5)Designed simulation interface of UCAV's autonomous attack decision-making.By simulation analysis,the rationality and applicability of the established model are verified.Meanwhile,it is shown that the model can help UCAV adapt to increasingly complex battlefield environment,enhance the independent combat capability under supervisory control mode.