Research On Task Planning Method For Aircraft Launch And Recovery Operations On Carrier Deck

Task planning for carrier aircraft launch and recovery operations on carrier deck is to develop a reasonable and feasible action plan for a given flight mission.This action plan is highly relevant to the efficiency of the carrier's mission,the safety and cost of deck operations.The task planning for aircraft launch and recovery operations on carrier deck has the characteristics of diverse tasks,distinct levels and complex constraints,and is difficult to solve.As more and more UAVs will be carried on aircraft carriers,it is inevitable that automated planning methods will play an important role in future carrier deck task planning.The purpose of this paper is to study the problem of task planning aircraft launch and recovery operations on carrier deck.Firstly,this paper takes the “nimitz” class aircraft carrier as the research background to build the task planning model for aircraft launch and recovery on carrier deck in detail.Second,according to the characteristics of the problem,a task planning algorithm for aircraft launch and recovery operations with complex constraints is proposed.In the algorithm,the idea of task decomposition of hierarchical task network planning is used for reference to decompose tasks so as to improve the efficiency of planning.The planning problem is described as a multi-stage decision-making problem.Third,for the three decision-making stages in the algorithm,a reinforcement learning algorithm is designed,then based on the historical action plans learn the strategies of task leadtime prediction,aircraft selection and transition spots selection,so as to guide the planner to better determine the parameters or rules of the decision-making stages and improve the performance of the action plan.Aiming at task planning for aircraft launch and recovery on carrier deck,this paper designed and implemented a automated planning algorithm,and embedded reinforcement learning algorithm in the process of planning to optimize the performance of the action plan.Combining the knowledge-based reasoning method with the data-based learning method,a new idea is explored to improve the performance of the action plan obtained by the automatic planning algorithm.It has certain reference value for the operation and management practice of the aircraft carrier deck.