Shipboard aircraft guarantee operator scheduling method for emergency situations

Abstract

The invention discloses a shipboard aircraft support personnel scheduling method, which is used for coping with uncertain emergencies of deck shipboard aircraft support operation scheduling. The method comprises the following steps: firstly, constructing a support process of support personnel on a shipboard aircraft into a Markov decision process; an improved Soft Actor Critic (SAC) scheduling algorithm is then designed according to the process characteristics: (1) in order to reduce the learning difficulty, the SAC algorithm is expanded into a multi-agent algorithm, environmental data processing is added, and environmental state information needing to be processed by agents is reduced; (2) in order to avoid action conflicts, designing an adaptive rate to increase scheduling quality; and (3) in order to optimize the overall training process, setting an invalid action shielding mechanism and a priority experience playback mechanism. And finally, the designed algorithm is put into training, and the trained agent can be put into scheduling. According to the method, the emergency situation of the deck can be well handled, so that deck scheduling has higher robustness for uncertainty, and the efficiency of deck scheduling is improved.

Description

(1) Technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to a method for dispatching carrier-based aircraft support operators for emergencies. (2) Background technology [0002] As the core of the ocean-going operations of the navy of a major country, the aircraft carrier is also a symbol of comprehensive national strength and naval strength, and has unparalleled military value. The index to measure the combat capability of an aircraft carrier is usually defined as the sortie rate of carrier-based aircraft, and the sortie rate is related to many factors. After the aircraft carrier is completed, its hardware configuration is basically fixed. The main way to improve the combat capability of the aircraft carrier is to improve resource utilization efficiency. An important part of reflecting the efficiency of various aspects in the sortie recovery rate of the carrier-based aircraft lies in the various support t...

AI Technical Summary

Problems solved by technology

An important part of reflecting the efficiency of various aspects in the sortie recovery rate of the carrier-based aircraft lies in the various support tasks performed by the carrier-based aircraft on the deck. At the same time, in fact, most of the support operations on the current aircraft carrier are carried out by support personnel. , so the scheduling of security personnel can have a direct impact on the improvement of security work efficiency

[0003] Although the traditional intelligent optimization algorithm can obtain a relatively good scheduling strategy, it mostly optimizes the overall shift schedule, and it is difficult to deal with emergencies on the deck

The thesis "Optimization of Operation Scheduling and Resource Allocation Integration of Shipboard Aircraft Deck Maintenance Support" builds an integrated optimization model of carrier-based aircraft deck maintenance support operation scheduling and resource allocation, which is solved by NSGA II, but does not consider emergency situations in support operations; The paper "The Joint Optimization Method of Personnel Configuration-Scheduling for Carrier-based Aircraft Group Dispatch Support Based on Marginal-Artificial Bee Colony Algorithm" optimizes the problem of personnel configuration and scheduling at the same time. The outer layer uses the marginal optimization algorithm to solve the problem of staffing The worker bee colony algorithm solves the scheduling problem, but it cannot cope with the impact of deck emergencies. It considers real-time scheduling, and can also respond immediately to emergencies on the deck, but it does not deal with execution conflicts accordingly, and only makes up for possible conflicts of actions through frequent decision-making, which will lengthen the decision-making time

Images