A method, device, and medium for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery.

Abstract

This application discloses a method, device, and medium for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery, relating to the interdisciplinary field of computer technology and aerospace technology. The method includes: defining a composite reward function in the aircraft recovery process; providing a scalar reward to the decision-making agent at each decision step of reinforcement learning based on the composite reward function; the scalar reward being the reward function used for aircraft recovery reinforcement learning; the reward function guiding the decision-making agent to learn a recovery strategy that balances fuel safety, recovery efficiency, and task priority; embedding the calculation process of the reward function into the reinforcement learning training process to construct an optimal recovery scheduling strategy model to generate an aircraft recovery scheduling sequence; and recovering all aircraft to be recovered based on the aircraft recovery scheduling sequence. This application can comprehensively quantify the performance of multiple optimization objectives such as safety, efficiency, and priority in the aircraft recovery process and achieve intelligent and optimized aircraft recovery decision-making.

Description

Technical Field

[0001] This application relates to the interdisciplinary field of computer technology and aerospace technology, and in particular to a method, device and medium for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery. Background Technology

[0002] Aircraft recovery, especially approach scheduling at busy airports, is an extremely complex dynamic scheduling problem. Dispatchers must make optimal recovery sequence decisions within a very short timeframe, comprehensively considering various dynamically changing factors such as the real-time fuel level, airframe health, mission priority, and recovery lane availability of each aircraft in the recovery group. This process demands not only extremely high decision-making efficiency but also stringent safety requirements that allow no room for error. Traditional aircraft recovery relies primarily on manual scheduling, with dispatchers making judgments based on extensive experience and established rules. However, when facing large-scale, high-intensity recovery tasks, the cognitive load on human personnel is enormous, making it difficult to guarantee decision quality and prone to oversights, posing a potential threat to flight safety. With technological advancements, some decision-making aids based on traditional operations research or expert systems have been proposed, but these methods often have rigid models and are ill-suited to the highly dynamic and uncertain civil aviation environment.

[0003] In recent years, reinforcement learning, as an artificial intelligence technique capable of autonomously learning optimal strategies through interaction with the environment, has provided a new approach to solving complex decision-making problems. In reinforcement learning, the design of the reward function is the core of guiding the agent's learning direction, directly determining the quality of the final strategy. Existing reinforcement learning methods applied to scheduling problems often have relatively simple reward function designs, or struggle to effectively balance multiple conflicting optimization objectives. For example, overemphasizing fuel safety may lead to low recovery efficiency, while a one-sided pursuit of efficiency may ignore the needs of high-priority tasks, resulting in a scheduling strategy that fails to meet the comprehensive requirements of practical applications. Therefore, how to design a reward function that comprehensively and accurately reflects the complex needs of aircraft recovery tasks, and based on this, generate a reliable optimal aircraft recovery scheduling sequence, is a pressing problem in the current technological field. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, and medium for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery, in order to solve the problems of a single reward function and scheduling strategies that cannot meet the comprehensive requirements of practical applications.

[0005] To achieve the above objectives, this application provides the following solution.

[0006] In a first aspect, this application provides a method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery, including the following methods.

[0007] Define a composite reward function for the aircraft recovery process; the composite reward function includes a fuel safety penalty, an efficiency penalty, a priority waiting penalty, and a mission completion reward.

[0008] Based on the composite reward function, a scalar reward is provided to the decision agent at each decision step of reinforcement learning; the decision agent is used to perform decision actions in an environment simulating the dynamic changes of the aircraft convoy to be recovered; the scalar reward is a reward function used for aircraft recovery reinforcement learning; the reward function is used to guide the decision agent to learn a recovery strategy that takes into account fuel safety, recovery efficiency, and task priority.

[0009] The calculation process of the reward function is embedded into the reinforcement learning training process to construct an optimal retrieval scheduling strategy model.

[0010] An aircraft recovery scheduling sequence is generated based on the optimal recovery scheduling strategy model.

[0011] All aircraft awaiting recovery will be recovered based on the aforementioned aircraft recovery scheduling sequence.

[0012] Secondly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for generating aircraft recovery scheduling sequences for reinforcement learning in aircraft recovery.

[0013] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery.

[0014] According to the specific embodiments provided in this application, this application has the following technical effects.

[0015] This application defines a composite reward function to provide scalar rewards to the reinforcement learning agent at each decision step. This composite reward function consists of a fuel safety penalty, an efficiency penalty, a priority waiting penalty, and a task completion reward, which can comprehensively quantify the performance of multiple optimization objectives such as safety, efficiency, and priority during aircraft recovery. Specifically, the fuel safety penalty penalizes low fuel conditions, the efficiency penalty encourages shortening the total recovery time, and the priority waiting penalty ensures that high-priority or faulty aircraft are processed first, effectively balancing multiple conflicting optimization objectives.

[0016] Furthermore, this application obtains the optimal recovery strategy model by training the decision-making agent in a simulated environment, and then uses this model to make sequential decisions in actual tasks, generating a comprehensive optimal aircraft recovery scheduling sequence that takes into account multiple objectives. This makes the scheduling strategy unable to meet the comprehensive requirements of practical applications, solves the limitations of traditional scheduling methods in handling complex dynamic constraints, and realizes the intelligentization and optimization of aircraft recovery decision-making. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a method for generating aircraft recovery scheduling sequences using reinforcement learning, provided as an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] To make the objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] like Figure 1 As shown in the figure, this application provides a method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery, including the following steps.

[0022] S1: Define a composite reward function for the aircraft recovery process; the composite reward function includes a fuel safety penalty, an efficiency penalty, a priority waiting penalty, and a mission completion reward. The fuel safety penalty... Efficiency penalty items and priority waiting penalty items All values ​​are non-positive; the task completion reward items... It is a non-negative value.

[0023] S2: Based on the composite reward function, a scalar reward is provided to the decision agent in each decision step of reinforcement learning; the decision agent is used to perform decision actions in an environment simulating the dynamic changes of the aircraft convoy to be recovered; the scalar reward is a reward function used for aircraft recovery reinforcement learning; the reward function is used to guide the decision agent to learn a recovery strategy that takes into account fuel safety, recovery efficiency and task priority.

[0024] Scalar reward The calculation formula is as follows.

[0025]

[0026] S3: Embed the calculation process of the reward function into the reinforcement learning training process to construct an optimal retrieval scheduling strategy model.

[0027] S4: Generate an aircraft recovery scheduling sequence based on the optimal recovery scheduling strategy model.

[0028] S5: Recover all aircraft to be recovered based on the aircraft recovery scheduling sequence.

[0029] In one exemplary embodiment, the following steps are included before S1.

[0030] The calculation of fuel safety penalties includes the following steps.

[0031] S11: Obtain the queue of all aircraft that have not yet been successfully recovered in the decision step.

[0032] S12: Obtain the current fuel quantity of each aircraft to be recovered in the aircraft to be recovered queue at the decision step.

[0033] S13: Calculate the fuel safety penalty for a single aircraft to be recovered based on the current fuel quantity and the minimum fuel safety threshold.

[0034] S14: Determine the fuel safety penalty item based on the fuel safety penalties of all aircraft to be recovered.

[0035] The fuel safety penalty program establishes a safety baseline by monitoring the amount of fuel in all unrecovered aircraft, imposing a penalty proportional to the amount of fuel loss for situations falling below a safety threshold, and applying strong negative feedback for extreme situations where fuel is completely exhausted.

[0036] In one exemplary embodiment, S13 specifically includes the following steps.

[0037] like ,Sure =0; where, For the first i The current fuel level of the aircraft awaiting recovery; The minimum fuel safety threshold; Fuel safety penalty for a single aircraft awaiting recovery.

[0038] like ,Sure A huge penalty is imposed for running out of fuel. The negative value.

[0039] like ,Sure Equals fuel loss amount and fuel safety penalty coefficient The negative value of the product; all fuel deficits are at the minimum fuel safety threshold. With the i Current fuel level of the aircraft awaiting recovery The difference.

[0040] Fuel depletion incurs huge penalties The value range is [500, 2000].

[0041] Minimum fuel safety threshold It is [1.1, 1.5] times the total amount of fuel required to complete one standard recovery approach and one go-around procedure, based on the aircraft model.

[0042] In one exemplary embodiment, the following steps are included before S1.

[0043] Calculate the efficiency penalty term; the efficiency penalty term as follows.

[0044]

[0045] in, To start from the previous decision step End of current decision-making step The length of time elapsed to complete the process. This is the preset efficiency penalty coefficient.

[0046] The efficiency penalty term is directly related to the time consumption between decision steps, and applies continuous negative feedback to the passage of time to incentivize the agent to complete all recycling tasks as quickly as possible.

[0047] In one exemplary embodiment, the following steps are included before S1.

[0048] Calculate the priority waiting penalty term; the priority waiting penalty term as follows.

[0049]

[0050] in, For the queue of aircraft awaiting recovery Aircraft index in the database; For airplane The relative completeness; For airplane The relative task priority; The preset priority waiting penalty coefficient.

[0051] Fuel safety penalty coefficient Efficiency penalty coefficient and priority waiting penalty coefficient The value range is [0.1, 5.0].

[0052] In practical applications, relative completeness and relative task priority It is based on the aircraft to be recovered The original fault level and mission level data are normalized in the [0,1] interval to obtain dimensionless values, which are used to quantify the scheduling cost of keeping aircraft with higher fault levels or higher mission levels waiting.

[0053] The priority waiting penalty takes into account the airframe integrity and mission importance of each aircraft to be recovered, and imposes penalties on behaviors that cause high-priority aircraft to remain in a waiting state, ensuring that critical missions and high-risk aircraft are given priority processing.

[0054] In one exemplary embodiment, the following steps are included before S1.

[0055] Calculation task completion reward items Specifically, it includes the following steps.

[0056] Determine the queue of aircraft to be recovered Is it an empty set? If yes, determine that all aircraft to be recovered have been successfully recovered, and use a fixed positive mission completion reward value as the mission completion reward item; if no, determine that the mission completion reward item is 0.

[0057] Task completion reward value The value range is [100, 500].

[0058] The task completion reward is a significant positive reward given at the final moment when all aircraft have been successfully recovered, providing a clear and ultimate convergence goal for the agent's learning process.

[0059] In practical applications, the preset parameters involved in this application have the following value ranges.

[0060] In one exemplary embodiment, S3 specifically includes the following steps.

[0061] S31: Embed the calculation process of the reward function into the reinforcement learning training process.

[0062] S32: In the reinforcement learning training process, the internal decision parameters of the decision agent are updated according to the scalar reward, optimized policy function, or value function obtained after the decision agent performs each decision action, until the decision agent's recovery strategy converges, thus constructing an optimal recovery scheduling strategy model. The decision action is to select an aircraft from the queue of aircraft yet to be recovered for recovery.

[0063] In one exemplary embodiment, S4 specifically includes the following steps.

[0064] S41: Input the initial state of the queue of aircraft to be recovered into the optimal recovery scheduling strategy model, and execute the optimal recovery scheduling strategy model sequentially, setting the decision of one aircraft to be recovered to be output at each decision step, until all aircraft to be recovered have been scheduled.

[0065] S42: Arrange the aircraft to be recovered from all decision steps in the order of their corresponding decisions to generate an aircraft recovery scheduling sequence that achieves the best overall balance between fuel safety, recovery efficiency, and mission priority.

[0066] In practical applications, this application utilizes the aforementioned reward function design method to train a decision-making agent in a high-fidelity simulation environment. Through continuous trial and error, the agent receives the composite reward function designed in this application after each decision and adjusts its internal decision network parameters accordingly, ultimately learning an optimal recovery strategy model that maximizes long-term cumulative rewards. Subsequently, in practical applications, the initial state of the aircraft queue to be recovered is input into the trained optimal recovery strategy model. The model then makes decisions sequentially and step-by-step based on the current state, selecting the aircraft most deserving of recovery at each decision. Arranging all the decisions made by the model in chronological order constitutes a comprehensively optimal aircraft recovery scheduling sequence, achieving automation and intelligence in the entire scheduling process.

[0067] This application, through the design of a refined multi-objective reward function, successfully transforms complex scheduling constraints into numerical signals that the decision-making agent can understand and optimize. The resulting scheduling strategy achieves a good balance between safety, efficiency, and task priority, and is significantly better than traditional methods.

[0068] Taking a typical reinforcement learning framework as an example, the technical solution of this application will be further explained.

[0069] In a typical reinforcement learning framework, the decision-making agent interacts with the environment at each decision step, and the environment provides a reward signal, i.e., a reward function, based on the agent's actions. The core of this application lies in the design of this reward signal.

[0070] At each decision step, this application calculates a composite scalar reward value, i.e., a composite reward function, which is the sum of four components.

[0071] The first component is the fuel safety penalty. To calculate it, the system first identifies the set of all aircraft currently in the air awaiting recovery, i.e., the aircraft queue. For each aircraft in this set, the system obtains its current fuel level and compares it to a preset minimum fuel safety threshold. This threshold is typically set based on the aircraft type and standard approach go-around fuel consumption, with a certain safety margin. If the aircraft has sufficient fuel above this threshold, no penalty is incurred. If the fuel is exhausted, a large, fixed penalty is applied to strongly prevent the agent from learning a strategy that leads to such a catastrophic consequence. If the fuel level is between the minimum safety threshold and zero, the penalty is proportional to the amount of fuel deficit; the greater the deficit, the larger the penalty. Finally, the fuel penalties for all aircraft awaiting recovery are summed to obtain the total fuel safety penalty for this decision step.

[0072] The second component is the efficiency penalty term. Its purpose is to encourage the entire recycling process to be completed as quickly as possible. Its calculation is very straightforward: a preset efficiency penalty coefficient is multiplied by the time elapsed from the previous decision step to the current decision step, and the result is negative. This means that for every unit of time elapsed, the agent is subject to a fixed penalty, thus incentivizing it to take actions that shorten the total time consumption.

[0073] The third component is the priority waiting penalty. This component aims to ensure that aircraft in dangerous situations or carrying out important missions are prioritized for recovery. For each unrecovered aircraft, this application first obtains its two key attributes: relative integrity and relative mission priority. These two attributes are typically obtained from the aircraft's original fault level and mission level data, through normalization, with values ​​between zero and one; a higher value indicates a worse aircraft condition or a more important mission. The penalty is calculated by summing the relative integrity and relative mission priority of all aircraft awaiting recovery, then multiplying by a preset priority waiting penalty coefficient and taking a negative value. Thus, as long as high-priority or high-risk aircraft are waiting in the queue, a large penalty will continuously be generated, forcing the agent to prioritize their processing.

[0074] The fourth component is the task completion reward. This is a sparse but crucial positive reward. At each decision step, the system checks if the set of aircraft to be recovered is empty. In the vast majority of cases, the set is not empty, and the reward is zero. Only when the last aircraft is successfully recovered and the set becomes empty will the system award a fixed, relatively large positive reward value. This final reward provides the agent with a clear endpoint and goal for learning.

[0075] This application provides a method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery, which is a direct application of the above-mentioned reward function design.

[0076] The aircraft recovery scheduling sequence generation process comprises two stages: training and generation. In the training stage, the calculation of the aforementioned reward function is embedded into a reinforcement learning training process. A decision-making agent, such as a deep neural network, is trained in a virtual environment that simulates aircraft flight, fuel consumption, and the recovery process. The agent's task at each decision moment is to select one aircraft from all available aircraft for recovery. After making a selection, the environment updates its state based on the action and calculates a composite reward using this method, feeding it back to the agent. Based on this reward signal, the agent updates its network parameters using algorithms such as policy gradient or value iteration, aiming to learn a policy that maximizes the total reward of the entire recovery task. This training process is repeated thousands of times until the agent's policy converges and stabilizes, forming an optimal recovery scheduling policy model.

[0077] During the generation phase, when faced with a real recovery task, the initial state information of the aircraft queue to be recovered is simply input into this pre-trained optimal recovery scheduling strategy model. The model immediately outputs its first decision: the currently optimal aircraft to recover. After the first aircraft is scheduled, the system state is updated, and the model then makes its second decision, and so on, until all aircraft have been scheduled. Arranging these sequentially output decisions forms a complete, balanced optimal scheduling sequence among multiple objectives. In this way, this application transforms the complex manual scheduling task into an automated, data-driven intelligent decision-making process.

[0078] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments. The computer device can be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device provides computational and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device stores data to be processed. The I / O interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with an external terminal via a network connection. When the computer program is executed by the processor, it implements a method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery.

[0079] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0080] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0081] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0082] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by hardware related to computer program instructions. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0083] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0084] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0085] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery, characterized in that, include: Calculate fuel safety penalties Specifically, it includes: Obtain the queue of all aircraft that have not yet been successfully recovered in the decision step; Obtain the current fuel level of each aircraft in the aircraft to be recovered queue at the decision step; The fuel safety penalty for a single aircraft awaiting recovery is calculated based on the current fuel level and the minimum fuel safety threshold, specifically including: like ,Sure =0; where, For the first i The current fuel level of the aircraft awaiting recovery; The minimum fuel safety threshold; For a single aircraft to be recovered i Fuel safety penalties; like ,Sure A huge penalty is imposed for running out of fuel. The negative value; like ,Sure Equals fuel loss amount and fuel safety penalty coefficient The negative value of the product; all fuel deficits are at the minimum fuel safety threshold. With the i Current fuel level of the aircraft awaiting recovery The difference; The fuel safety penalty items are determined based on the fuel safety penalties of all aircraft to be recovered; Calculate the efficiency penalty term; the efficiency penalty term for: ； in, To start from the previous decision step End of current decision-making step The length of time elapsed to complete the process. This is a preset efficiency penalty coefficient; Calculate the priority waiting penalty term; the priority waiting penalty term for: ； in, For the queue of aircraft awaiting recovery Aircraft index in the database; For airplane The relative completeness; For airplane The relative task priority; The pre-defined priority waiting penalty coefficient; Calculation task completion reward items Specifically, it includes: Determine the queue of aircraft to be recovered Is it an empty set? If so, and it is confirmed that all aircraft to be recovered have been successfully recovered, the positive mission completion reward value will be used as the mission completion reward item; If not, the task completion reward item is set to 0; Define a composite reward function for the aircraft recovery process; the composite reward function includes a fuel safety penalty, an efficiency penalty, a priority waiting penalty, and a mission completion reward. Based on the composite reward function, a scalar reward is provided to the decision agent at each decision step of reinforcement learning; the decision agent is used to perform decision actions in an environment simulating the dynamic changes of the aircraft convoy to be recovered; the scalar reward is the reward function used for aircraft recovery reinforcement learning; the reward function is used to guide the decision agent to learn a recovery strategy that takes into account fuel safety, recovery efficiency, and task priority; scalar reward The calculation formula is ; The calculation process of the reward function is embedded into the reinforcement learning training process to construct an optimal retrieval scheduling strategy model; Generate an aircraft recovery scheduling sequence based on the optimal recovery scheduling strategy model; All aircraft awaiting recovery will be recovered based on the aforementioned aircraft recovery scheduling sequence.

2. The method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery according to claim 1, characterized in that, The calculation process of the reward function is embedded into the reinforcement learning training process to construct an optimal retrieval scheduling strategy model, specifically including: The calculation process of the reward function is embedded into the reinforcement learning training process; In the reinforcement learning training process, the internal decision parameters of the decision agent are updated according to the scalar reward, optimization policy function or value function obtained after the decision agent performs each decision action, until the recycling policy of the decision agent converges, and the optimal recycling scheduling policy model is constructed.

3. The method for generating aircraft recovery scheduling sequences using reinforcement learning for aircraft recovery according to claim 1, characterized in that, The aircraft recovery scheduling sequence is generated based on the optimal recovery scheduling strategy model, specifically including: The initial state of the queue of aircraft to be recovered is input into the optimal recovery scheduling strategy model, and the optimal recovery scheduling strategy model is executed sequentially. It is set that a decision of one aircraft to be recovered is output at each decision step until all aircraft to be recovered are scheduled. The aircraft to be recovered, output from all decision steps, are arranged in the order of their corresponding decisions to generate an aircraft recovery scheduling sequence that achieves the best overall balance between fuel safety, recovery efficiency, and mission priority.

4. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the aircraft recovery scheduling sequence generation method for reinforcement learning of aircraft recovery as described in any one of claims 1-3.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the aircraft recovery scheduling sequence generation method for reinforcement learning of aircraft recovery as described in any one of claims 1-3.

Images