A modeling method and system for a hydrogen-powered drone swarm refueling scheme

Abstract

This invention discloses a modeling method and system for a hydrogen-powered drone swarm refueling scheme. The method involves: decomposing the drone's flight process into multiple independent events; managing these independent events using a three-alignment principle; calculating the energy consumption of the drones during flight based on a nonlinear energy consumption model; rasterizing the continuous flight space; planning obstacle avoidance for the mission drones' flight paths; and using stacked genetic programming to cyclically insert random points as refueling rendezvous points into the mission drones' flight paths, assigning them to the corresponding refueling drone paths to form a refueling scheduling scheme. This invention can effectively model complex constraints such as time, space, and energy in a scenario, providing a detailed model and planning framework for collaborative refueling tasks of drone swarms. This reduces the difficulty of subsequent design and optimization algorithms, improves solution efficiency and feasibility, and is expected to enhance the accuracy of drone planning.

Description

Technical Field

[0001] This invention belongs to the field of drone swarm optimization scheduling technology, specifically involving a modeling method and system for a hydrogen-powered drone swarm refueling scheme. Background Technology

[0002] With the development of drone technology, hydrogen-powered drones have upgraded their power systems to hydrogen fuel cells, building upon conventional drones. Leveraging the high power density and environmental adaptability of hydrogen fuel cells, drones have seen significant improvements in endurance, takeoff weight, and cruising speed, enabling them to operate in harsher environments and becoming ideal platforms for emergency rescue, long-duration reconnaissance, and inspection missions. During missions, hydrogen-powered drones utilize hydrogen fuel cell systems to achieve high endurance and mission efficiency. However, despite the significant improvement in endurance, energy consumption remains a limiting factor in wide-area patrols or long-duration missions. Traditional refueling methods primarily rely on drones traveling to and from ground bases for refueling, increasing energy consumption and mission duration, especially leading to frequent and inefficient trips, particularly during long-range missions. Therefore, achieving in-flight refueling for drones during missions has become a pressing technical challenge.

[0003] Currently, existing refueling solutions mostly focus on wireless charging for electric drones. However, unlike battery-based drone refueling planning models, the energy consumption curve of hydrogen-powered drones exhibits complex, non-linear characteristics. This makes existing battery-powered drone planning models unsuitable for direct application to hydrogen-powered drones. Therefore, there is currently a lack of suitable research and effective planning models to address the refueling needs of hydrogen-powered drones. Modeling in-flight refueling solutions for hydrogen-powered drones involves complex issues such as the coordinated scheduling of multiple mission drones and refueling drones, and refueling path planning, presenting the following technical challenges: Complex scenarios involving multiple objects: In-flight refueling tasks involve multiple refueling drones and mission drones, while also considering multiple task points, obstacle areas, and other factors, resulting in high complexity. Non-linear energy consumption: The energy consumption rate of hydrogen-powered drones is affected by changes in the weight of the hydrogen cylinder, resulting in a non-linear overall energy consumption curve. Difficulty in time and space alignment: During the in-flight refueling phase, the refueling drone and the mission drone need to be precisely aligned in time and space to successfully complete the refueling. However, due to the differences in flight speed, path planning, and arrival time between the two types of UAVs, modeling the temporal and spatial synchronization is very difficult, increasing the scheduling complexity of the refueling UAV and the mission UAV. The optimization objectives are difficult to determine: refueling schemes involve multiple optimization objectives, including minimizing range, minimizing energy consumption, shortening time, and reducing the number of refueling attempts, requiring comprehensive consideration of various requirements to formulate the optimal solution. The complexity of solution encoding: Designing specific in-flight refueling schemes requires effective encoding of the paths and refueling nodes of both the mission UAV and the refueling UAV, which is difficult to describe and implement. Therefore, in summary, current modeling of in-flight refueling for hydrogen-powered UAVs suffers from relatively complex model planning and multiple constraints, failing to provide relatively accurate and reliable mathematical model support for subsequent optimization algorithms. Summary of the Invention

[0004] This invention provides a modeling method and system for hydrogen-powered drone swarm refueling schemes. The purpose is to solve the problem that the current modeling process for hydrogen-powered drone in-flight refueling has relatively complex model planning and multiple constraints, which cannot provide relatively accurate and reliable mathematical model support for subsequent optimization algorithms.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] This invention provides a modeling method for a hydrogen-powered drone swarm refueling scheme, comprising the following steps:

[0007] S1. Based on the event-driven model, the flight process of the UAV is decomposed into several independent events. Each independent event includes the start position, end position, start time, end time, start payload, end payload, and energy consumption information.

[0008] S2. The three-alignment principle is adopted to manage several independent events of the mission drone and the refueling drone to align the time, position and energy of each independent event; and the energy consumption of the drone during flight is dynamically calculated based on the remaining fuel of the drone according to the nonlinear energy consumption model.

[0009] S3. The continuous flight space is rasterized, the flight path is discretized into several path grid points, obstacle avoidance planning is performed on the flight path of the mission UAV, and the flight path of the mission UAV is formed.

[0010] S4. Using stack genetic programming, random points are cyclically inserted into the flight path of the mission drone as refueling rendezvous points and assigned to the corresponding refueling drone paths to form a refueling scheduling scheme for hydrogen energy drone clusters.

[0011] S5. For the energy replenishment scheduling scheme of hydrogen-powered drone swarm, calculate the overall energy consumption ratio and energy depletion time ratio of the mission drone and the replenishment drone, and use the overall energy consumption ratio and energy depletion time ratio as the optimization objectives of the model.

[0012] In some implementations, in S1, each independent event represents an action at a specific stage during flight. Several independent events include takeoff events, flight events, refueling events, mission events, waiting events, and landing events.

[0013] In some implementations, in S1, the energy consumption of the takeoff event is a fixed value; in the flight event, the energy consumption of the UAV per unit distance is inversely proportional to its total weight; in the refueling event, the energy of the mission UAV decreases linearly over time, and the refueling UAV will consume additional energy supplied to the mission UAV; the energy consumption in the mission event depends on the mission type; and the energy consumption in the landing event is set to a fixed value.

[0014] In some implementations, in S1, in the event-driven model, a unified energy consumption model is used to calculate the energy consumption for independent events of the same type.

[0015] In some implementations, in S2, the three alignment principles include time alignment, position alignment, and energy alignment.

[0016] In some implementations, in S2, the nonlinear energy consumption model establishes an energy consumption function based on the impact of the UAV's remaining fuel on energy consumption.

[0017] In some implementations, in S3, several path grid points are numbered sequentially, and obstacle avoidance planning is performed on the mission UAV path to obtain the mission UAV path in the baseline solution. The path of the refueling UAV in the baseline solution consists only of the coordinate numbers of its take-off and landing airports.

[0018] In some implementations, in S4, stack genetic programming is used to sequentially insert refueling rendezvous points into the departure and return paths of the mission drones and match refueling drones until all mission drone paths are filled.

[0019] Furthermore, S4 specifically includes:

[0020] S51. Randomly select a path point in the energy replenishment area as the first energy replenishment rendezvous point;

[0021] S52. Randomly select a mission drone and a refueling drone for matching;

[0022] S53. Insert the refueling rendezvous point between any two adjacent path points in the departure or return path of the mission UAV.

[0023] S54. Insert the refueling rendezvous point between any two adjacent path points in the flight path of the refueling drone.

[0024] This invention also provides a refueling system for a hydrogen-powered drone swarm, the system comprising an event-driven model, an event management module, a path planning module, and a refueling scheduling module, wherein:

[0025] The event decomposition module is used to decompose the flight process of the UAV into several independent events, including information on the start position, end position, time, payload, and energy consumption.

[0026] The event management module is used to manage independent events using a three-alignment principle and dynamically calculate energy consumption based on a nonlinear energy consumption model.

[0027] The path planning module is used to perform gridding processing on the continuous flight space, discretize the flight path into several path grid points, perform obstacle avoidance planning on the flight path of the mission UAV, and form the flight path of the mission UAV.

[0028] The energy replenishment scheduling module is used to use stack genetic programming to cyclically insert random points as energy replenishment rendezvous points in the flight path of the mission drone and assign them to the corresponding energy replenishment drone paths to form an energy replenishment scheduling scheme for the hydrogen energy drone cluster; calculate the overall energy consumption ratio and energy depletion time ratio of the mission drone and the energy replenishment drone, and use the overall energy consumption ratio and energy depletion time ratio as the optimization objectives of the model.

[0029] Compared with existing technologies, the modeling method for a hydrogen-powered drone swarm refueling scheme of the present invention has the following beneficial effects:

[0030] This invention establishes an event-chain-based refueling modeling approach, providing effective model support for subsequent scheduling optimization. This enables accurate analysis of path planning and refueling scheduling for both mission-oriented and refueling drones, thus facilitating the continuous and efficient task execution of drone swarms. Compared to current modeling methods that directly encode the path of each drone, this invention more effectively ensures the temporal, spatial, and energy alignment of the swarm refueling scheme. Through event-driven model construction and the three-alignment principle, this invention integrates refueling rendezvous point insertion, energy consumption nonlinear models, and path obstacle avoidance models. Furthermore, it introduces stacked genetic programming into the solution encoding process, systematically modeling key aspects such as the flight process, energy consumption, and refueling rendezvous points of both mission-oriented and refueling drones. This allows for precise description of the dynamic behavior and energy consumption changes of both drones during in-flight refueling. This ensures the accuracy and operability of the modeling process, providing a reliable foundation for path optimization and task scheduling algorithms in complex mission scenarios, and significantly improving the flexibility and scalability of the modeling process. The method of this invention provides a detailed model and planning framework for the collaborative refueling mission of UAV swarms, which greatly reduces the difficulty of subsequent design and optimization algorithms, improves solution efficiency and feasibility, and is expected to improve the accuracy of UAV planning and realize efficient and reliable aerial refueling missions for hydrogen-powered UAVs. Attached Figure Description

[0031] The accompanying drawings are provided to further understand the invention and constitute a part of this invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0032] Figure 1 This is a flowchart illustrating a modeling method for a hydrogen-powered drone swarm refueling scheme according to the present invention.

[0033] Figure 2 This is a schematic diagram of the energy replenishment problem in the modeling method of a hydrogen-powered drone swarm energy replenishment scheme of the present invention;

[0034] Figure 3 This is a schematic diagram illustrating the need for time and space alignment in the modeling method of a hydrogen-powered drone swarm refueling scheme according to the present invention.

[0035] Figure 4 This is a schematic diagram of the event-event chain-based modeling method in the modeling method of a hydrogen energy drone swarm refueling scheme of the present invention;

[0036] Figure 5 This is a schematic diagram of the timeline of the mission drone 2 and the replenishment drones 1 and 4 in an embodiment of the modeling method for a hydrogen-powered drone swarm refueling scheme of the present invention.

[0037] Figure 6 This is a schematic diagram of the baseline solution in the modeling method of a hydrogen-powered drone swarm refueling scheme of the present invention;

[0038] Figure 7 This is a schematic diagram of inserting a refueling rendezvous point into the baseline solution in an embodiment of the modeling method for a hydrogen-powered drone swarm refueling scheme of the present invention;

[0039] Figure 8 This is a schematic diagram comparing the modeling method of the hydrogen energy drone swarm refueling scheme of the present invention with direct coding path modeling in scenario 1.

[0040] Figure 9 This is a comparative diagram of scenario 2 and direct coding path modeling in the modeling method of the hydrogen energy drone swarm refueling scheme of the present invention.

[0041] Figure 10 This diagram illustrates the comparison of the number of successful refueling rendezvous between two encoding methods in a comparative experiment of the modeling method for a hydrogen-powered drone swarm refueling scheme of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0043] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0044] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0045] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0046] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0047] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0048] This paper proposes a modeling method for refueling schemes to achieve efficient and reliable in-flight refueling planning for hydrogen-powered drones. This method effectively addresses complex constraints and helps design efficient refueling scheme planning algorithms. It also enhances the continuous mission capability of drone swarms, optimizes resource allocation, and improves execution efficiency.

[0049] Based on this, such as Figure 1 As shown, the present invention provides a modeling method for a hydrogen-powered drone swarm refueling scheme, comprising the following steps:

[0050] S1. Based on the event-driven model, the flight process of the UAV is decomposed into several independent events. Each independent event includes the start position, end position, start time, end time, start payload, end payload, and energy consumption information.

[0051] S2. The three-alignment principle is adopted to manage several independent events of the mission drone and the refueling drone to align the time, position and energy of each independent event; and the energy consumption of the drone during flight is dynamically calculated based on the remaining fuel of the drone according to the nonlinear energy consumption model.

[0052] S3. The continuous flight space is rasterized, the flight path is discretized into several path grid points, obstacle avoidance planning is performed on the flight path of the mission UAV, and the flight path of the mission UAV is formed.

[0053] S4. Using stack genetic programming, random points are cyclically inserted into the flight path of the mission drone as refueling rendezvous points and assigned to the corresponding refueling drone paths to form a refueling scheduling scheme for hydrogen energy drone clusters.

[0054] S5. For the energy replenishment scheduling scheme of hydrogen-powered drone swarm, calculate the overall energy consumption ratio and energy depletion time ratio of the mission drone and the replenishment drone, and use the overall energy consumption ratio and energy depletion time ratio as the optimization objectives of the model.

[0055] The method of this invention, when applied to the design of swarm air replenishment planning algorithms, effectively models various complex constraints such as time, space, and energy in the scenario. By establishing a replenishment modeling method based on event-event chains, it provides effective model support for subsequent scheduling optimization, enabling scientific analysis of path planning and replenishment scheduling between mission UAVs and replenishment UAVs, thereby helping to achieve continuous and efficient task execution of UAV swarms. Furthermore, based on event-driven model construction and the three-alignment principle, this invention integrates replenishment rendezvous point insertion, energy consumption model, and path obstacle avoidance model. In the solution encoding process, it introduces the idea of ​​stack genetic programming, which can effectively model various complex constraints such as time, space, and energy in the scenario, thereby reducing the difficulty of algorithm design and providing technical support for improving the accuracy of UAV scheduling.

[0056] The following detailed description of the modeling method and system for a hydrogen-powered drone swarm refueling scheme of the present invention, in conjunction with specific embodiments and accompanying drawings, provides further details.

[0057] like Figure 2The diagram illustrates an in-flight refueling scenario for a swarm of hydrogen-powered drones, showcasing the scheduling of multiple hydrogen-powered drones for in-flight refueling during a mission. Two different types of controlled objects are involved: refueling drones and mission drones. The optimization objective is to minimize the overall energy consumption of both mission and refueling drones while ensuring the safe and accurate return of mission drones to and from the mission area. The entire airspace is divided into refueling and potential obstacle zones by solid black lines. The left-hand zone is the refueling zone, while the right-hand zone poses potential challenges to the flight safety of refueling drones due to obstacles such as tall buildings and complex terrain. Therefore, to ensure flight safety, in-flight refueling can only be conducted within the refueling zone. The ellipses on the left and right diagonal lines represent the refueling and obstacle zones, respectively. Circular and triangular nodes represent waypoints for different mission and refueling drones, respectively. Correspondingly, different types of lines passing through them represent the flight paths of different refueling and mission drones. The solid circle represents the drone's takeoff and landing airport. The square area is the mission area. The entire flight process of the mission drones is as follows: several mission drones take off from the airport and perform their first in-flight refueling upon reaching the corresponding refueling point. Then, they fly to the mission area to complete the mission. Afterward, they return to the refueling point along the planned path for a second in-flight refueling, and finally return to the landing airport. Similarly, the entire flight process of the refueling drone is as follows: taking off from the airport, flying over all the ordered in-flight refueling mission points in sequence, completing each refueling mission, and finally returning to the landing airport.

[0058] Based on the various complex factors in the scenario of hydrogen-powered drone swarms refueling in the air, the model established in this invention is based on the following assumptions:

[0059] Assumption 1: In aerial refueling missions, only hydrogen fuel can be transported, and mission materials such as sensors and cargo of the mission drone can only be loaded at ground airports.

[0060] Assumption 2: Each mission drone will recharge twice during the entire round-trip mission: the first time on its path after leaving the airport to ensure sufficient energy to reach the mission area from the departure airport; the second time on its return path to ensure sufficient energy to return from the mission area to the landing airport. Each recharging mission uses a one-to-one rendezvous method to connect the mission drone and the recharging drone.

[0061] Assumption 3: The continuous flight space is discretized into several path grid points. Each grid point represents a coordinate in the flight and is numbered sequentially to form the UAV's flight path. The entire path can be viewed as a set of multiple path points arranged sequentially, where the first and last points represent the UAV's start and end points, respectively.

[0062] Assumption 4: When the drones perform their missions, the obstacles they need to avoid, such as tall buildings and complex terrain, are set up as several circular obstacle zones. Since the flight paths of all refueling drones have already avoided potential obstacle zones, the refueling drones will fly between the airport and each aerial refueling mission point.

[0063] Assumption 5: The energy consumption of each drone during takeoff and landing is calculated based on average energy consumption.

[0064] Assumption 6: The drone flies at cruising speed throughout the entire operation, with its altitude and speed remaining constant. The mission drone flies between every two waypoints in the air.

[0065] Assumption 7: The drone does not take into account the effects of weather, wind direction and wind speed.

[0066] like Figure 1 As shown, this invention systematically analyzes the aerial refueling problem. We model the aerial refueling problem hierarchically based on event-event chains. The overall modeling approach is to first establish model assumptions based on mission requirements, including energy consumption models for the mission UAV and the refueling UAV, path planning models, obstacle area constraints, and event-driven flight processes. The energy consumption model considers the nonlinear energy consumption relationship of hydrogen-powered UAVs at specific altitudes. In path planning, the continuous space is discretized into path grid points, and the flight path of the UAV is accurately described by path numbering. In addition, the flight process of the mission UAV and the refueling UAV is decomposed into flight events, refueling events, mission events, and waiting events through the event-driven model. By aligning the time, position, and energy between events, the synchronization of the mission UAV and the refueling UAV during the aerial refueling process is ensured. In the solution encoding process, by separating the path planning of the mission UAV from the refueling task, the refueling rendezvous point and task allocation are gradually inserted using the idea of ​​stack genetic programming, thereby generating an aerial refueling scheduling scheme with precise alignment of time, space, and energy. Finally, regarding optimization objectives, the goals are to minimize energy consumption and reduce the proportion of drones flying in low-energy conditions.

[0067] This invention relates to the construction of an event-driven model.

[0068] To accurately describe the dynamic behavior and energy consumption changes of mission-oriented and refueling UAVs during in-flight refueling, this invention employs an event-event chain-based modeling method. The entire flight process of the UAV is divided into multiple independent events, each representing an action at a specific stage of the flight. Specifically, the event types include the following categories:

[0069] Takeoff event: The mission drone or refueling drone takes off from the ground airport and enters cruise mode.

[0070] Flight event: The process of a drone flying through the air from one waypoint to another. Each flight event records the starting position, ending position, starting time, and flight energy consumption.

[0071] like Figure 3 As shown, the refueling event involves the mission drone and the refueling drone docking in mid-air to replenish hydrogen energy, ensuring the mission drone has sufficient energy to complete its subsequent flight mission. During this process, it is crucial to ensure that the mission drone and the refueling drone start at the same time and from the same starting position.

[0072] Mission events: Mission events refer to the reconnaissance, aerial survey, and other operations performed by the UAV at a specific mission point. The duration of a mission event is determined according to mission requirements.

[0073] Waiting events: When the arrival times of the mission drone and the refueling drone at the refueling rendezvous point are inconsistent, a waiting event is triggered, increasing the flexibility and accuracy of the scheduling process.

[0074] Landing event: The mission or refueling drone returns to the airport and lands safely after completing its mission.

[0075] like Figure 4 and Figure 5 As shown, the events above occur sequentially, forming the event chain of the entire mission. Each event has clearly defined attributes, including start position, end position, start time, end time, start payload, end payload, and energy consumption. To ensure smooth transitions between events, we employ a three-alignment principle: time, position, and energy alignment. The end state of each event is used as the start state of the next event. For example, the end position of a flight event is the start position of a refueling event. This state transition allows for seamless integration of energy consumption and mission scheduling.

[0076] In the event-driven model, a unified energy consumption model can be used for events of the same type to simplify energy consumption calculations. For example, takeoff events, flight events, and refueling events all have their corresponding energy consumption models, as follows:

[0077] Takeoff event: The energy consumption during the takeoff phase is set to a fixed value, representing the energy required for the drone to take off from the ground and reach cruise mode.

[0078] Flight Events: At a fixed flight altitude, the energy consumption of a UAV per unit distance is inversely proportional to its total weight. Total weight includes the UAV's empty weight and payload weight, with the payload weight comprising fuel and mission supplies. During flight, as fuel is continuously consumed, the UAV's weight gradually decreases, thus changing the rate of energy consumption. Therefore, we established an energy consumption function that considers the impact of remaining fuel on energy consumption, effectively describing the dynamic changes in UAV energy consumption during flight.

[0079] Energy replenishment event: During the aerial energy replenishment process, both the mission drone and the replenishment drone are in a hovering state. The mission drone's energy decreases linearly over time, while the replenishment drone will consume additional energy supplied to the mission drone.

[0080] Waiting event: The drone is hovering, and its energy decreases linearly over time.

[0081] Task Events: The energy consumption of a task event depends on the task type. Different types of tasks consume different fixed amounts of energy.

[0082] Landing event: The energy consumption during the landing phase is set to a fixed value, representing the energy required for the drone to land from cruise mode to the ground.

[0083] By adopting this event-event chain modeling approach, we can comprehensively capture all the key actions and energy consumption of mission drones and refueling drones during execution, thereby achieving refined management of the entire mission process.

[0084] This invention relates to the encoding of solutions and the generation of baseline solutions.

[0085] The goal of aerial refueling scheduling for hydrogen-powered drone swarms consists of four parts: the flight path of each task drone, the rendezvous point of each aerial refueling task, the refueling task allocation, and the flight path of each refueling drone. Since the computational complexity of simultaneously performing task scheduling and path planning is high, we separate the path arrangement of task drones from the refueling scheduling. After planning the task drone paths, we use a stack-based genetic programming approach to sequentially insert the refueling rendezvous point and task allocation, thereby simplifying the computational complexity.

[0086] like Figure 6 As shown, the initial baseline solution is obtained by the mission UAV without recharging during flight. The algorithm optimizes the path of the mission drone and the path of the refueling drone, which only includes the take-off and landing airports. In the baseline solution, we assume that the aircraft can continue to fly at the original speed when its energy is depleted, so the baseline solution that cannot perform any refueling is reasonable to exist in the optimization process.

[0087] After obtaining the baseline solution, we introduce the idea of ​​stack genetic programming to sequentially insert a refueling rendezvous point into the paths of each mission UAV before and after the mission, until all the departure and return paths of all mission UAVs are filled, thus obtaining a complete in-flight refueling scheduling scheme. The specific operation is as follows:

[0088] Randomly select a path point in the energy replenishment area as the first energy replenishment rendezvous point;

[0089] Randomly select a mission drone and a refueling drone for matching;

[0090] Will Insert it between any two adjacent path points in the departure or return path of the mission drone;

[0091] This invention inserts a refueling rendezvous point between any two adjacent path points in the flight path of the refueling drone; according to the above steps, the operation of inserting a refueling rendezvous point can be encoded as follows:

[0092] [ , , , , ];

[0093] in: Indicates the coordinate number of the point where the energy complement meets; It is the serial number of the mission drone; This indicates that the refueling rendezvous point is inserted into the path of the mission drone. The path point and the first +1 path point in the middle; It is the serial number of the refueling drone; This indicates that the refueling rendezvous point is inserted into the path of the refueling drone. The path point and the first +1 path point in the middle.

[0094] like Figure 7 As shown, in the baseline solution, the energy replenishment point will be... =168 (the number of coordinates [600,400]) is inserted between the second and third path points of mission UAV 1 and between the first and second path points of refueling UAV 1. The encoding operation is: [168,1,2,1,1].

[0095] Based on the above, a complete solution can be obtained from 2* ( The present invention uses an encoding representation of (number of mission UAVs) insertion rendezvous points. Through this encoding method, the solution provided by this invention is no longer a description of the entire aerial refueling process, but rather a series of refueling event insertion schemes. This description of the refueling event insertion scheme is called a super-solution. The super-solution can accurately represent the path planning and mission scheduling schemes of the mission UAV and the refueling UAV, thereby further adjusting the path in subsequent optimization processes to minimize energy consumption.

[0096] In the aerial refueling problem, the optimization objective of this invention is to minimize the overall energy consumption of the mission UAV and the refueling UAV. However, since we assume that the mission UAV can still fly with zero energy in the solution encoding, in addition to minimizing energy consumption, it is also necessary to calculate the proportion of the flight time of all mission UAVs in the zero-energy state to the total time in each solution, so that this value is as small as possible, thereby improving the safety and execution efficiency of the mission.

[0097] This invention assumes that the replenishment drone and the mission drone can still fly when their energy is depleted, and that their flight speed remains constant. Therefore, the duration during which energy is depleted to 0 within the time window of each event can be statistically analyzed. By statistically analyzing all the durations of zero energy consumption throughout the entire flight event chain and dividing this by the total time, the optimization objective of the energy depletion time ratio is obtained. Clearly, the optimization objective for the energy depletion time ratio should be to reduce it to 0.

[0098] This modeling method for energy drone swarm refueling solutions can effectively help optimize endurance and mission execution efficiency. Through three-aligned event management and dynamic refueling insertion points, it can effectively model various complex constraints such as time, space, and energy in the scenario, thereby reducing the difficulty of algorithm design.

[0099] The following comparative experiments will provide a more detailed explanation of the modeling method and system for a hydrogen-powered drone swarm refueling scheme of the present invention.

[0100] A comparison between this invention and direct coding path modeling.

[0101] To verify the effectiveness of this invention, it was compared with traditional methods that directly encode and model paths in both simple and complex scenarios. Scenario 1 included one refueling drone and one mission drone, while Scenario 2 included four refueling drones and three mission drones. These two scenarios simulated mission execution environments ranging from low to high complexity, aiming to evaluate the applicability of the two modeling methods.

[0102] like Figure 8 As shown, from Figure 8As can be seen from the simulation, in Scenario 1, the mission drone needs to recharge twice with the recharge drone during the mission execution. Simulation of the mission process reveals that while the traditional direct path coding method is intuitive in terms of path planning, the lack of precise management of multiple constraints such as time, space, and energy leads to the mission drone and the recharge drone failing to dock successfully during the second recharge.

[0103] like Figure 9 As shown, in Scenario 2, the complexity of the mission and refueling increases. The three mission drones need to complete multi-mission flights and dock with four refueling drones multiple times during their flight. Traditional modeling methods exhibit certain limitations in the simulation. Although the first refueling mission of each refueling drone can be matched with the mission drone due to the controllable takeoff time, in subsequent refueling missions, due to the complex interaction of time and path, traditional modeling methods cannot coordinate the temporal and spatial alignment of the refueling drones and mission drones in all refueling events, resulting in the failure of some refueling missions. The modeling method of this invention achieves precise temporal and spatial alignment of all refueling events.

[0104] Table 1. Comparison of rendezvous success rates between the two encoding methods.

[0105]

[0106] like Figure 10 As shown in the simulation results, the present invention demonstrates significant advantages in both scenarios. Based on the event-driven three-alignment principle, the present invention tightly integrates time, location, and energy during the modeling process, enabling more systematic management of refueling task scheduling. It provides a detailed modeling framework for collaborative refueling tasks in UAV swarms, greatly reducing the difficulty of subsequent algorithm design and optimization, improving solution efficiency and feasibility, and enhancing the effectiveness of UAV scheduling planning.

[0107] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Anyone skilled in the art can readily implement the present invention according to the description and above. Any modifications, alterations, or variations made based on the disclosed technical content are equivalent embodiments of the present invention. Furthermore, any modifications, alterations, or variations made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the present invention.

Claims

1. A modeling method for a hydrogen-powered drone swarm refueling scheme, characterized in that, Includes the following steps: S1. Based on the event-driven model, the flight process of the UAV is decomposed into several independent events. Each independent event includes the start position, end position, start time, end time, start payload, end payload, and energy consumption information. S2. The three-alignment principle is adopted to manage several independent events of the mission drone and the refueling drone to align the time, position and energy of each independent event; and the energy consumption of the drone during flight is dynamically calculated based on the remaining fuel of the drone according to the nonlinear energy consumption model. S3. The continuous flight space is rasterized, the flight path is discretized into several path grid points, obstacle avoidance planning is performed on the flight path of the mission UAV, and the flight path of the mission UAV is formed. S4. Using stack genetic programming, random points are cyclically inserted into the flight path of the mission drone as refueling rendezvous points and assigned to the corresponding refueling drone paths to form a refueling scheduling scheme for hydrogen energy drone clusters. S5. For the energy replenishment scheduling scheme of hydrogen-powered drone swarm, calculate the overall energy consumption ratio and energy depletion time ratio of the mission drone and the replenishment drone, and use the overall energy consumption ratio and energy depletion time ratio as the optimization objectives of the model.

2. The modeling method for the hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S1, each independent event represents an action at a specific stage during flight. Several independent events include takeoff events, flight events, refueling events, mission events, waiting events, and landing events.

3. The modeling method for the hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S1, the energy consumption of the takeoff event is a fixed value; the energy consumption of the UAV per unit distance during the flight event is inversely proportional to its total weight; during the refueling event, the energy of the mission UAV decreases linearly over time, and the refueling UAV will consume additional energy supplied to the mission UAV; the energy consumption in the mission event depends on the mission type; and the energy consumption of the landing event is set to a fixed value.

4. The modeling method for the hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S1, in the event-driven model, a uniform energy consumption model is used to calculate the energy consumption for independent events of the same type.

5. The modeling method for the hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S2, the three alignment principles include time alignment, position alignment, and energy alignment.

6. The modeling method for the hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S2, the nonlinear energy consumption model establishes an energy consumption function based on the impact of the UAV's remaining fuel on energy consumption.

7. The modeling method for the hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S3, several path grid points are numbered sequentially, and obstacle avoidance planning is performed on the mission UAV path to obtain the mission UAV path in the baseline solution. The path of the refueling UAV in the baseline solution consists only of the coordinate numbers of its take-off and landing airports.

8. The modeling method for a hydrogen-powered drone swarm refueling scheme according to claim 1, characterized in that, In S4, stack genetic programming is used to sequentially insert refueling rendezvous points into the departure and return paths of the mission drones and match the refueling drones until all the paths of the mission drones are filled.

9. The modeling method for a hydrogen-powered drone swarm refueling scheme according to claim 8, characterized in that, S4 specifically includes: S51. Randomly select a path point in the energy replenishment area as the first energy replenishment rendezvous point; S52. Randomly select a mission drone and a refueling drone for matching; S53. Insert the refueling rendezvous point between any two adjacent path points in the departure or return path of the mission UAV. S54. Insert the refueling rendezvous point between any two adjacent path points in the flight path of the refueling drone.

10. A system upon which the modeling method for a hydrogen-powered drone swarm refueling scheme according to any one of claims 1-9 is based, characterized in that, The system includes an event-driven model, an event management module, a path planning module, and a power replenishment scheduling module, wherein: The event decomposition module is used to decompose the flight process of the UAV into several independent events, including information on the start position, end position, time, payload, and energy consumption. The event management module is used to manage independent events using a three-alignment principle and dynamically calculate energy consumption based on a nonlinear energy consumption model. The path planning module is used to perform gridding processing on the continuous flight space, discretize the flight path into several path grid points, perform obstacle avoidance planning on the flight path of the mission UAV, and form the flight path of the mission UAV. The energy replenishment scheduling module is used to use stack genetic programming to cyclically insert random points as energy replenishment rendezvous points in the flight path of the mission drone and assign them to the corresponding energy replenishment drone paths to form an energy replenishment scheduling scheme for the hydrogen energy drone cluster; calculate the overall energy consumption ratio and energy depletion time ratio of the mission drone and the energy replenishment drone, and use the overall energy consumption ratio and energy depletion time ratio as the optimization objectives of the model.

Images