Airway network construction method based on simulated annealing

Abstract

The invention discloses an air route network construction method based on simulated annealing, and the method comprises the following steps: 1, setting a target airspace, obtaining an available airspace range and a no-fly zone range in the target airspace, and forming an airspace information database; 2, acquiring all flight task data in a target airspace to form a task database; step 3, carrying out random initialization on the air route network in the target airspace; step 4, designing a loss function for the flight mission to represent the advantages and disadvantages of the air route network; and step 5, iteratively optimizing the air route network by adopting a simulated annealing method and taking minimization of the loss function as a target, finally obtaining an optimal air route network, and completing air route network construction based on simulated annealing. The method is an auxiliary decision-making method aiming at the current situation of artificially constructing the air route network in the air management field, and is beneficial to the optimization work of the follow-up management department on the air route network.

Description

Technical Field

[0001] The invention relates to a route network construction method, in particular to a route network construction method based on simulated annealing. Background Art

[0002] This section merely provides background information related to the present disclosure and is not necessarily prior art.

[0003] The route network is a general term for routes and waypoints in the airspace. All aircraft flying in the airspace need to fly along the routes, so the construction of the route network is an important part of ensuring the normal flight of flights. There are three main difficulties in the construction of the route network:

[0004] First, the number of waypoints. Assuming in an extreme case, there are no waypoints in the airspace, and all flights fly directly, then the total distance must be the shortest, but it will be difficult for controllers to manage, which will lead to a decrease in airspace capacity. In another extreme case, there are a large number of waypoints in the airspace, then each flight will most likely find a path similar to a direct flight, which can also shorten the total distance, but will cause pilots and controllers to face an extremely complex flight plan, which is also not conducive to daily flight operations. Therefore, it is necessary to find an appropriate number of waypoints to meet the needs of both shorter mission distances and lower control difficulty;

[0005] Second, the arrangement of waypoints. Routes are constructed based on waypoints. Under the premise of limiting the number of waypoints, the arrangement of waypoints will affect the arrangement of routes, thereby affecting the flight distance of flights in the airspace, and further affecting the flight time and fuel consumption. Therefore, the reasonable arrangement of waypoints is the second biggest problem in the construction of route networks.

[0006] Third, route design: which waypoints need to be connected by routes and which waypoints do not need to have routes is a major difficulty in building a route network. Too many routes connecting the same waypoint will bring about control difficulties, while too few routes will affect the connectivity of the entire route network. Therefore, it is necessary to find a suitable algorithm to solve this problem.

[0007] In the current air traffic control field, the construction of route networks mainly relies on the experience of professionals, which is inefficient and cannot guarantee that the constructed route network is optimal. Therefore, there is an urgent need to design an automated route network construction method to provide planning suggestions for staff and find the optimal route network construction plan under the set rules, thereby reducing manual compliance and minimizing the total flight distance in the airspace.

[0008] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0009] Purpose of the invention: The technical problem to be solved by the present invention is to provide a route network construction method based on simulated annealing in view of the shortcomings of the prior art.

[0010] In order to solve the above technical problems, the present invention discloses a route network construction method based on simulated annealing, comprising the following steps:

[0011] Step 1: Set the target airspace and obtain the available airspace range and the no-fly zone range in the target airspace to form an airspace information database;

[0012] Step 2, obtaining all flight mission data within the target airspace to form a mission database;

[0013] Step 3, randomly initialize the route network in the target airspace;

[0014] Step 4: Design a loss function for the flight mission to characterize the quality of the route network;

[0015] Step 5, using the simulated annealing method, with the goal of minimizing the loss function, iteratively optimizing the route network, finally obtaining the optimal route network, and completing the construction of the route network based on simulated annealing.

[0016] Furthermore, the forming of the airspace information database described in step 1 includes:

[0017] Step 1-1, in the set target airspace range, obtain the available airspace range information and store it in the airspace information database in the form of polygons;

[0018] The polygon is the coordinate information of each vertex arranged in order, and the last point is connected to the first point to form a closed polygon;

[0019] Step 1-2, in the set target airspace range, obtain the no-fly zone range information and store it in the airspace information database in the form of polygons.

[0020] Furthermore, the forming of the task database described in step 2 includes:

[0021] Step 2-1, flight mission data, including: the take-off airport, i.e., the take-off point, and the landing airport, i.e., the landing point, in the flight mission;

[0022] Step 2-2, associating the task database with the airspace information database to obtain a task set database; the task set database includes: a task list, and the airspace corresponding to the task.

[0023] Furthermore, the random initialization of the route network in the target airspace described in step 3 includes:

[0024] Step 3-1, according to the target airspace size S and the expected interval size d between waypoints, determine the expected number of waypoints Np in the target airspace, and perform spatially uniform random initialization with the number of waypoints N within the available airspace; wherein the number of waypoints Np is calculated as follows:

[0025]

[0026] Among them, α represents the redundancy of waypoints during random initialization;

[0027] Step 3-2, after random initialization, connect all waypoints whose intervals are less than or equal to d to form a route;

[0028] Step 3-3, connecting the airport in the target airspace with the waypoints in the surrounding area whose distance is less than or equal to d to form a route;

[0029] Step 3-4, compare the no-fly airspace information and delete all routes that cross the no-fly airspace.

[0030] Furthermore, the design loss function described in step 4 includes:

[0031] Design the total loss function L of the route network under the current flight mission All , the calculation method is as follows:

[0032] L All =λ a L a +λ b L b +λ c L c

[0033] Among them, λ a ,λ b and λ c is the weighting coefficient, L a is the first loss function, which represents the loss caused by the total voyage; L b is the second loss function, which represents the loss caused by the short distance of the waypoint; L c is the third loss function, which represents the loss caused by the inability to connect the starting point and the end point of the flight mission.

[0034] Furthermore, the calculation method of the first loss function includes:

[0035] For all flight missions [T 1 , T 2 , …, T N], based on the route network obtained in step 3, the shortest routes are planned in sequence as follows:

[0036] [R 1 , R 2 , …, R N ]

[0037] For each route, calculate its total distance [L 1 , L 2 , …, L N ], if a flight mission is not connected in the route network, the total distance of the route is temporarily recorded as 0, and the total distance of N missions is L a As the first loss function, the calculation method is as follows:

[0038] L a =L 1 +L 2 +,…,+L N .

[0039] Furthermore, the second loss function is calculated by:

[0040] For each route, check whether the length of each route section is less than λd, where λ is the proportional coefficient; when it is detected that the length of a route section is less than λd, the difference D′ between the length D of the route section and λd is recorded as:

[0041] D′=λd-D

[0042] After detecting and calculating each section of all routes, all the differences D′ are added together as the second loss function L b , the calculation method is as follows:

[0043] L b =∑D′.

[0044] Furthermore, the calculation method of the third loss function includes:

[0045] For the unconnected tasks in the route network, a separate loss L′ is set for it, and the third loss function L is calculated c , the calculation method is as follows:

[0046] L c =N′×L′

[0047] Wherein, N′ represents the number of tasks that have no access in the route network among all tasks.

[0048] Further, the simulated annealing method described in step 5 includes:

[0049] Step 5-1, calculate the total loss function L according to the waypoints and route distribution in the route network after random initialization in step 3 All , and define the first parameter current_loss and the second parameter best_loss to represent the current total loss function and the best total loss function in this cycle;

[0050] Step 5-2, according to the waypoints passed by all flight missions, the waypoint list is divided into two categories: in-use waypoints and out-of-use waypoints, and random disturbances of different amplitudes are added to the two types of waypoints;

[0051] Step 5-3, calculate the new total loss new_loss after adding random perturbations. If the total loss is less than the first parameter current_loss, use the new total loss new_loss as the first parameter current_loss, record the new waypoint coordinates, and update the second parameter best_loss according to the new total loss new_loss.

[0052] Step 5-4, if the new total loss new_loss after adding the disturbance is greater than the first parameter current_loss, but the random number random between 0 and 1 is less than e (current_loss-new_loss) / temperature , the new waypoint coordinates are also recorded, and the new total loss new_loss is used as the first parameter current_loss, but the second parameter best_loss is not updated, where temperature is the heat value, which is a hyperparameter in the simulated annealing method;

[0053] Step 5-5, repeat steps 5-2 to 5-4, and reduce the temperature value after each repetition until the exit condition is met, and then execute step 5-6;

[0054] Step 5-6, determine the final waypoint update result according to the waypoint coordinates corresponding to the second parameter best_loss; record the final waypoint coordinates, the connection between waypoints, and the connection between airports and waypoints, and select the waypoints and routes used by at least one task as the final route network.

[0055] Furthermore, the exit conditions described in step 5-5 include:

[0056] Condition 1: The total loss is less than the set threshold;

[0057] Condition 2: The number of iterations is greater than the set threshold;

[0058] Condition 3: The number of iterations during which the second parameter best_loss does not change is greater than the set threshold.

[0059] Beneficial effects:

[0060] The method proposed in the present invention transforms from manual route network construction to automatic route network construction, automatically adjusts the positions of waypoints and routes based on limited rules, and finds the optimal construction plan under the limited rules, so as to reduce the total range of the flight mission while meeting the current control status. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0062] Figure 1 This is a schematic diagram of the airspace information database.

[0063] Figure 2 Schematic diagram of vertex coordinate database.

[0064] Figure 3 Schematic diagram of the task collection database.

[0065] Figure 4 This is a schematic diagram of the task database.

[0066] Figure 5 Flowchart of the route network construction method based on simulated annealing.

[0067] Figure 6 This is the flow chart of the simulated annealing algorithm.

[0068] Figure 7 The effect diagram of multiple rounds of iterations for route network planning. DETAILED DESCRIPTION

[0069] The principle of the present invention is to randomly scatter waypoints in the available airspace, and connect the waypoints with routes according to certain rules to form an initial route network, and then design a loss function to characterize the pros and cons of the current channel network for the current task set, and then minimize the loss function through multiple rounds of iterations based on the simulated annealing algorithm, so as to obtain the route network construction plan in the available airspace that is most suitable for the current task set.

[0070] The present invention provides a route network construction method based on simulated annealing, such as Figure 5 As shown, the following steps are included:

[0071] Step 1: Obtain the available airspace range and the no-fly zone range information therein, and store them in the airspace information database. Step 1 specifically includes:

[0072] Step 1-1, obtain airspace range information and store it in an airspace information database in a polygonal form, including coordinate information of each vertex of the polygon arranged in sequence, and the last point is connected to the first point to form a closed polygon;

[0073] Step 1-2, obtaining the range information of the no-fly zone and storing it in the airspace information database in the same polygonal form as step 1-1;

[0074] Step 2, obtaining all flight mission data within the airspace and storing them as a mission database. Step 2 specifically includes:

[0075] Step 2-1, the mission database must contain at least the mission's departure airport (take-off point) and landing airport (landing point) information;

[0076] Step 2-2, the task database needs to be associated with the task set database, which contains the task list and the airspace corresponding to the task set;

[0077] Step 3: Randomly initialize the route network, including:

[0078] Step 3-1, first determine the expected number of waypoints N in the airspace according to the airspace size S and the expected interval size d between waypoints, and use this number to perform spatially uniform random initialization within the available airspace, where 2 in the formula can be replaced with other values ​​according to the mission situation;

[0079]

[0080] Among them, α represents the redundancy of waypoints during random initialization. The larger the value, the lower the calculation efficiency and the higher the accuracy.

[0081] In one embodiment of the present invention, it is preferably set to 2.

[0082] Step 3-2, connect all waypoints whose intervals are less than or equal to d to form a route;

[0083] Step 3-3, connect the waypoints with a distance less than or equal to d between the airport and the surrounding area to form a route;

[0084] Step 3-4, compare the no-fly airspace information and delete all routes that cross the no-fly airspace;

[0085] Step 4, design a loss function for the flight mission to characterize the quality of the route network. Step 4 specifically includes:

[0086] Step 4-1, for all flight tasks [T1, T2... TN], plan their shortest paths [R1, R2... RN] in sequence based on the route network;

[0087] Step 4-2, for each path, calculate its total distance [L1, L2, ... LN]. If a flight mission is not connected in the route network, the total distance is temporarily recorded as 0. The total distance of N missions is the sum of their L a =L1+L2+... +LN;

[0088] Step 4-3, for each path, check whether the length of each route segment is less than λd, where λ is a proportional coefficient. The smaller the value, the looser the distance restriction between waypoints in the route network construction, and the larger the value, the more fixed the distance between waypoints. Whenever it is detected that the length of a route segment is less than λd, the difference between the length D of the route segment and λd is recorded as D'=λd-D. After detecting and calculating each route segment in all paths, add up all D' to get L b =∑D′;

[0089] Step 4-4: For the tasks detected in step 4-2 that are not connected in the route network, set a separate loss L' for them. The number of tasks that have no access in the route network multiplied by the loss L' is the loss L. c =N′×L′;

[0090] Step 4-5, summarize the three types of losses to obtain the total loss function L of the channel network under the current flight mission All =λ a L a +λ b L b +λ c L c ;

[0091] Step 5, using simulated annealing algorithm to minimize the loss function, so as to obtain the optimal route network construction plan; Step 5 specifically includes:

[0092] Step 5-1, calculate the total loss function L according to the initialized waypoints and route distribution All , and define the parameters current_loss and best_loss to store the current total loss function and the best total loss function in the cycle;

[0093] Step 5-2, according to the waypoints passed by all tasks, the waypoint list is divided into two categories: "waypoints in use" and "waypoints not in use", and random disturbances of different amplitudes are added to these two types of waypoints. The disturbance of the waypoints in use is lower, and the disturbance of the waypoints not in use is larger. In this embodiment, they are uniformly distributed within 0.1 kilometers and uniformly distributed within 10 kilometers respectively;

[0094] Step 5-3, calculate the total loss new_loss after adding the disturbance. If the total loss is less than the current_loss, use the new loss function as the current_loss, record the new waypoint coordinates, and update the best_loss according to the total loss.

[0095] Step 5-4, if the total loss after adding the perturbation is greater than current_loss, but the random number between 0 and 1 is less than e (current_loss-new_loss) / temperature , the new waypoint coordinates are also recorded, and the new loss function is used as current_loss, but best_loss is not updated. Temperature is the heat value, which is a hyperparameter of the simulated annealing algorithm. It is used to determine whether the updated loss must be less than the loss before the update during parameter update.

[0096] Step 5-5, repeat steps 5-2 to 5-4, and lower the temperature after each repetition until the exit condition is met. The exit condition includes: ① the total loss is less than a certain threshold; ② the number of iterations is greater than a certain threshold; ③ the number of iterations in which best_loss does not change is greater than a certain threshold. Finally, the final waypoint update result is determined based on the waypoint coordinates corresponding to best_loss. The final waypoint location, the connection between waypoints, and the connection between airports and waypoints are recorded, and the waypoints and routes used by at least one task are selected as the final route network construction result.

[0097] Example:

[0098] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described below in conjunction with the accompanying drawings and embodiments. Since the description of the embodiments is specific and cannot cover all embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making innovative work belong to the protection scope of the present invention.

[0099] Figure 1 It is a schematic diagram of the airspace information database, which includes the airspace ID, airspace availability and a list of vertex IDs that constitute the airspace. The availability of 1 indicates that the airspace is an available airspace, and 0 indicates that the airspace is a prohibited airspace. In the polygon vertex ID list, AAA, BBB, etc. are vertex IDs, and the airspace is a closed polygon formed by the lines from AAA to BBB, the line from BBB to CCC, the line from CCC to DDD, and the line from DDD to AAA.

[0100] Figure 2 This is a schematic diagram of the waypoint coordinate database, which includes the vertex ID and its longitude and latitude.

[0101] Figure 3 It is a schematic diagram of the task set database, which includes the task set ID, the involved airspace ID and the included task ID list, which is used to indicate the available airspace of each task set and which specific tasks are included in the set.

[0102] Figure 4 It is a schematic diagram of the task database, which includes the take-off and landing airports for each task. In this embodiment, it is set that the take-off and landing airports of all tasks are located in the available airspace. If there are tasks that cross the available airspace during implementation, the airspace entry point and exit point can be filled in the table as the take-off and landing airport coordinates.

[0103] Figure 5 It is a flow chart of the route network construction method based on simulated annealing, which shows the overall process of the route network construction method proposed in the present invention.

[0104] Figure 6 This is a simulated annealing algorithm flow chart, which shows the specific process of the simulated annealing algorithm adopted by the present invention.

[0105] Figure 7 This is a diagram showing the effect of multiple rounds of iterations in route network planning. The dot starting with D represents the take-off airport, which is the first letter of departure. The dot starting with L represents the landing airport, which is the first letter of landing. The dot starting with W represents the waypoint, which is the first letter of waypoint. D and L with the same number are a pair, indicating the location of the take-off airport and landing airport of a mission. Figure 6 In the example, each 1000 calculations of new_loss are recorded as one generation. It can be seen that at the beginning, the arrangement of waypoints and routes is not very scientific. Some tasks cannot even find a suitable path from the starting point to the end point. As the number of iterations increases, the arrangement of waypoints and routes gradually becomes more reasonable. In this example, no no-fly airspace is set in the available airspace.

[0106] In a specific embodiment, the specific process of the route network construction method based on simulated annealing is as follows:

[0107] Assume that a task is received to plan the route network for the airspace ID "123456", and the flight mission set ID in this airspace is "654321";

[0108] First obtain Figure 1 The airspace information database shown in the figure gives the ID of each airspace. The airspace with ID "123456" is found, and its corresponding polygon vertex list is [AAA, BBB, CCC, DDD];

[0109] Then get Figure 2The waypoint coordinate database shown corresponds to the airspace information database, and the specific position information of the polygon vertices of each airspace is known;

[0110] According to the flight mission set ID "654321", from Figure 3 The flight mission list covered by the mission set database is [aaa, bbb.....];

[0111] Then from Figure 4 The coordinates of the take-off and landing airports of these tasks are obtained by querying the task database shown;

[0112] This completes the data collection work;

[0113] Then, based on the area S of the available airspace and the reasonable spacing d between the desired waypoints, the formula Calculate the number of initial waypoints. In the embodiment, S is 10,000 square kilometers, d is 30 kilometers, so N is 23;

[0114] Then randomly initialize the waypoint positions in the available airspace, and initialize the coordinates of each waypoint in a spatially uniform distribution (i.e., if each individual waypoint is initialized multiple times, its position is uniformly distributed in space);

[0115] Then add routes according to the rules for connecting routes between waypoints whose distance is less than d, and use routes to connect the take-off and landing airports and their surrounding waypoints whose distance is less than d;

[0116] Then, for each task, the shortest path under the current route network is solved (in this embodiment, the shortest path is defined only by the total length of the route, and in different embodiments, other penalty functions can be added on the basis of the route length according to weather, flight flow, ground facilities, etc., and the shortest path is calculated based on this), and then the lengths of the shortest paths of all tasks are added together to obtain the total flight loss L a ;

[0117] Then, for each section of the planned route for each mission, check whether its length is less than 0.8d. If it is less than 0.8d, the loss of the part less than 0.8d is calculated according to a coefficient of 10. For example, for mission aaa, if a section of its planned route is 20km long, d is 30km, and 0.8d is 24km, then for this section of the route, a loss of (24-20)*10=40 is included. The loss of all routes of all missions is added together to obtain the loss L due to the waypoints being too close. b ;

[0118] Then for tasks where no path can be found (e.g. Figure 6D4 in the 0th generation has no path connected to L4), and the loss of 10,000 is directly increased as the loss L caused by the unconnected path c ;

[0119] In this embodiment, the three are directly added together to obtain the total loss L ALL ,In different task scenarios, the three types of losses can be weighted according to the needs;

[0120] Then, according to the simulated annealing method, the waypoint coordinates are randomly adjusted, and all the steps starting from adding the route are repeated to calculate the adjusted total loss, which is compared with the total loss in the 0th generation to determine the coordinate positions of the waypoints in the 1st generation;

[0121] In this embodiment, this step is repeated until the difference between the total loss of the Nth iteration and the N-5, N-4, N-3, N-2, and N-1 generations is less than 1, then the optimization is considered complete, and the current waypoints and route information are output as the final route network planning result.

[0122] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, the invention content of a route network construction method based on simulated annealing provided by the present invention and some or all steps in each embodiment can be executed. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0123] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of computer programs and their corresponding general hardware platforms. Based on such an understanding, the technical solutions in the embodiments of the present invention can be essentially or partly contributed to the prior art in the form of computer programs, i.e., software products, which can be stored in a storage medium and include several instructions for enabling a device including a data processing unit (which can be a personal computer, a server, a single-chip microcomputer, an MCU or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0124] The present invention provides a method and idea for constructing a route network based on simulated annealing. There are many methods and approaches to implement the technical solution. The above is only a preferred implementation of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A route network construction method based on simulated annealing, characterized in that: The following steps are involved: Step 1: Set the target airspace and obtain the available airspace range and the no-fly zone range in the target airspace to form an airspace information database; Step 2, obtaining all flight mission data within the target airspace to form a mission database; Step 3, randomly initialize the route network in the target airspace; Step 4: Design a loss function for the flight mission to characterize the quality of the route network; Step 5, using the simulated annealing method, with the goal of minimizing the loss function, iteratively optimizing the route network, finally obtaining the optimal route network, and completing the construction of the route network based on simulated annealing.

2. The method for constructing a route network based on simulated annealing according to claim 1, characterized in that: The step 1 of forming an airspace information database includes: Step 1-1, in the set target airspace range, obtain the available airspace range information and store it in the airspace information database in the form of polygons; The polygon is the coordinate information of each vertex arranged in order, and the last point is connected to the first point to form a closed polygon; Step 1-2, in the set target airspace range, obtain the no-fly zone range information and store it in the airspace information database in the form of polygons.

3. A route network construction method based on simulated annealing according to claim 2, characterized in that: The formation of the task database described in step 2 includes: Step 2-1, flight mission data, including: the take-off airport, i.e., the take-off point, and the landing airport, i.e., the landing point, in the flight mission; Step 2-2, associating the task database with the airspace information database to obtain a task set database; the task set database includes: a task list, and the airspace corresponding to the task.

4. A route network construction method based on simulated annealing according to claim 3, characterized in that: The random initialization of the route network in the target airspace described in step 3 includes: Step 3-1, according to the target airspace size S and the expected interval size d between waypoints, determine the expected number of waypoints Np in the target airspace, and perform spatially uniform random initialization with the number of waypoints N within the available airspace; wherein the number of waypoints Np is calculated as follows: Among them, α represents the redundancy of waypoints during random initialization; Step 3-2, after random initialization, connect all waypoints whose intervals are less than or equal to d to form a route; Step 3-3, connecting the airport in the target airspace with the waypoints in the surrounding area whose distance is less than or equal to d to form a route; Step 3-4, compare the no-fly airspace information and delete all routes that cross the no-fly airspace.

5. The method for constructing a route network based on simulated annealing according to claim 4, characterized in that: The design loss function described in step 4 includes: Design the total loss function L of the route network under the current flight mission All , the calculation method is as follows: L All =λ a L a +λ b L b +λ c L c Among them, λ a ,λ b and λ c is the weighting coefficient, L a is the first loss function, which represents the loss caused by the total voyage; L b is the second loss function, which represents the loss caused by the short distance of the waypoint; L c is the third loss function, which represents the loss caused by the inability to connect the starting point and the end point of the flight mission.

6. A route network construction method based on simulated annealing according to claim 5, characterized in that: The first loss function, the calculation method includes: For all flight tasks [T1,T2,…,T N ], based on the route network obtained in step 3, the shortest routes are planned in sequence as follows: [R1,R2,…,R N ] For each route, calculate its total distance [L1, L2, …, L N ], if a flight mission is not connected in the route network, the total distance of the route is temporarily recorded as 0, and the total distance of N missions is L a As the first loss function, the calculation method is as follows: L a =L1+L2+,…,+L N 。 7. A route network construction method based on simulated annealing according to claim 6, characterized in that: The second loss function is calculated by: For each route, check whether the length of each route section is less than λd, where λ is the proportional coefficient; when it is detected that the length of a route section is less than λd, the difference D between the length D of the route section and λd is ′ Recorded as: D′=λd-D After detecting and calculating each section of all routes, all the differences D′ are added together as the second loss function L b , the calculation method is as follows: L b =∑D′。 8. The method for constructing a route network based on simulated annealing according to claim 7, characterized in that: The third loss function is calculated by: For the unconnected tasks in the route network, a separate loss L′ is set for it, and the third loss function L is calculated c , the calculation method is as follows: L c =N′×L′ Wherein, N′ represents the number of tasks that have no access in the route network among all tasks.

9. A route network construction method based on simulated annealing according to claim 8, characterized in that: The simulated annealing method described in step 5 comprises: Step 5-1, calculate the total loss function L according to the waypoints and route distribution in the route network after random initialization in step 3 All , and define the first parameter current_loss and the second parameter best_loss to represent the current total loss function and the best total loss function in this cycle; Step 5-2, according to the waypoints passed by all flight missions, the waypoint list is divided into two categories: in-use waypoints and out-of-use waypoints, and random disturbances of different amplitudes are added to the two types of waypoints; Step 5-3, calculate the new total loss new_loss after adding random perturbations. If the total loss is less than the first parameter current_loss, use the new total loss new_loss as the first parameter current_loss, record the new waypoint coordinates, and update the second parameter best_loss according to the new total loss new_loss. Step 5-4, if the new total loss new_loss after adding the disturbance is greater than the first parameter current_loss, but the random number random between 0 and 1 <e (current_loss-new_loss) / temperature , the new waypoint coordinates are also recorded, and the new total loss new_loss is used as the first parameter current_loss, but the second parameter best_loss is not updated, where temperature is the heat value, which is a hyperparameter in the simulated annealing method; Step 5-5, repeat steps 5-2 to 5-4, and reduce the temperature value after each repetition until the exit condition is met, and then execute step 5-6; Step 5-6, determine the final waypoint update result according to the waypoint coordinates corresponding to the second parameter best_loss; record the final waypoint coordinates, the connection between waypoints, and the connection between airports and waypoints, and select the waypoints and routes used by at least one task as the final route network.

10. A route network construction method based on simulated annealing according to claim 9, characterized in that: The exit conditions described in step 5-5 include: Condition 1: The total loss is less than the set threshold; Condition 2: The number of iterations is greater than the set threshold; Condition 3: The number of iterations during which the second parameter best_loss does not change is greater than the set threshold.

Images