Multi-target distribution and route planning method based on simulated annealing algorithm

Abstract

The invention discloses a multi-target distribution and route planning method based on a simulated annealing algorithm, and the method comprises the steps: firstly, building a multi-target distribution model, defining related variables, and constructing a target function, so as to minimize the total flight cost, and meanwhile, considering the task execution time and fuel limitation; secondly, constructing a double-loop model, abstracting a problem into a Hamilton loop problem, including an outer Hamilton loop and an inner Hamilton loop which are respectively responsible for track planning between regional nodes and track planning of targets in a region; and finally, solving the TSP model of the double Hamilton loop by adopting a simulated annealing algorithm, and finding an optimal multi-target allocation strategy and a corresponding track plan through control parameter setting, initial solution generation, solution transformation, Metropolis criterion application and a cooling strategy. According to the method, large-scale, multi-target and multi-constraint complex problems can be effectively solved, the calculation efficiency is improved, a solution close to global optimum is found, and the method has important practical application value and market prospects.

Description

Technical Field

[0001] The invention belongs to the technical field of aircraft, and in particular relates to a multi-target allocation and track planning method based on a simulated annealing algorithm. Background Art

[0002] In modern military and civilian fields, the problem of multi-target allocation and trajectory planning is becoming increasingly important. This type of problem involves how to effectively allocate and plan multiple targets within limited resources and time to achieve optimal mission execution efficiency and effectiveness. For example, in military operations, accurate trajectory planning is required for strike and reconnaissance missions to ensure that the attack is completed during the period when the enemy's defense is weakest, while avoiding unnecessary waste of resources and risks.

[0003] Limitations of traditional methods: Traditional multi-objective allocation and trajectory planning methods, such as linear programming, integer programming, and dynamic programming, can provide effective solutions in some cases, but they often show low computational efficiency and insufficient optimization of solutions when faced with large-scale, multi-objective, and multi-constrained complex problems. These problems are usually highly nonlinear and dynamically variable, making it difficult for traditional optimization algorithms to adapt. In addition, these methods often require a lot of computing resources and time when dealing with practical problems, and it is difficult to guarantee that the global optimal solution will be found.

[0004] Application background of simulated annealing algorithm: Simulated annealing algorithm (SA) was originally proposed by Metropolis et al. as a heuristic search algorithm based on the annealing process in physics. The algorithm provides an effective search strategy for solving complex optimization problems by simulating the thermodynamic phenomena in the annealing process of materials. The core idea of ​​the SA algorithm is to allow poor solutions to be accepted during the search process to avoid falling into the local optimum and gradually approach the global optimal solution. This algorithm has shown significant advantages in solving the traveling salesman problem (TSP) and other combinatorial optimization problems.

[0005] In the field of multi-objective allocation and trajectory planning, the application of SA algorithm has gradually gained attention. The algorithm can handle the dynamics and uncertainty of the problem, and effectively explore the solution space through iterative search and adaptive adjustment to find the optimal or approximately optimal solution. However, when the existing SA algorithm is applied to multi-objective allocation and trajectory planning problems, there are still some challenges, such as the complexity of parameter setting, algorithm convergence speed and solution quality.

[0006] Potential of simulated annealing algorithm: Despite the challenges, the potential of simulated annealing algorithm in the field of multi-objective allocation and trajectory planning cannot be ignored. Through reasonable parameter adjustment and algorithm improvement, SA algorithm can effectively handle large-scale problems and improve the quality of solution and computational efficiency. In addition, the flexibility and adaptability of SA algorithm enable it to be combined with other optimization techniques to form a hybrid algorithm to further improve the optimization degree of solution and the robustness of the algorithm. Summary of the invention

[0007] In order to overcome the shortcomings of the prior art, the present invention provides a multi-objective allocation and trajectory planning method based on a simulated annealing algorithm. First, a multi-objective allocation model is established, relevant variables are defined, and an objective function is constructed to minimize the total flight cost while considering the mission execution time and fuel limit; secondly, a dual-loop model is constructed to abstract the problem into a Hamilton loop problem, including an outer Hamilton loop and an inner Hamilton loop, which are responsible for the trajectory planning between regional nodes and the trajectory planning of targets within the region, respectively; finally, a simulated annealing algorithm is used to solve the TSP model of the dual Hamilton loop, and the optimal multi-objective allocation strategy and the corresponding trajectory planning are found through control parameter setting, initial solution generation, solution transformation, Metropolis criterion application, and cooling strategy. The method of the present invention can effectively handle large-scale, multi-objective, multi-constrained complex problems, improve computational efficiency, and find solutions close to the global optimal solution, which has important practical application value and market prospects.

[0008] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0009] Step 1: Establish a multi-objective allocation model;

[0010] The key variables in the multi-objective allocation process are defined, including flight cost, mission execution time, and detention time;

[0011] Construct an objective function that comprehensively considers the cost of the regional loop, the cost of the target loop, and the total cost of completing the mission for all targets, with the goal of minimizing the total flight cost;

[0012] Step 2: Establish a dual-loop model;

[0013] The multi-objective allocation and trajectory planning problems are abstracted into Hamilton loop problems, including outer Hamilton loops and inner Hamilton loops; the outer Hamilton loop is responsible for trajectory planning between regional nodes, and the inner Hamilton loop is responsible for trajectory planning of targets within the region; through the classification of target nodes in the inner loop and the allocation of regional nodes in the outer loop, the multi-objective allocation problem is converted into a TSP problem;

[0014] Step 3: Solution based on simulated annealing algorithm;

[0015] The simulated annealing algorithm is used to solve the constructed TSP model of double Hamilton loops;

[0016] The implementation of the algorithm includes the setting of control parameters, the generation of initial solutions, the transformation of solutions to generate new solutions, the application of the Metropolis criterion, and the cooling strategy;

[0017] A multi-objective allocation strategy and corresponding trajectory planning close to the global optimum are obtained.

[0018] Preferably, the step 1 is specifically:

[0019] Step 1-1: Define variables for multiple targets as follows:

[0020]

[0021] set up represents the cost of the mth track of the nth mission, from regional node j to target j'; It is analyzed in two parts, that is, when flying between different regional nodes, it is necessary to To execute the task in the same region, you only need to calculate the straight-line distance between the two targets. The flight time is calculated by the flight speed; S represents the running time of the mission. represents the time when goal j is completed, It means that the aircraft is between two targets at a certain moment. represents the retention time of the track at target j;

[0022] Step 1-2: Select the function with the minimum total flight cost as the objective function. The objective function is composed of the cost of the regional loop, the cost of the target loop, and the total cost required to complete the task at all targets, that is:

[0023]

[0024] in, They respectively represent the target cost of passing through the i regional node when planning the mth trajectory of the nth mission, the target cost of passing through the j regional node when planning the mth trajectory of the nth mission, the cost of flying between different regional nodes, and the stay time of the trajectory at the j target.

[0025] Preferably, the step 2 is specifically:

[0026] Assume that there are multiple targets J in each regional node ii =(j i1 ,j i2 ,j i3 ,j i4 ,…,j in ), get j i ∈i; When planning and designing the route, the aircraft carrying out the mission takes the airport as the starting point and the end point to consider the guarantee and supply, thereby continuously constructing the Hamilton closed loop, and finally forming multiple Hamilton closed loops traversing the regional node i, that is, the outer Hamilton loop; for the internal region, for the target J to be attacked or reconnaissance i Perform trajectory planning, each time from J i Limited selection of J ik Perform trajectory planning and finally traverse all targets J = (j1, j2, j3, ..., j n ), so it is abstracted as an open Hamilton loop, that is, the inner Hamilton loop;

[0027] Step 2-1: Classification of inner loop target nodes;

[0028] For all target nodes J = (j1, j2, j3, ..., j n ) are classified into three categories:

[0029] a. Adjustment node: used to adjust the unreasonable aspects in the model;

[0030] b. Long-distance nodes: Long-distance nodes are constructed based on the longest flight range. If the flight route cannot reach directly between two regional nodes, a long-distance node is added in the middle for adjustment;

[0031] c. Ordinary nodes: In target allocation and route planning, ordinary nodes are nodes that are allocated as target problems;

[0032] Step 2-2: outer ring regional node allocation;

[0033] When modeling and solving the outer Hamilton loop, it is treated as a multiple traveling salesman problem (MTSP). Hamilton loops are continuously constructed. Each loop must eventually return to the starting point. By constructing a "copy node", the MTSP is converted into a TSP problem for simplified calculation and solution.

[0034] Preferably, the step 3 is specifically:

[0035] Step 3-1: Setting of control parameters: The control parameters that need to be set are cooling rate q, initial problem T0, end temperature Tend, and chain length L;

[0036] Step 3-2: Initial solution: Generate an initial solution S by random permutation. During the algorithm operation, the planned target nodes are continuously deleted until all target nodes are traversed. At this time, the stored target nodes are empty.

[0037] Step 3-3: Generate a new solution by solution transformation: Generate a new multi-target allocation strategy and the planned trajectory by transforming the current solution S1; the transformation adopted is to generate two target nodes to be exchanged by using a random number generation method, and to generate a new path, i.e., a new feasible solution S2, by using a two-neighborhood transformation method;

[0038] Step 3-4: Metropolis criterion: If the cost under the current solution S1 is f(S1), and the cost under the new solution S2 is f(S2), the cost difference is df = f(S2) - f(S1), then the Metropolis criterion is:

[0039]

[0040] If df<0, the new target allocation strategy is accepted with probability 1; otherwise, the new target allocation strategy is accepted with probability exp(-df / T).

[0041] Step 3-5: Cooling: Cooling is performed using a cooling rate q, that is, T′=qT. If T′ is less than the end temperature, the iteration is stopped and the current target allocation strategy and the planned path are output, otherwise the iteration continues;

[0042] Step 4: Get the optimal multi-objective allocation strategy and the corresponding planned trajectory.

[0043] A computer program enables a computer to execute the above-mentioned multi-objective allocation and trajectory planning method.

[0044] An electronic device comprises: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device executes the above-mentioned multi-target allocation and trajectory planning method.

[0045] A computer-readable storage medium stores a computer program, which implements the above-mentioned multi-target allocation and trajectory planning method when executed by a processor.

[0046] A chip includes: a processor, which is used to call and run a computer program from a memory, so that a device equipped with the chip executes the above-mentioned multi-target allocation and trajectory planning method.

[0047] A computer program product, the computer program product comprising a computer storage medium storing a computer program, the computer program comprising instructions executable by at least one processor, and the multi-target allocation and trajectory planning method is implemented when the instructions are executed by the at least one processor.

[0048] The beneficial effects of the present invention are as follows:

[0049] 1. Comprehensive consideration of multi-objective allocation and trajectory planning: The present invention not only considers the problem of target allocation, but also comprehensively considers trajectory planning, thereby achieving optimal allocation of resources and efficient execution of tasks.

[0050] 2. Construction of dual-loop model: By abstracting the problem into a Hamilton loop problem, the present invention provides a new perspective to deal with multi-objective allocation and trajectory planning problems, making the problem easier to solve.

[0051] 3. Optimization application of simulated annealing algorithm: The present invention improves the search efficiency and solution quality of the algorithm by optimizing the parameter setting and search strategy of the simulated annealing algorithm, so that better solutions can be found in complex multi-objective allocation and trajectory planning problems.

[0052] 4. In summary, the present invention provides an innovative multi-objective allocation and trajectory planning method, which can effectively solve complex problems in practical applications and has important practical application value and broad market prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A general map of the flight routes of each aircraft;

[0054] Figure 2 Zoom in on routes in key areas;

[0055] Figure 3 It is a schematic diagram of the double-loop model;

[0056] Figure 4 Schematic diagram of simplifying the MTSP problem to the TSP problem;

[0057] Figure 5 This is the flow chart of the simulated annealing algorithm, (a) route map of area A, (b) route map of area B, (c) route map of area C, (d) route map of area D, and (e) route map of area E. DETAILED DESCRIPTION

[0058] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0059] This invention proposes a multi-objective allocation and trajectory planning method based on simulated annealing algorithm, aiming to overcome the limitations of traditional methods, take advantage of the SA algorithm, and provide a new solution to multi-objective allocation and trajectory planning problems. This method transforms complex multi-objective allocation and trajectory planning problems into solvable optimization problems by constructing a multi-objective allocation model and a dual-loop model, and solves them using a simulated annealing algorithm, in order to achieve the optimal resource allocation and trajectory planning effect.

[0060] The present invention realizes effective allocation and trajectory planning of multiple targets by comprehensively considering the complexity and variability of the tasks. The technical solution of the present invention includes the following key steps:

[0061] Establishment of multi-objective allocation model: The present invention first defines the key variables in the multi-objective allocation process, including flight cost, mission execution time, and detention time. By constructing an objective function, the function comprehensively considers the cost of the regional loop, the cost of the target loop, and the total cost of all target missions, with the goal of minimizing the total flight cost.

[0062] Establishment of dual-loop model: The present invention abstracts the multi-objective allocation and trajectory planning problems into Hamilton loop problems, including an outer Hamilton loop and an inner Hamilton loop. The outer loop is responsible for trajectory planning between regional nodes, while the inner loop is responsible for trajectory planning of targets within the region. Through the classification of target nodes in the inner loop and the allocation of regional nodes in the outer loop, the present invention effectively transforms the complex multi-objective allocation problem into a more tractable TSP problem.

[0063] Solution based on simulated annealing algorithm: The present invention uses simulated annealing algorithm to solve the constructed TSP model of double Hamilton loop. The implementation of the algorithm includes setting of control parameters, generation of initial solution, generation of new solution by solution transformation, application of Metropolis criterion and cooling strategy. Through these steps, the present invention can find a multi-objective allocation strategy and corresponding trajectory planning close to the global optimum while ensuring computational efficiency.

[0064] Example:

[0065] The multi-objective allocation and trajectory planning method based on simulated annealing algorithm mainly includes three steps: multi-objective allocation model establishment, dual-loop model establishment, and multi-objective allocation and trajectory planning solution based on simulated annealing algorithm.

[0066] Step 1: Establish target allocation model.

[0067] First, define the variables for multiple objectives as follows:

[0068]

[0069] set up represents the cost of the mth track of the nth mission, from regional node j to target j'. In order to simplify the model, It is analyzed in two parts, that is, when flying between different regional nodes, it is necessary to Go to another regional node and then perform the task; when performing tasks in the same region, if the distance is short, you only need to calculate the straight-line distance between the two targets. The flight time is calculated by the flight speed. S represents the running time of the task (e.g. represents the time when goal j is completed, and Indicates that the aircraft is at a certain moment between two targets. This is to limit the flight time to the time required by the mission and the onboard fuel). Indicates the time that the track stays at target j.

[0070] The function with the minimum total flight cost is selected as the objective function. The objective function is composed of the cost of the regional loop, the cost of the target loop, and the total cost required to complete the task at all targets, that is:

[0071]

[0072] Assume that there are 31 regional nodes and 201 multi-targets:

[0073]

[0074] When planning multiple targets, the strike mission cannot exceed 5 hours, while the reconnaissance mission cannot exceed 8 hours, and both must be carried out within the time window of 7:00 to 19:00:

[0075]

[0076] The target must be reached within the time window when the enemy is unprepared or under-prepared:

[0077]

[0078] Based on the above considerations, the target allocation model is established as follows:

[0079]

[0080] Step 2: Establishment of dual-loop model.

[0081] The Hamilton graph is an undirected graph that goes from a specified starting point to a specified end point, passing through all other nodes only once. In graph theory, it refers to a graph containing a Hamiltonian circuit. A closed Hamiltonian path is called a Hamiltonian circuit, and a path containing all vertices in the graph is called a Hamiltonian path. For the model problem in step 1, it can be abstracted as a Hamiltonian circuit that starts from the airport and eventually needs to return to the airport starting point within a specified time.

[0082] When building the model, assume that for each regional node i, there are multiple targets J in this region. i =(j i1 ,j i2 ,j i3 ,j i4 ,…,j in ), easy to get j i ∈i, from the overall point of view, when planning and designing the route, the aircraft performing the mission needs to consider the guarantee and supply with the airport as the starting point and the end point, so as to continuously build Hamilton closed loops, and finally form multiple Hamilton closed loops traversing the regional node i, that is, the outer Hamilton loop; and for the internal region, it is necessary to target J to be attacked or detected. i Perform trajectory planning, each time from J i Limited selection of J ik Perform trajectory planning and finally traverse all targets J = (j1, j2, j3, ..., j n ), so it can be abstracted as an open Hamilton loop, that is, the inner Hamilton loop. The inner and outer loop models are as follows Figure 1 shown.

[0083] Step 1: Classification of inner loop target nodes.

[0084] For all target nodes J = (j1, j2, j3, ..., j n ), they can be roughly divided into three categories:

[0085] a. Adjustment nodes: As the name implies, adjustment nodes are used to adjust the unreasonable aspects of the model. Because this type of node has the property that multiple task nodes can be reached in a short time, whether the adjustment node is used as the previous target node (or the next target node) of a region will affect the global target allocation and route planning. Therefore, after the program is run, these adjustment nodes can be manually modified to make the program run results more reasonable.

[0086] b. Distant nodes: Distant nodes are constructed based on the longest flight range. Due to the long distance between regions, the flight route sometimes cannot reach directly between two regional nodes, so other nodes must be added in the middle for adjustment. Therefore, distant nodes are also a very important aspect that must be considered in the overall target allocation plan.

[0087] c. Ordinary nodes: Ordinary nodes are neither adjustment nodes nor long-distance nodes. In target allocation and route planning, these nodes are considered to be simple nodes that can be allocated for general target problems.

[0088] Step 2: Outer ring road regional node allocation

[0089] When modeling and solving the outer Hamilton loop, it can be regarded as a Multiple Traveling Salesman Problem (MTSP), where the traveling salesman needs to continuously construct Hamilton loops, and each loop must eventually return to the starting point. The TSP problem is an NP-hard problem, that is, a nonlinear problem that cannot be represented and calculated using polynomials, and the MTSP cannot be solved using ordinary methods. However, for this type of MTSP problem with a single starting point and closed-loop Hamilton loop, the MTSP can be converted into a TSP problem by constructing a "copy node" for simplified calculation and solution.

[0090] like Figure 2 As shown in the figure, it is very difficult to solve this MTSP problem. Using the "copy node" method, the starting point is copied. According to the number of Hamilton loops N, it is copied N-1 times. At this time, it can be converted into a normal TSP problem as shown in the right figure, which greatly simplifies the difficulty of solving the model and improves the calculation efficiency.

[0091] Finally, the inner and outer loops are integrated for multi-objective allocation and trajectory planning, and can be transformed into a TSP problem for modeling and model solving.

[0092] Step 3: Multi-objective allocation and trajectory planning solution based on simulated annealing algorithm.

[0093] The above TSP model belongs to an NP-hard problem, so it cannot be expressed by polynomials and solved by mathematical functions. Here, the heuristic algorithm of simulated annealing algorithm is used to calculate the above model.

[0094] The idea of ​​simulated annealing (SA) algorithm was first proposed by Metropolis et al. Its starting point is based on the similarity between the annealing process of solid matter in physics and general combinatorial optimization problems. SA provides an effective way and general framework for solving TSP problems that are difficult to handle with traditional methods, and gradually develops into an iterative adaptive heuristic probabilistic search algorithm. The algorithm flow chart for solving the constructed double Hamilton loop TSP model based on the simulated annealing algorithm is as follows Figure 3 shown.

[0095] For this model, the implementation of the SA algorithm mainly includes the following aspects:

[0096] (1) Setting of control parameters: The main control parameters that need to be set are cooling rate q, initial temperature T0, end temperature Tend, and chain length L.

[0097] (2) Initial solution: For the TSP problem with 201 target nodes, a random permutation method is used to generate an initial solution S. During the algorithm operation, the planned target nodes are continuously deleted until all target nodes are traversed. At this time, the stored target nodes are empty.

[0098] (3) Solution transformation to generate new solution: By transforming the current solution S1, a new multi-target allocation strategy and the planned trajectory are generated. The transformation adopted is to generate a random number method to generate the two target nodes to be exchanged, and use the two-neighborhood transformation method to generate a new path, that is, the new feasible solution S2.

[0099] (4) Metropolis criterion: If the cost under the current solution S1 is f(S1), and the cost under the new solution S2 is f(S2), the cost difference is df = f(S2) - f(S1), then the Metropolis criterion is:

[0100]

[0101] If df<0, the new target allocation strategy is accepted with probability 1; otherwise, the new target allocation strategy is accepted with probability exp(-df / T).

[0102] (5) Cooling: Cool down using the cooling rate q, that is, T′=qT. If T is less than the end temperature, stop the iteration and output the current target allocation strategy and the planned path; otherwise, continue the iteration.

[0103] Combine steps one, two and three to obtain the optimal multi-objective allocation strategy and the corresponding planned trajectory.

[0104] According to the established multi-objective allocation and trajectory planning model, and the optimization solution in the simulated annealing algorithm, the overall flight route diagram of each aircraft can be obtained as follows: Figure 4 As shown, it can be seen that the aircraft can plan a reasonable route according to the target node to achieve the purpose of completing the mission quickly.

[0105] In order to more clearly show the situation of the aircraft performing the mission in the mission area, the route map of the aircraft performing the mission near the target is enlarged as shown below: Figure 5 It can be seen that the aircraft can effectively allocate targets in each trajectory through the constructed method, and the obtained trajectory can ensure the optimal trajectory while ensuring the reachability of each target.

Claims

1. A multi-objective allocation and trajectory planning method based on simulated annealing algorithm, characterized in that: The steps include: Step 1: Establish a multi-objective allocation model; The key variables in the multi-objective allocation process are defined, including flight cost, mission execution time, and detention time; Construct an objective function that comprehensively considers the cost of the regional loop, the cost of the target loop, and the total cost of completing the mission for all targets, with the goal of minimizing the total flight cost; Step 2: Establish a dual-loop model; The multi-objective allocation and trajectory planning problems are abstracted into Hamilton loop problems, including outer Hamilton loops and inner Hamilton loops; the outer Hamilton loop is responsible for trajectory planning between regional nodes, and the inner Hamilton loop is responsible for trajectory planning of targets within the region; through the classification of target nodes in the inner loop and the allocation of regional nodes in the outer loop, the multi-objective allocation problem is converted into a TSP problem; Step 3: Solution based on simulated annealing algorithm; The simulated annealing algorithm is used to solve the constructed TSP model of double Hamilton loops; The implementation of the algorithm includes the setting of control parameters, the generation of initial solutions, the transformation of solutions to generate new solutions, the application of the Metropolis criterion, and the cooling strategy; A multi-objective allocation strategy and corresponding trajectory planning close to the global optimum are obtained.

2. The multi-objective allocation and trajectory planning method based on simulated annealing algorithm according to claim 1 is characterized in that: The step 1 is specifically as follows: Step 1-1: Define variables for multiple targets as follows: set up represents the cost of the mth track of the nth mission, from regional node j to target j'; It is analyzed in two parts, that is, when flying between different regional nodes, it is necessary to To execute the task in the same region, you only need to calculate the straight-line distance between the two targets. The flight time is calculated by the flight speed; S represents the running time of the mission. represents the time when goal j is completed, It means that the aircraft is between two targets at a certain moment. represents the retention time of the track at target j; Step 1-2: Select the function with the minimum total flight cost as the objective function. The objective function is composed of the cost of the regional loop, the cost of the target loop, and the total cost required to complete the task at all targets, that is: in, They respectively represent the target cost of passing through the i regional node when planning the mth trajectory of the nth mission, the target cost of passing through the j regional node when planning the mth trajectory of the nth mission, the cost of flying between different regional nodes, and the stay time of the trajectory at the j target.

3. The multi-objective allocation and trajectory planning method based on simulated annealing algorithm according to claim 2 is characterized in that: The step 2 is specifically as follows: Assume that there are multiple targets J in each regional node i i =(j i1 ,j i2 ,j i3 ,j i4 ,…,j in ), get j i ∈i; When planning and designing the route, the aircraft carrying out the mission takes the airport as the starting point and the end point to consider the guarantee and supply, thereby continuously constructing the Hamilton closed loop, and finally forming multiple Hamilton closed loops traversing the regional node i, that is, the outer Hamilton loop; for the internal region, for the target J to be attacked or reconnaissance i Perform trajectory planning, each time from J i Limited selection of J ik Perform trajectory planning and finally traverse all targets J = (j1, j2, j3, ..., j n ), so it is abstracted as an open Hamilton loop, that is, the inner Hamilton loop; Step 2-1: Classification of inner loop target nodes; For all target nodes J = (j1, j2, j3, ..., j n ) are classified into three categories: a. Adjustment node: used to adjust the unreasonable aspects in the model; b. Long-distance nodes: Long-distance nodes are constructed based on the longest flight range. If the flight route cannot reach directly between two regional nodes, a long-distance node is added in the middle for adjustment; c. Ordinary nodes: In target allocation and route planning, ordinary nodes are nodes that are allocated as target problems; Step 2-2: outer ring regional node allocation; When modeling and solving the outer Hamilton loop, treat it as a multiple traveling salesman problem (MTSP), and continuously construct Hamilton loops. Each loop must eventually return to the starting point. By constructing a "copy node", the MTSP is converted into a TSP problem for simplified calculation and solution.

4. The multi-objective allocation and trajectory planning method based on simulated annealing algorithm according to claim 3 is characterized in that: The step 3 is specifically as follows: Step 3-1: Setting of control parameters: The control parameters that need to be set are cooling rate q, initial problem T0, end temperature Tend, and chain length L; Step 3-2: Initial solution: Generate an initial solution S by random permutation. During the algorithm operation, the planned target nodes are continuously deleted until all target nodes are traversed. At this time, the stored target nodes are empty. Step 3-3: Generate a new solution by solution transformation: Generate a new multi-target allocation strategy and the planned trajectory by transforming the current solution S1; the transformation adopted is to generate two target nodes to be exchanged by using a random number generation method, and to generate a new path, i.e., a new feasible solution S2, by using a two-neighborhood transformation method; Step 3-4: Metropolis criterion: If the cost under the current solution S1 is f(S1), and the cost under the new solution S2 is f(S2), the cost difference is df = f(S2) - f(S1), then the Metropolis criterion is: If df<0, the new target allocation strategy is accepted with probability 1; otherwise, the new target allocation strategy is accepted with probability exp(-df / T); Step 3-5: Cooling: Cooling is performed using a cooling rate q, that is, T′=qT. If T′ is less than the end temperature, the iteration is stopped and the current target allocation strategy and the planned path are output, otherwise the iteration continues; Step 4: Get the optimal multi-objective allocation strategy and the corresponding planned trajectory.

5. A computer program, characterized in that The computer program enables a computer to execute the method according to any one of claims 1 to 4.

6. An electronic device, characterized in that: include: Processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device executes the method as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

8. A chip, characterized in that: include: A processor, configured to call and run a computer program from a memory, so that a device equipped with the chip executes a method as claimed in any one of claims 1 to 4.

9. A computer program product, characterized in that The computer program product comprises a computer storage medium storing a computer program, wherein the computer program comprises instructions executable by at least one processor, and when the instructions are executed by the at least one processor, the method according to any one of claims 1 to 4 is implemented.

Images