Satellite task planning method based on ant colony algorithm

Abstract

This invention discloses a satellite mission planning method based on ant colony optimization (ACO). The method utilizes ACO to rapidly generate suboptimal solutions, improves the quality of optimal solutions by performing local searches on the optimal solutions in each generation, and dynamically adjusts the pheromone influence factor and pheromone evaporation factor based on the iteration number and fitness rate. The satellite mission planning method based on ACO provided by this invention not only has high execution efficiency but also meets the requirements of mission planning problems in complex applications. It avoids the algorithm getting trapped in local optima too early, enhancing global search capabilities; reduces computation time, and improves the efficiency of solving large-scale problems. Furthermore, it can employ two ant colonies to handle observation and data transmission tasks separately, enabling joint optimization of these two tasks.

Description

Technical Field

[0001] This invention belongs to the field of satellite observation technology and relates to a satellite mission planning method based on ant colony algorithm. Background Technology

[0002] Earth observation satellites acquire information about the Earth's surface from space using onboard sensors, playing a crucial role in disaster prevention, weather forecasting, environmental protection, military reconnaissance, and modern agriculture. With the increasing demand for remote sensing, users are placing higher requirements on the diversification of observation targets and the breadth of observation ranges, making satellite mission planning technology a key element in the efficient utilization of satellite resources.

[0003] Satellite mission planning is essentially the process of matching and scheduling observation missions with satellite resources under complex constraints (such as time windows, satellite maneuverability, storage capacity, energy limitations, etc.). Its purpose is to develop a scientific work plan to drive satellite resources to execute missions efficiently.

[0004] Multi-satellite cooperative task planning problems are often modeled as combinatorial optimization problems. Due to their NP-hard nature, traditional exact algorithms struggle to solve them effectively. Existing techniques often employ metaheuristic algorithms, such as genetic algorithms, particle swarm optimization, and ant colony optimization. Among these, ant colony optimization (ACO) simulates ant foraging behavior, using pheromone concentration to guide the search direction and gradually converge to the optimal solution. Its positive feedback mechanism and distributed computing advantages make it particularly effective in path planning problems.

[0005] However, traditional ant colony algorithms have significant limitations in satellite mission planning applications: Easily trapped in local optima: Premature concentration of pheromones leads to a loss of algorithmic diversity; Slow convergence speed: Especially in large-scale problems, the search space grows exponentially, resulting in low algorithm efficiency; Static parameter limitations: Fixed parameters are difficult to adapt to dynamically changing environments during the search process; Separate scheduling problem: Observation tasks and data transmission tasks are usually scheduled separately, which makes it impossible to achieve overall efficiency optimization. Summary of the Invention

[0006] The technical problem solved by this invention is to provide a satellite mission planning method based on ant colony algorithm, which improves solution quality and accelerates convergence.

[0007] This invention is achieved through the following technical solution: A satellite mission planning method based on ant colony optimization includes the following operations: : Define the encoding rules for the ant colony algorithm: use numerical encoding, with the encoding length equal to the number of tasks. Each gene bit corresponds to a task, and the gene value is the time window number selected for that task; the gene value can be 1 or 0, where 0 means do not execute and 1 means execute. Initialization includes the pheromone matrix Heuristic information matrix Pheromones influencing factors Heuristic factors pheromone volatile factors Ant colony size and maximum number of iterations Ant colony algorithm parameters, including those included; Input to the ant colony algorithm: Satellite collection M represents the number of satellites; Task Collection N represents the number of tasks, and the priority attribute of each task is represented as... ; The mission and satellite visibility window is ,in Indicates the task. Indicates satellite, Indicates the start time. Indicates the end time; Construct a candidate list, which includes the available time windows for all tasks, and perform conflict detection on the time windows, deleting time windows that conflict due to constraints. The constraints include time resource constraints. If the time windows of multiple tasks overlap on the same satellite, the conflict will be resolved according to the task priority. Construct an ant path, where each ant starts from the starting node, selects the next task node based on the state transition probability, and traverses all nodes; Where the ant k in the t-th iteration is... i Node transfer j The state transition probability of a node is : in For pheromone concentration, For inspirational information, , For changes in orbital transfer speed; Given the set of task nodes that are currently allowed to be accessed, the ant can only choose from these paths; To traverse the pheromone concentration and heuristic information of task nodes; Local pheromone updates: After each ant completes path construction, the pheromones along that path are updated immediately. ; in, For the first Only ants on the path The increase in pheromones on the surface; This indicates the pheromone concentration along the path in the new iteration; : Construct a fitness function to calculate the fitness value of the solution constructed by each ant, and perform fitness evaluation; The fitness function is: in This indicates that the task has been successfully scheduled; otherwise, its value is 0. A weighting factor for the number of tasks completed, used to adjust the weight of the number of tasks completed in the overall goal; This is a weighting coefficient for the overall task weights, used to adjust the overall weight of each task in the total objective, balancing the weights of the number and priority of tasks. The preset task weight value represents the relative importance or priority of task i. The task weight value is preset based on factors including task type and level. The higher the weight, the more important the task. After each iteration, only the top-ranked ants are allowed to release pheromones for a global pheromone update. in, For the number of elite ants, The ranking weight for each elite ant; The pheromone influence factor is dynamically adjusted based on the iteration process. ; Indicates the time of the t-th iteration The value, This represents the fitness function value in the t-th iteration. This represents the logarithm of the rate of change of the fitness function; If the fitness change rate is less than the threshold for N consecutive iterations, the pheromone evaporation factor is adjusted as follows: ; Indicates the time of the t-th iteration The value; By performing a local search on the optimal solution in each generation, the quality of the optimal solution can be improved. The resulting next generation is used as a new population; and steps 7, 8, and 9 are repeated iteratively. If the maximum number of iterations is reached or the optimal solution no longer changes significantly, the iteration is terminated, and the resulting allocation scheme is the optimal scheme that satisfies the constraints. Output the optimal solution, which includes mission planning such as the satellites allocated to each mission, start time, and end time.

[0008] Furthermore, two ant colonies were used to handle the observation task and the data transmission task respectively.

[0009] Furthermore, the local search method includes swap operations and inverse transformation operations.

[0010] Compared with the prior art, the present invention has the following beneficial technical effects: The satellite mission planning method based on the ant colony algorithm provided by this invention not only has high execution efficiency, but also meets the mission planning problems of complex applications: Improve solution quality: Avoid the algorithm getting trapped in local optima too early and enhance global search capabilities; Accelerated convergence: Reduces computation time and improves the efficiency of solving large-scale problems; Adaptive optimization: The pheromone influence factor is dynamically adjusted based on the number of iterations and the fitness change rate to adapt to different stages of the search process; Integrated scheduling: Two ant colonies are used to handle observation tasks and data transmission tasks respectively, which can realize joint optimization of observation tasks and data transmission tasks. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0012] The present invention will be further described in detail below with reference to embodiments. These descriptions are for illustrative purposes only and are not intended to limit the scope of the invention.

[0013] See Figure 1 A satellite mission planning method based on ant colony optimization includes the following operations: : Define the encoding rules for the ant colony algorithm: use numerical encoding, with the encoding length equal to the number of tasks. Each gene bit corresponds to a task, and the gene value is the time window number selected for that task; the gene value can be 1 or 0, where 0 means do not execute and 1 means execute. Initialization includes the pheromone matrix Heuristic information matrix Pheromones influencing factors Heuristic factors pheromone volatile factors Ant colony size and maximum number of iterations Ant colony algorithm parameters, including those included; Input to the ant colony algorithm: Satellite collection M represents the number of satellites; Task Collection N represents the number of tasks, and the priority attribute of each task is represented as... ; The mission and satellite visibility window is ,in Indicates the task. Indicates satellite, Indicates the start time. Indicates the end time; the superscript 's' indicates the start time, and 'e' indicates the end time; Construct a candidate list, which includes the available time windows for all tasks, and perform conflict detection on the time windows, deleting time windows that conflict due to constraints. The constraints include time resource constraints. If the time windows of multiple tasks overlap on the same satellite, the conflict will be resolved according to the task priority. The task priority is set by the user. Construct an ant path, where each ant starts from the starting node, selects the next task node based on the state transition probability, and traverses all nodes; Where the ant k in the t-th iteration is... i Node transfer j The state transition probability of a node is : in For pheromone concentration, For inspirational information, , For changes in orbital transfer speed; Given the set of task nodes that are currently allowed to be accessed, the ant can only choose from these paths; To traverse the pheromone concentration and heuristic information of task nodes; Local pheromone updates: After each ant completes path construction, the pheromones along that path are updated immediately. ; in, For the first Only ants on the path The pheromone increment is initially constant. This indicates the pheromone concentration along the path in the new iteration; : Construct a fitness function to calculate the fitness value of the solution constructed by each ant, and perform fitness evaluation; The fitness function is: in This indicates that the task has been successfully scheduled; otherwise, its value is 0. A weighting factor for the number of tasks completed, used to adjust the weight of the number of tasks completed in the overall goal; This is a weighting coefficient for the overall task weights, used to adjust the overall weight of each task in the total objective, balancing the weights of the number and priority of tasks. The preset task weight value represents the relative importance or priority of task i. This preset value is based on factors including task type and level. The higher the weight, the more important the task. After each iteration, only the top-ranked ants are allowed to release pheromones for a global pheromone update. in, The number of elite ants is generally defined as the top 10% of ants in the population. The ranking weight for each elite ant; The pheromone influence factor is dynamically adjusted according to the iterative process to adapt to different stages of the search process. ; Indicates the time of the t-th iteration The value, This represents the fitness function value in the t-th iteration. This represents the logarithm of the rate of change of the fitness function; when When the value is greater than 1, increase This reinforces historical experience; when the value is less than 1, it decreases. To increase random search; If the fitness change rate is less than the threshold for N consecutive iterations, the pheromone evaporation factor is adjusted as follows: ; Indicates the time of the t-th iteration The value of pheromone; a smaller volatile factor can reduce the evaporation of pheromones, thus preserving the optimal solution during the algorithm calculation process; The quality of the optimal solution is improved by performing a local search on the optimal solution of each generation; the local search method includes swap operations and inverse transformation operations. The resulting next generation is used as a new population; and steps 7, 8, and 9 are repeated iteratively. If the maximum number of iterations is reached or the optimal solution no longer changes significantly, the iteration is terminated, and the resulting allocation scheme is the optimal scheme that satisfies the constraints. Output the optimal solution, which includes mission planning such as the satellites allocated to each mission, start time, and end time.

[0014] Furthermore, by using two ant colonies to handle the observation task and the data transmission task respectively, joint optimization of the observation task and the data transmission task can be achieved.

[0015] Table 1 shows a comparison between the traditional ant colony algorithm and the ant colony algorithm of this invention for satellite mission planning under the same constraints. It can be seen that the present invention achieves a higher mission completion rate in a shorter time.

[0016] Table 1 Comparison between the traditional ant colony algorithm and the ant colony algorithm of this invention The embodiments given above are preferred examples for implementing the present invention, and the present invention is not limited to the above embodiments. Any non-essential additions or substitutions made by those skilled in the art based on the technical features of the present invention are within the protection scope of the present invention.

Claims

1. A satellite mission planning method based on ant colony optimization, characterized in that, Includes the following operations: : Define the encoding rules for the ant colony algorithm: use numerical encoding, with the encoding length equal to the number of tasks. Each gene bit corresponds to a task, and the gene value is the time window number selected for that task; the gene value can be 1 or 0, where 0 means do not execute and 1 means execute. Initialization includes the pheromone matrix Heuristic information matrix Pheromones influencing factors Heuristic factors pheromone volatile factors Ant colony size and maximum number of iterations Ant colony algorithm parameters, including those included; Input to the ant colony algorithm: Satellite collection M represents the number of satellites; Task Collection N represents the number of tasks, and the priority attribute of each task is represented as... ; The mission and satellite visibility window is ,in Indicates the task. Indicates satellite, Indicates the start time. Indicates the end time; Construct a candidate list, which includes the available time windows for all tasks, and perform conflict detection on the time windows, deleting time windows that conflict due to constraints. The constraints include time resource constraints. If the time windows of multiple tasks overlap on the same satellite, the conflict will be resolved according to the task priority. Construct an ant path, where each ant starts from the starting node, selects the next task node based on the state transition probability, and traverses all nodes; Where the ant k in the t-th iteration is... i Node transfer j The state transition probability of a node is : in For pheromone concentration, For inspirational information, , For changes in orbital transfer speed; Given the set of task nodes that are currently allowed to be accessed, the ant can only choose from these paths; To traverse the pheromone concentration and heuristic information of task nodes; Local pheromone updates: After each ant completes path construction, the pheromones along that path are updated immediately. ; in, For the first Only ants on the path The increase in pheromones on the surface; This indicates the pheromone concentration along the path in the new iteration; : Construct a fitness function to calculate the fitness value of the solution constructed by each ant, and perform fitness evaluation; The fitness function is: in This indicates that the task has been successfully scheduled; otherwise, its value is 0. A weighting factor for the number of tasks completed, used to adjust the weight of the number of tasks completed in the overall goal; This is a weighting coefficient for the overall task weights, used to adjust the overall weight of each task in the total objective, balancing the weights of the number and priority of tasks. The preset task weight value represents the relative importance or priority of task i. The task weight value is preset based on factors including task type and level. The higher the weight, the more important the task. After each iteration, only the top-ranked ants are allowed to release pheromones for a global pheromone update. in, For the number of elite ants, The ranking weight for each elite ant; The pheromone influence factor is dynamically adjusted based on the iteration process. ； Indicates the time of the t-th iteration The value, This represents the fitness function value in the t-th iteration. This represents the logarithm of the rate of change of the fitness function; If the fitness change rate is less than the threshold for N consecutive iterations, the pheromone evaporation factor is adjusted as follows: ; Indicates the time of the t-th iteration The value; : Improve the quality of the optimal solution by performing a local search on the optimal solution of each generation; The resulting next generation is used as a new population; and steps 7, 8, and 9 are repeated iteratively. If the maximum number of iterations is reached or the optimal solution no longer changes significantly, the iteration is terminated, and the resulting allocation scheme is the optimal scheme that satisfies the constraints. Output the optimal solution, which includes mission planning such as the satellites allocated to each mission, start time, and end time.

2. The satellite mission planning method based on ant colony algorithm as described in claim 1, characterized in that, Two ant colonies were used to handle the observation task and the data transmission task respectively.

3. The satellite mission planning method based on ant colony algorithm as described in claim 1, characterized in that, The top 10% of ants in the population are considered elite ants.

4. The satellite mission planning method based on ant colony algorithm as described in claim 1, characterized in that, The local search method includes swap operations and inverse transformation operations.

Images