Energy supply and task path optimization method for cluster operation of mobile platform

Abstract

The invention provides an energy supply and task path optimization method for cluster operation of a mobile platform, and relates to the technical field of multi-agent task allocation and path planning. According to the method, information of a carrier layer, a platform layer and a task layer is coded into individuals, and the information of the carrier layer is a transport vehicle parking place set including a path starting place set and a path ending place set; the platform layer information is a matching relationship between the unmanned platform and the path starting place and the path ending place; the task layer information comprises a matching relationship between the unmanned platform and various tasks including a battery replacement task, and a task execution sequence; initializing individuals according to the sequence of a carrier layer, a platform layer and a task layer, and generating an initial population; and performing population optimization by using a culture gene algorithm, and outputting a multi-agent cluster operation energy supply and task path collaborative planning scheme. Energy supply and task execution paths of the unmanned platform can be planned at the same time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of multi-agent task allocation and path planning, and particularly relates to an energy supply and task path optimization method for mobile platform cluster operation. BACKGROUND

[0002] With the rapid development of intelligent technology, multi-unmanned platforms have been widely used in many fields due to many advantages, such as environmental ecological monitoring, commodity logistics transportation, power facility inspection, smart city security, disaster emergency rescue, etc. In multi-unmanned systems, the collaborative operation efficiency is effectively improved through multi-body capability cooperation, multi-resource integration and multi-source information sharing. However, the diversity of task types, the difference of platform capabilities and the cooperation of heterogeneous platforms, etc. have led to the complication of multi-unmanned system task allocation strategy, the enhancement of multi-dimensional system parameter coupling, and the severe test on the improvement of multi-unmanned system collaborative efficiency.

[0003] The present application relates to a multi-type task demand scenario: multi-transport vehicle carrying unmanned platforms, one transport vehicle can carry one or more than one unmanned platform, and the carrying quantity is related to the loading capacity of the transport vehicle. The transport vehicle transports the unmanned platform to the candidate site, the unmanned platform leaves the transport vehicle to execute the task, and in the task execution process, the unmanned platform can go to the candidate site for energy supply, and after completing the task, it arrives at the candidate site and gets on the vehicle, and is taken away by the transport vehicle. The unmanned platform can be heterogeneous, and the heterogeneous unmanned platform is not tightly coupled with the transport vehicle. The unmanned platform can be carried back by any transport vehicle, and can also be supplied with energy at any transport vehicle.

[0004] In the above task demand scenario, the task allocation and path planning problem of energy-limited multi-agent cluster operation belongs to the field of combinatorial optimization, which focuses on solving the following problems: in a given task environment, how to allocate and coordinate multi-unmanned platforms with different functions by combining task demand and agent performance, design energy supply and return strategy, and unmanned platform-task adaptation criteria, and reasonably plan the task execution path, while meeting the capability limit, task demand and other related constraints, to improve the cluster operation efficiency as much as possible, including but not limited to minimizing the task completion time, reducing the overall operation cost or loss, and improving the task completion quality and effect, so that the dispersed individual behavior is integrated into a highly coordinated intelligent collaborative whole through systematic planning, and intelligence emerges at the system level.

[0005] Considering that the unmanned platform is driven by electricity, its endurance time cannot be comparable with that of a fuel vehicle, therefore, in order to effectively deal with the endurance limit and improve the operation efficiency, the transport vehicle with the moving ability in the above-mentioned scenario can realize the position deployment and energy supply of the unmanned platform. Therefore, it is necessary to optimize the site selection of the transport vehicle to complete the position deployment of the unmanned platform before the task execution, so as to ensure that it is deployed in the shortest time; during the task execution, the transport vehicle supplies energy for the unmanned platform to ensure the sustainability of the system; after the task is completed, the transport vehicle carries the unmanned platform back to the base or goes to the subsequent task area. The above operation process ensures the continuity and efficiency of the task execution, maximizes the operation coverage range of the unmanned platform under the limited endurance, and enhances the group operation efficiency.

[0006] The above-mentioned problem scenario involves multiple collaborative relationships among transport vehicles, between transport vehicles and unmanned platforms, and between heterogeneous unmanned platforms, there is a multi-level coupling among tasks, platforms and vehicles, the parking position of the transport vehicle, the platform-task adaptation relationship, and the cooperative decision of the unmanned platform task execution path are needed, therefore, the multi-agent cluster operation optimization problem in this scenario is highly similar to the location-routing problem (LRP). The so-called location-routing problem (LRP) integrates three aspects of decision-making: Determining a subset of candidate sites (facility location problem), corresponding to the determination of the position set of the transport vehicle loading (unloading) the unmanned platform in the present application, the energy supply and position deployment of multiple unmanned platforms are realized by optimizing the parking position of the transport vehicle; Task and platform correspondence (task assignment problem), corresponding to the assignment of tasks to platforms in the present application, the platform-task adaptation relationship is optimized to fully exert the ability advantage of the unmanned platform; Task execution path optimization (vehicle routing problem), corresponding to the optimization of the task execution path of each unmanned platform in the present application, to minimize the task completion time and improve the task execution efficiency.

[0007] At present, the optimization objectives of LRP are usually to minimize transportation cost, maximize system revenue, and maximize customer satisfaction, etc. The solution algorithms for location-routing problem are mainly divided into exact method and intelligent optimization algorithm. The former mainly includes column generation, branch and bound, dynamic programming, etc. Since the exact algorithm can only solve small-scale problems within an acceptable time range, it is difficult to obtain a feasible solution for large-scale LRP within a limited time. Therefore, many scholars have designed various intelligent optimization algorithms with excellent performance based on the characteristics of LRP problem. The algorithm framework involved mainly includes genetic algorithm, simulated annealing algorithm, tabu search algorithm, ant colony algorithm, and neighborhood search algorithm, etc. Through the introduction of heuristic rules, parallel computing and hybrid strategy, intelligent optimization algorithm shows strong global search ability and computing efficiency in solving large-scale LRP. However, through research, it is found that the current research on location-routing problem has the following limitations when applied to the task allocation and path planning problem of multi-type task demand scenario of energy-limited multi-mobile platform cluster operation: (1) Few mobile carriers and vehicle inter-station operations are introduced The research background of existing location-routing problem is mainly logistics transportation or home medical care, and the start and end positions of vehicles are warehouses or hospitals without mobility. Moreover, the vehicles that execute tasks need to return to the starting position, so there is a lack of vehicle inter-station operation and mobile carrier collaborative planning. In actual application, such as disaster relief or urban security scenarios, unmanned platforms that execute tasks end their journey in a place different from the starting position can effectively reduce energy consumption during the journey, and the cooperation of multiple transport vehicles and multiple unmanned platforms can improve the efficiency of task execution under limited resources. This problem has important research value and practical significance. However, the introduction of unmanned platform inter-station operation and mobile carrier collaborative mechanism will lead to complex coupling between the task sets corresponding to each mobile carrier, making it difficult to effectively solve the problem by task grouping and decomposition. Therefore, it is necessary to design intelligent optimization algorithms for efficient solution.

[0008] (2) Poor exploration ability in high-dimensional complex solution space Existing algorithms face limitations such as dimension disaster, easy to fall into local optimum, and contradiction between computing efficiency and real-time performance when exploring complex solution space, making it difficult to efficiently solve high-dimensional and strongly coupled complex problems. Especially in the multi-mobile platform cluster operation scenario, the introduction of transport carrier collaborative mechanism improves the efficiency of task execution, but also intensifies the multi-level constraint nesting of task-resource-location relationship, further increasing the difficulty of problem solving and reducing the practical application effect of existing algorithms in multi-agent cluster operation field. Therefore, it is necessary to design targeted algorithms based on problem characteristics to effectively solve the energy supply and task path collaborative optimization problem of multi-agent cluster operation in multi-type task demand scenario, improve the exploration ability in complex solution space, and significantly enhance the effectiveness and intelligence of multi-agent cluster operation. SUMMARY

[0009] Therefore, the application provides a mobile platform cluster operation energy supply and task path optimization method, which models the task planning problem of energy-limited multi-agent cluster operation in multiple types of task demand scenarios as a site-path planning problem with movable carriers and energy supply, reasonably selects the stopping positions of transport vehicles in combination with task demand and unmanned platform performance, and plans the energy supply of unmanned platforms and the task execution path. While meeting the capability limitations, task demand and other related constraints, the task completion time is minimized as much as possible.

[0010] To solve the above technical problems, the application is implemented as follows.

[0011] A mobile platform cluster operation energy supply and task path optimization method, comprising: Step one: determining the information of candidate sites, transport vehicles, unmanned platforms and tasks; Step two: modeling the task planning problem, and the planning problem constraints include: (1) the starting point positions of the transport vehicle stops constitute the path starting site set The transport vehicle stopping points where the unmanned platforms return after completing various tasks constitute the path termination site set , The elements in the path starting site set and the path termination site set are also optional positions for the unmanned platforms to perform the battery replacement task; the elements in the path starting site set and the path termination site set are selected from the candidate sites; the contents of the path starting site set and the path termination site set are the same or different; (2) the unmanned platforms cannot consume all the energy while driving, and the unmanned platforms are fully charged after each energy supply; Step three: encoding the carrier layer, platform layer and task layer information into individuals; the carrier layer information is the transport vehicle stopping site set, including the path starting site set and the path termination site set ; the platform layer information is the matching relationship between the unmanned platforms and the path starting sites and the path termination sites; the task layer information includes the matching relationship between the unmanned platforms and various tasks including the battery replacement task, and the task execution sequence; Step four: initializing the individuals in the order of the carrier layer, the platform layer and the task layer to generate an initial population; Step five: using the cultural gene algorithm to optimize the population, and outputting a multi-agent cluster operation energy supply and task path collaborative planning scheme.

[0012] Preferably, in the step one, the information of the candidate sites, the transport vehicles, the unmanned platforms and the tasks is: Determine the location and number of candidate locations, the number and loading capacity of transport vehicles, the number of unmanned platforms, the mission execution capability matrix and endurance time vector, and the location and requirements of the mission. Elements in the task execution capability matrix of unmanned platforms for Type of unmanned platform performs tasks per unit time The amount completed.

[0013] Preferably, in step two, the overall optimization objective of the task planning problem is to minimize the cluster job completion time.

[0014] Preferably, in step two, the constraints of the planning problem further include: (3) Loading capacity constraints of transport vehicles: A transport vehicle can carry a maximum of An unmanned platform; (4) Unmanned platform Arrive at the mission Time spent The following conditions must be met:

[0015] In the formula, For type unmanned platform The speed of travel; For unmanned platforms In the region and The driving distance between; M The total number of tasks. L The total number of candidate locations. This represents the total number of unmanned platforms. (5) The task requirements cannot be broken down, and each unmanned platform cannot travel back and forth to the task location; they must leave after the task is completed. (6) Each transport vehicle can park independently, meaning that at most one transport vehicle can park at each candidate location.

[0016] Preferably, step four, which initializes individuals in the order of carrier layer, platform layer, and task layer to generate the initial population, specifically includes: Step 41: Initialize the carrier layer and determine the starting path set for the unmanned platform. and path termination set For each task, determine the nearest candidate location, and for each candidate location, count the number of times that candidate location is the nearest candidate point for the task. The number of times Reflects the density of task distribution around the candidate location; based on the number of times Determine the global selection probability of candidate locations ,frequency The higher the probability, the greater the global selection probability. The larger the value; based on the global selection probability. The starting set of paths is determined by using a round-robin method. and path termination set Elements in; Step 42: Perform platform-level initialization and decide on the allocation of candidate locations for unmanned platforms, including the allocation of path start and end locations for unmanned platforms. Based on the set of starting and ending points of the path determined in task and step 41. The distance between each element is used to group all tasks, and each task group corresponds to a path start point. The total time required for each unmanned platform to be assigned to each task group to execute the corresponding task is calculated. The unmanned platform-task group matching is performed according to the task assignment scheme with the minimum total time required. The path start point corresponding to the task group is used as the path start point for matching the unmanned platform. Based on the mission execution status of each unmanned platform, the endurance constraints of the unmanned platform, and the loading capacity constraints of the transport vehicle, the distance greedy principle is used to determine the path termination location matched by the unmanned platform. Step 43: Perform task layer initialization, and determine the matching tasks and task execution paths for the unmanned platform: A weighted ranking value for tasks is constructed using the transfer cost of unmanned platforms and the task execution time. When matching tasks, random numbers are generated. If random number Less than the set first threshold If a candidate task is randomly selected from the task group as a pending follow-up task, otherwise the task with the smallest weighted ranking value is selected as the pending follow-up task. Calculate whether the remaining energy after completing the pending follow-up task can reach the nearest transport vehicle. If it can, the pending follow-up task is confirmed as a follow-up task. If not, determine to add a battery swapping task and return to the nearest transport vehicle to perform the battery swapping operation. For each specified follow-up task, update the weighted ranking value of the remaining tasks and then perform follow-up task allocation until all tasks are allocated and the task execution path is obtained.

[0017] Preferably, in step 43, the weighted ranking value for constructing the task using the unmanned platform transfer cost and task execution time is: Let the population size be To set intervals generate Weight coefficients For the first When initializing an individual, select the first... Weight coefficients As a weighting coefficient for the cost of relocating unmanned platforms, as the weighted coefficient of the task execution time; the weighted ranking value of the i-th candidate task in the task group is x is the transfer cost of the candidate task is the task execution time of the candidate task

[0018] Preferably, the step five, when performing the population optimization operation, the individual crossover includes the carrier layer, the platform layer and the task layer three levels of crossover selection; Carrier layer crossover selection: for the parent individuals and , the path starting point set and the path ending point set of the offspring individuals are determined by crossover selection in the following way: the to-be-determined or is recorded as ; when is determined, the path starting point set of each unmanned platform involved in the parent individual is recorded as , and the path starting point set of each unmanned platform involved in the parent individual is recorded as ; when is determined, the path ending point set of each unmanned platform involved in the parent individual is recorded as , and the path ending point set of each unmanned platform involved in the parent individual is recorded as ; and The common element set of , and is ; if the number of elements in is the total number of transport vehicles P , then the set ; if the elements in are less than P , then the set , wherein the set is selected by the roulette method in , and represents the union set; Platform layer crossover selection: combined with the allocation of unmanned platforms-candidate locations in the parent, the path starting point set ​​​​​​The corresponding unmanned platform number of each candidate location element in the parent, and constitute a set ; wherein, is a candidate location p The corresponding unmanned platform number set; Let be The number of unmanned platforms in is the maximum loading capacity of the transport vehicle to the unmanned platform; according to , , The priority order of each candidate location is processed; For candidate locations, the unmanned platform number in the child from the candidate location is the same as the parent; For candidate locations, sort unmanned platforms according to the task execution time, and select the first unmanned platform with the smallest execution time and no multi-candidate location conflict to assign to the current candidate location of the child; If there is only one candidate location, which meets , first assign the unmanned platform element to the current candidate location without multi-candidate location conflict, and if there is still a remaining unmanned platform after assignment, assign the remaining unmanned platform to the current candidate location of the child; If there are multiple locations, all of which meet , first assign the unmanned platform element without multi-candidate location conflict; if there is still a remaining unmanned platform after assignment, assign the remaining unmanned platform in combination with the task execution time and the transport vehicle loading capacity constraint; In the platform layer cross selection, update the elements in the set PL after completing the assignment of each candidate location; Task layer cross selection: according to the location assigned to the unmanned platform after the platform layer cross selection, determine the task set in the parent whose location is the starting location or the ending location of the path, and take the union of the two task sets as the advantage gene set of the current unmanned platform; use the advantage gene set to assign the task execution sequence to the unmanned platform.

[0019] Preferably, in the task layer cross selection, using the advantage gene set to assign the task execution sequence to the unmanned platform includes two stages: First stage, select the first task executed by each unmanned platform: generate a random number , if the random number is less than a set second threshold then randomly select the unmanned platform in turn, and select a task in the advantage gene set corresponding to the unmanned platform to add to the execution path of the unmanned platform as the first task; if the random number is not less than the second threshold value then calculate the execution cost of the selectable tasks in the advantage gene set corresponding to each unmanned platform, and add the task corresponding to the minimum execution cost to the task execution path of the corresponding unmanned platform as the first task; In the second stage, the subsequent tasks executed by each unmanned platform are selected: using the "first complete first select" strategy, the unmanned platform that completes the current task first enters the active state selects a subsequent task in the corresponding advantage gene set and adds it to the task execution path; if the remaining energy of the unmanned platform is insufficient to execute the tasks in the advantage gene set, it is determined whether other unexecuted tasks can be executed, if yes, the executable tasks are added to the task execution path, otherwise, a battery replacement task location is selected based on the candidate locations in the offspring path termination location set and the task selection process of the first stage and the second stage is continued to be repeated; after each task selection, the elements in the advantage gene set of each unmanned platform and the elements in the unexecuted task set are updated.

[0020] Preferably, when the population optimization operation is performed in step five, a multi-granularity neighborhood search is used for local search of the solution space in local search, including path adjustment based on task point exchange, path adjustment based on candidate location exchange, path adjustment based on task segment exchange, and path adjustment based on inter-path task reallocation mechanism.

[0021] Preferably, the path adjustment based on the inter-path task reallocation mechanism includes: calculating the inter-path distance corresponding to each task execution sequence, selecting the two paths with the shortest distance and ; randomly deleting a set proportion of tasks in the two paths and ; respectively calculating the task execution time increment corresponding to the insertion of each deleted task in the deleted task set into different positions of the current all paths; selecting the deleted task with the minimum time increment to insert into the corresponding position; updating the insertable positions, the remaining tasks in the deleted task set, and the corresponding task execution time increment, and continuing to select the deleted task with the minimum task execution time increment to insert into the corresponding position; After the insertion operation of all deleted tasks is completed, the energy supplement location is adjusted or added for the unmanned platform that does not meet the endurance time constraint.

[0022] Beneficial effects: The present application is directed to the task planning problem of energy-limited multi-agent cluster operation in multi-type task demand scenarios, which is modeled as a site-path planning problem with movable carriers and energy supply, the stop position of the transport vehicle is reasonably selected in combination with the task demand and the performance of the unmanned platform, and the energy supply of the unmanned platform and the task execution path are planned, while meeting the capability limit, task demand and other related constraints, the task completion time is minimized as much as possible, and then a mobile platform cluster operation energy supply and task path optimization method is disclosed. The contributions of the present application mainly reflect in the following aspects: (1) The present application analyzes and describes multiple complex coupling constraints, fully considers the nonlinear, strong coupling and other characteristics in actual application, and constructs a site-path planning problem model with movable carriers and energy supply.

[0023] (2) In an preferred scheme, in order to effectively consider the quality and diversity of the initial population, a hierarchical heuristic-assisted population initialization rule is designed, which involves carrier layer, platform layer and task layer collaborative decision-making, and comprehensively considers task distribution, platform execution capability and workload balance; In order to effectively improve the planning scheme quality, the unmanned platform transfer cost and task matching degree are introduced, and the high-quality initial population is quickly generated; (3) In an preferred scheme, in order to improve the population quality and convergence speed, a crossover operator with vehicle site exploration and superior gene rearrangement is proposed, the offspring individuals are generated by selecting and rearranging the excellent genes of the parent individuals, and exploring new task sequence and site selection scheme, which significantly improves the search efficiency of the algorithm for complex solution space, speeds up the convergence speed of the algorithm, and enhances the exploration ability of the population for potential high-quality solutions; (4) In an preferred scheme, combined with the characteristics of the problem, a multi-granularity neighborhood search structure is designed, including path adjustment based on task point exchange, path adjustment based on candidate location exchange, path adjustment based on task segment exchange and path adjustment based on task reallocation mechanism between paths, which realizes the efficient cooperation of global and local search. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The present application is directed to the task planning problem of energy-limited multi-agent cluster operation in multi-type task demand scenarios, which is modeled as a site-path planning problem with movable carriers and energy supply, the stop position of the transport vehicle is reasonably selected in combination with the task demand and the performance of the unmanned platform, and the energy supply of the unmanned platform and the task execution path are planned, while meeting the capability limit, task demand and other related constraints, the task completion time is minimized as much as possible, and then a mobile platform cluster operation energy supply and task path optimization method is disclosed. The contributions of the present application mainly reflect in the following aspects:

[0025] Figure 2 The present application is directed to the task planning problem of energy-limited multi-agent cluster operation in multi-type task demand scenarios, which is modeled as a site-path planning problem with movable carriers and energy supply, the stop position of the transport vehicle is reasonably selected in combination with the task demand and the performance of the unmanned platform, and the energy supply of the unmanned platform and the task execution path are planned, while meeting the capability limit, task demand and other related constraints, the task completion time is minimized as much as possible, and then a mobile platform cluster operation energy supply and task path optimization method is disclosed. The contributions of the present application mainly reflect in the following aspects:

[0026] Figure 3 The present application is directed to the task planning problem of energy-limited multi-agent cluster operation in multi-type task demand scenarios, which is modeled as a site-path planning problem with movable carriers and energy supply, the stop position of the transport vehicle is reasonably selected in combination with the task demand and the performance of the unmanned platform, and the energy supply of the unmanned platform and the task execution path are planned, while meeting the capability limit, task demand and other related constraints, the task completion time is minimized as much as possible, and then a mobile platform cluster operation energy supply and task path optimization method is disclosed. The contributions of the present application mainly reflect in the following aspects:

[0027] Figure 4An individual encoding diagram for the CLRP-MCER problem under the condition that the transport vehicle is not fully loaded.

[0028] Figure 5 An individual decoding method diagram for the CLRP-MCER problem.

[0029] Figure 6 A population initialization rule diagram assisted by hierarchical heuristics.

[0030] Figure 7 A crossover operator diagram with vehicle location exploration and dominant gene rearrangement.

[0031] Figure 8 A candidate site-unmanned platform pointing diagram.

[0032] Figure 9 A path task reassignment mechanism diagram. DETAILED DESCRIPTION

[0033] The present application models the task allocation and path planning problem of energy-limited multi-agent cluster operation in multi-type task demand scenarios as a location-routing problem variant, i.e., a location-routing problem with mobile carriers and energy replenishment, which is referred to as CLRP-MCER (Complete location-routing problem with mobile carriers and energy replenishment) in the present application. A method and device for energy replenishment and task path collaborative optimization for heterogeneous mobile platform cluster operation are proposed in the framework of Memetic algorithm, with the introduction of domain knowledge.

[0034] For the abstract description of the above problem, see Figure 1 The unmanned platform is loaded on the transport vehicle and transported to the task area by the transport vehicle. The transport vehicle unloads the unmanned platform, and the unmanned platform completes the tasks in its task set according to the planned path. During task execution, if the energy of the unmanned platform is low, it moves to the transport vehicle stop and performs battery replacement operation. The unmanned platform completes all tasks and also returns to the transport vehicle. Given some candidate sites, the starting point and stop point of the transport vehicle are selected from the candidate sites.

[0035] See Figure 1The dashed line in the figure is the path of the unmanned platform before and after energy replenishment. Due to the introduction of the inter-station operation required for battery replacement energy replenishment, the coupling relationship between the task sets corresponding to each candidate site occurs, the mobility of the transport vehicle causes the elements in the candidate site set as the starting point and the end point of the path to be different, and as the number of transport vehicles and candidate sites increases, the solution space of the CLRP-MCER problem of the application increases rapidly. At the same time, as the capacity of the transport vehicle increases, more complex coupling relationships will occur between the task sets corresponding to each candidate site. The above situations make it more difficult to effectively solve the CLRP-MCER problem, and the algorithm can effectively solve the above problems. Compared with the prior art, the algorithm has a significant advantage in task completion time.

[0036] The application will be described in detail below with reference to the accompanying drawings.

[0037] Step one: determine the information of candidate sites, transport vehicles, unmanned platforms and tasks.

[0038] This step includes determining the location and number of candidate sites, the number and loading capacity of transport vehicles, the number of unmanned platforms, the task execution capability matrix and the endurance time vector of the unmanned platforms, and the location and demand of tasks. Among them, the element in the task execution capability matrix of the unmanned platform is . The type of unmanned platform is the amount of task completion per unit time.

[0039] This step also determines the algorithm parameters, including population size, index evaluation times, etc.

[0040] Step two: build a task planning problem model.

[0041] A known number of heterogeneous unmanned platforms are deployed by transport vehicles to specified locations to execute tasks that are scattered and have different demands, and return to the transport vehicle to replenish energy within the endurance time or are carried by the transport vehicle to the subsequent task area after completing the task. Considering that different types of unmanned platforms have different capabilities to execute different types of tasks, it is necessary to reasonably optimize the task execution path of each unmanned platform, including the starting and ending sites of the path and the energy replenishment site, to minimize the completion time of all tasks as much as possible. The constraints and assumptions of the planning problem are as follows: (1) Constraints of the planning problem Each unmanned platform has an endurance time constraint and needs to return to the transport vehicle stop to replenish energy within the endurance time. It is assumed that the power of the unmanned platform is when full; Each transport vehicle has an upper limit of loading capacity ; Task demands are indivisible, and each UAV cannot go back and forth at the task, and must leave after the task is completed; Each candidate site can at most accommodate one transport vehicle.

[0042] (2) Planning problem assumptions Assume that each UAV has the same volume, i.e., the number of platforms that each transport vehicle can carry is not affected by the platform type; Assume that each UAV has a fixed endurance time, which is only related to the platform type and is not affected by the task performed; Allow more than one UAV to simultaneously replenish energy at the same transport vehicle, as the energy replenishment time (battery replacement) is much smaller than the travel time between areas and the task execution time, and is therefore ignored; Assume that the transport vehicle goes to the next location after completing all UAV unloading (if the transport vehicle needs to change the stop location), and can arrive at the designated location before the UAV arrives.

[0043] (3) UAV capability description and modeling ① UAV task execution capability The time required for the same UAV to perform different tasks is different, affected by the task type and scale; the time required for the same task to be performed by different UAVs is different, affected by the UAV execution capability, as shown in equation (1.1), where, is the completion amount matrix of each task by multiple UAVs per unit time, is the amount of task performed by a UAV of type , there are types of UAVs, M and tasks: (1.1) ② UAV endurance capability During cluster operation, the time consumption of the UAV mainly comes from two aspects: travel between areas and task execution. Since the UAVs are heterogeneous, the transport vehicle site selection, transport vehicle-UAV matching, UAV-task selection and execution path planning are mutually influenced and coupled, and need to be jointly optimized.

[0044] (4) Task demand modeling Assume that the time for a UAV to arrive at a task is , and the total demand at the task is , then the following relationship holds: (1.2) where, is the number of tasks, is the number of candidate locations, is the number of unmanned platforms, is the decision variable, if the unmanned platform moves from region to region , , , , , , . is the completion time of task . and denote candidate locations and .

[0045] (5) Modeling of the heterogeneous unmanned platform task planning problem The CLRP-MCER problem can be defined on an undirected graph , where, is the set of points, including the set of tasks and the set of candidate locations , which can be used as the location for the unmanned platforms to be unloaded (or loaded) by the transport vehicles, and also as the location for the unmanned platforms to be recharged. is the set of edges, whose elements are the travel distances of the unmanned platforms between regions (including task regions and candidate locations). The set of transport vehicles is , the number of which is and , and the set of unmanned platforms is .

[0046] The mathematical model of the CLRP-MCER problem is as follows: (1.3) (1.4) (1.5) (1.6) (1.7) (1.8) (1.9) (1.10) (1.11) (1.12) (1.13) (1.14) (1.15) Formula (1.3) is the optimization objective of the problem. f This indicates minimizing the cluster job completion time. For the task The completion time.

[0047] Formulas (1.4)-(1.5) indicate that each transport vehicle stops independently, meaning that at most one transport vehicle can stop at each candidate location. Here, the set of transport vehicle stopping locations at the beginning of the operation is called the path starting location set. The set of locations where transport vehicles return after completing various tasks is called the path termination point set. , The elements in the list also serve as optional locations for the unmanned platform to perform battery swapping tasks. (Set of path start and end points) Set of path termination locations The elements in the data may differ, but the quantity is always the same. This indicates the mobility of the transport vehicle. ( ) is the decision variable. If the transport vehicles are at the beginning (end) of the cluster operation... Located at candidate site ,but ( ),on the contrary, ( ); This represents the number of candidate locations.

[0048] Formula (1.6) indicates that there is a loading capacity constraint on transport vehicles; a transport vehicle can carry a maximum of [number] loads. One unmanned platform, among which ( ) is the decision variable. If the unmanned platform is at the beginning (end) of the cluster operation... By transport vehicle Loading, then ( ),on the contrary, ( ).

[0049] Formula (1.7) represents the candidate locations as the start and end points of the mission execution paths for each unmanned platform. and This is the set of candidate locations corresponding to two site selections for multiple transport vehicles. and These represent unmanned platforms. From candidate locations Departure and move to mission The incident, unmanned platform Completed the mission Return to candidate location If the above events occur, then the corresponding... =1 ( =1). The second formula in formula (1.7) indicates that the path termination point may be visited multiple times because there will be a battery swapping task, so the unmanned platform may return to the transport vehicle multiple times.

[0050] Formulas (1.8)-(1.10) stipulate that the unmanned platform must not consume all its energy during the journey, and that the unmanned platform must be fully charged after each energy replenishment. , and Unmanned platforms The rate of power consumption, maximum power, and area reached. The remaining battery power at that time For unmanned platforms in the region ( or )and ( or The driving distance between ( ). Indicates unmanned platform Arrive at the mission Time; As a decision variable, if the platform is unmanned By region ( or Move to area ( or ),but ;on the contrary, .

[0051] Formula (1.11) demonstrates the unmanned platform By region Move to area In this situation, unmanned platforms Arrive at the mission time The calculation method. For type unmanned platform Movement speed.

[0052] Constraint (1.12) represents the task completion requirement and the task requirement cannot be split, that is, each unmanned platform cannot go back and forth at the task, and needs to leave after the task is completed; Equation (1.13) is a flow balance constraint, and respectively represent the unmanned platform moving from the region to the task , and moving to the region after completing the task ; Equation (1.14) is a sub-loop elimination constraint, and are the indices of the tasks and in the planned path.

[0053] Equation (1.15) represents the decision variable as a Boolean variable.

[0054] The description of the decision variable in the CLRP-MCER problem is shown in Table 1.

[0055] Table 1 Symbol and description of decision variable

[0056] The following steps three to five are the enhanced Memetic algorithm designed by the application for hierarchical initialization and advantageous gene rearrangement.

[0057] Memetic algorithm (MA) combines the strong global search ability of genetic algorithm and the efficient local search ability of local search, and is a powerful algorithm framework. It has been successfully applied to various complex optimization problems and has achieved good results. Unlike genetic algorithm, mutation operation in MA algorithm is guided by problem knowledge, which can effectively improve the exploration ability and search efficiency of the algorithm for complex solution space. Therefore, combined with the characteristics of the CLRP-MCER problem, based on the Memetic algorithm framework, the application proposes an enhanced Memetic algorithm with hierarchical initialization and advantageous gene rearrangement to efficiently solve the CLRP-MCER problem.

[0058] Step three: design CLRP-MCER problem encoding and decoding.

[0059] ① Encoding method: The CLRP-MCER problem encoding includes carrier layer, platform layer and task layer information, and the encoding forms the individual of the population in the subsequent steps. The carrier layer information is the set of vehicle stopping places, including the set of path starting places and the set of path ending places The platform layer information is the matching relationship between the unmanned platform and the candidate sites (path starting site and path ending site); the task layer information includes the matching relationship between the unmanned platform and the task, and the task execution sequence. The task here includes the battery replacement task and various tasks.

[0060] In this embodiment, the individual is represented by using the encoding mode as shown in Figure 3 and Figure 4 The capital letter in the example represents the candidate site number, the Roman numeral represents the unmanned platform number, and the Arabic numeral represents the task number. Figures 3-4 The following is introduced: In Figure 3 , the number of transport vehicles is 3, the number of unmanned platforms is 6, and the loading capacity of the transport vehicles is 2, so each long-distance transport platform is in a full load state; among them, candidate sites A / B / C constitute the starting set , and candidate sites B / C / E constitute the ending set ; the unmanned platforms II and IV start from candidate site A, respectively sequentially execute tasks 1→13→15→14 and 5→6→7, and then return to candidate sites C and B. Element 0 represents energy supply. In this example, the unmanned platform II supplements energy at candidate site C before executing task 15 due to the endurance time limit. The unmanned platforms I and VI start from candidate site B, respectively sequentially execute tasks 9→4→8 and 11→10→20→19, and then return to candidate sites B and E. Similarly, the unmanned platform VI supplements energy at candidate site B before executing task 20. Similarly, the unmanned platforms III and V start from candidate site C, respectively sequentially execute tasks 12→3→2 and 16→17→18, and then return to candidate sites C and E.

[0061] In Figure 4 , the number of transport vehicles is 3, the number of unmanned platforms is 5, and the loading capacity of the transport vehicles is 2, so each long-distance transport platform is in a full load state; among them, candidate sites A / B / C constitute the starting set , and candidate sites B / C / E constitute the ending set ; the task execution and energy supply of the unmanned platforms are similar to Figure 3 , which will not be described here; since the transport vehicle located at candidate site B is not fully loaded, the empty position is completed by element NA.

[0062] ②Decoding mode: The individual is decoded by using an event triggering mechanism, that is, when the unmanned platform completes the current task, the unmanned platform becomes active and selects the subsequent task to be executed according to the encoding scheme. When the encoding scheme in the task layer is “0”, the unmanned platform moves to the corresponding candidate site to supplement energy. Figure 5 The encoding and decoding of the individual are respectively shown in Figures 3-4The diagram shows the decoding of each encoding scheme. Each unmanned platform executes tasks sequentially according to the given encoding order. Task 14 and Task 4 are the last tasks completed in two different scenarios. The dashed lines represent the round-trip paths of the unmanned platform before and after energy replenishment, and the solid lines represent the paths of the unmanned platform leaving / returning to the candidate location and executing the task.

[0063] Step 4: Generate a high-quality initial population using hierarchical heuristic-assisted initialization rules: Initialize individuals in the order of carrier layer, platform layer, and task layer to generate the initial population.

[0064] See Figure 2 The second step involves considering various factors, including candidate location selection, location-platform matching, platform-task allocation, and execution path planning. Therefore, heuristic methods designed considering only a single factor cannot generate high-quality execution plans for computational examples with different characteristics. Based on the above analysis, a layered heuristic-assisted initialization rule (LHa-IR) was designed, comprehensively considering task distribution, platform execution capabilities, and workload balancing. It mainly includes three decision-making aspects: the carrier layer, the platform layer, and the task layer. Figure 6 As shown, the details are as follows: Step 41: Carrier Layer Initialization: Decision-making on the set of parking locations for transport vehicles, i.e., the starting path set for the unmanned platform. and path termination set .

[0065] To minimize energy consumption during unmanned platform movement between areas and shorten the overall operation time of the cluster, candidate locations with a relatively dense distribution of surrounding tasks should be prioritized to form the initial path set. and path termination set Therefore, for each task, the nearest candidate location is determined, and for each candidate location, the number of times the candidate location is the nearest candidate point for the task is counted. The number of times Reflects the density of task distribution around the candidate location; based on the number of times Determine the global selection probability of candidate locations ,frequency The higher the probability, the greater the global selection probability. The larger the probability, the better; then, based on the global selection probability of each candidate location. The method of round betting is used to determine the results. and The elements in the design aim to maximize the quality of the initial individuals while satisfying the need for diversity.

[0066] Global selection probability of each candidate location According to formula (4.1), wherein, The distance between the candidate site The number of the nearest tasks.

[0067] (4.1) Step 42: Platform layer initialization: decision of unmanned platform-candidate site matching, including unmanned platform-path starting site and unmanned platform-path ending site matching.

[0068] Selection of unmanned platform path starting site: according to the set of tasks and step 41 The distance between each element in the set groups the tasks, and each task group corresponds to a path starting site. The total time required for each unmanned platform to be assigned to each task group to perform the corresponding task is calculated (at this time, the endurance time constraint and the distance between tasks are ignored). Assume that each task grouping is numbered as Then is the unmanned platform performing the task grouping The total time of each task in the set, is the decision variable, if the unmanned platform performs each task in the task grouping , then , otherwise, . Among them, the number of unmanned platforms loaded by each transport vehicle cannot exceed the upper limit, and each unmanned platform can only be assigned to one task partition.

[0069] Then according to formula (4.2), the multi-unmanned platform grouping assignment should follow the principle of minimum total time required, that is, the unmanned platform-task group matching is performed according to the task assignment scheme with the minimum total time required, and the path starting site corresponding to each task group is used as the path starting site of the matching of the unmanned platform.

[0070] (4.2) (4.3) (4.4) Selection of unmanned platform path ending site: according to the task execution, combined with the endurance constraint of the unmanned platform and the loading capacity constraint of the transport vehicle, the distance greedy principle is adopted. A specific operation is as follows: when the unmanned platform decides the path ending site, the principle of "first complete first select" is adopted, that is, the unmanned platform that completes the last task of the path first selects the path ending position. Considering that the transport vehicle has a loading capacity constraint, therefore, each time the unmanned platform selects the path ending site, within the allowable range of the transport vehicle capacity, the distance greedy principle is used to match the path ending site for the selectable candidate site.

[0071] Step 43: task layer initialization: decision of matching tasks and task execution path of unmanned platform.

[0072] In order to effectively improve the quality of the planning scheme, the transfer cost TC of the unmanned platform and the ability matching degree AM are introduced, wherein the transfer cost TC refers to the time required for the unmanned platform to reach the next task, in order to shorten the operation completion time as much as possible, the task execution path of each unmanned platform needs to be optimized, therefore the moving distance between the unmanned platform and the task can be used as a reference factor when selecting the task; and the ability matching degree AM mainly includes the task execution time, which refers to the time required for the unmanned platform to execute the task.

[0073] In the decision process, the active unmanned platform selects the subsequent executed task or the candidate location for returning, the priority of the candidate task is obtained by weighting and sorting the transfer cost TC of the unmanned platform and the ability matching degree AM of of the unmanned platform, so as to better balance the transfer cost and demand of different tasks.

[0074] Specifically, in the task layer initialization stage, the transfer cost TC of the unmanned platform and the ability matching degree AM value expressed by the task execution time are sorted in ascending order according to the weighted values. When matching the task, a random number is generated, if the random number is less than the set first threshold , a candidate task is randomly selected from the task group as the pending subsequent task, otherwise the task with the smallest weighted sorting value is selected as the pending subsequent task; when the unmanned platform selects each subsequent task, it is calculated whether the remaining energy after executing the pending subsequent task can reach the nearest transport vehicle, if yes, the pending subsequent task is confirmed as the subsequent task, if no, a battery replacement task is added and the unmanned platform returns to the nearest transport vehicle for battery replacement. After a specified subsequent task is selected, the weighted sorting values of the remaining tasks are updated, and the subsequent task or the battery replacement task execution position is allocated, until all task allocation is completed and the task execution path is obtained.

[0075] In a preferred embodiment, the weighted sorting value is set as follows: Suppose the population size is , and the interval is set as , then weight coefficients are generated, the interval has a size of , and the corresponding weight coefficient is used when generating each individual, for example, when the th individual is generated, the weight coefficient is , then the weighted coefficient of the transfer cost TC is , and the weighted coefficient of the ability matching degree AM is , that is, when the individual At that time, candidate tasks weighted value The calculation method is as follows: Then, according to Arrange the tasks in ascending order to determine the order of each task. When matching tasks, if the random number does not meet the conditions, then select... The smallest task is assigned as the next task.

[0076] Step 5: Utilize the cultural gene algorithm to optimize the population and output a collaborative planning scheme for energy replenishment and task paths for multi-agent cluster operations.

[0077] The inventive point in this step lies in the design of a crossover operator with vehicle location exploration and dominant gene rearrangement, as well as the use of multi-granularity neighborhood search in local search.

[0078] See Figure 2 This step specifically includes the following steps: Step 51: Use a crossover operator with vehicle location exploration and dominant gene rearrangement to generate offspring individuals and explore the complex solution space.

[0079] Offspring generation is a crucial step in evolutionary computation. Effective gene inheritance and spatial exploration can enhance population diversity while ensuring individual quality. Therefore, it is necessary to design efficient offspring generation rules based on problem characteristics to obtain high-quality offspring individuals.

[0080] Based on the characteristics of the CLRP-MCER problem, this invention proposes a crossover operator with vehicle location exploration and superior gene rearrangement (VleSGr-C). It should be noted that VleSGr-C does not directly inherit gene sequences from the parent generation, but rather selects and rearranges superior genes from the parent generation, and explores new task sequences and location schemes.

[0081] VleSGr-C is mainly discussed from three aspects: carrier layer, platform layer, and task layer. Now, let's take... Figure 7 For example, the following is an introduction, in which and Each represents a parent individual: Carrier layer cross selection: Taking the cross selection of the path start location set as an example, based on whether the path start positions are the same, the parent generation is selected. The set of starting and ending points of the path and father generation The set of starting and ending points of the path The elements in the database are divided into two types: and , respectively, represent the union of the common element set in and and the elements unique to each of them. If , represents the number of elements in the set, then the initial set of child individuals is ; if , , where is selected in the set using the roulette method, and the selection probability of each element in the set is proportional to and satisfies . The determination of each element in is similar to the above process.

[0082] As shown in Figure 7 , , and , so two elements need to be selected in by the roulette method, together with to form . Assuming that the selected elements are , then . Similarly, .

[0083] Platform layer cross selection: combined with the allocation of unmanned platforms-candidate sites in the parent generation, the corresponding unmanned platform number of each candidate site element in the parent generation is obtained in the set , and the set is formed, where is the set of unmanned platform numbers corresponding to the candidate site p .

[0084] Let be the number of unmanned platforms in , and be the maximum loading capacity of the transport vehicle to the unmanned platform. In order to inherit the excellent genes of the parent generation as much as possible, the priority of the matching results of each candidate site and unmanned platform is: first , then , and finally .

[0085] For the candidate site , the unmanned platform number starting from the candidate site in the child generation is the same as that in the parent generation; For the candidate site , sort the unmanned platforms according to the task execution time, and select the first The unmanned platform with the minimum total execution time and no multi-candidate location conflict is assigned to the current candidate location of the offspring. The multi-candidate location non-conflict means that the unmanned platform already assigned to one candidate location cannot be assigned to other unmanned platforms.

[0086] For example, in Figure 7 , the set of elements in the parent corresponds to the set of unmanned platform numbers of the candidate location . The optional platform number for the candidate location is .

[0087] If there is only one candidate location that meets , the unmanned platform element without multi-candidate location conflict is first assigned to the current candidate location; if there are still remaining unmanned platforms after the assignment, the remaining unmanned platforms are all assigned to the current candidate location of the offspring.

[0088] If there are multiple locations that all meet , the unmanned platform element without multi-candidate location conflict is first assigned; if there are still remaining unmanned platforms after the assignment, the remaining unmanned platforms are assigned in combination with the task execution time and the transport vehicle loading capacity constraints, as shown in Figure 8 .

[0089] It should be noted that after the assignment of each candidate location is completed, the elements in the set Figure 8 are updated in the manner shown in .

[0090] Task layer cross selection: the advantage genes of the parent are combined, and the advantage gene sequence is rearranged, so as to effectively inherit the excellent genes of the parent while improving the quality of the offspring individual. The advantage genes of each unmanned platform are the union of the task numbers corresponding to the matching candidate locations in the parent. For example, the unmanned platform starts from the candidate location in the offspring , while the task sets corresponding to the candidate locations as the starting point or ending point of the path in the parent and are and respectively, then the advantage gene set is the union of the two, that is .

[0091] The advantage gene set is used to assign the task execution sequence to the unmanned platform. In a preferred embodiment, the assignment of the path of the offspring individual, i.e. the task execution sequence, is divided into two stages: In the first stage, a first task performed by each unmanned platform is selected: in this stage, path construction is initiated, and a random number is generated If , then the unmanned platforms are randomly selected in turn, and a task is selected from the corresponding set of dominant genes of the unmanned platform and added to the execution path of the unmanned platform as the first task; otherwise, the execution cost of the selectable tasks in the corresponding set of dominant genes of each unmanned platform is calculated, which can be calculated by summing the moving time and the task execution time; and the task corresponding to the minimum execution cost is added to the execution path of the corresponding unmanned platform as the first task. In the second stage, subsequent tasks performed by each unmanned platform are selected: in this stage, each unmanned platform has an execution task, and a "first complete first selection" strategy is adopted, that is, the unmanned platform that first completes the current task enters an active state, selects a subsequent task from the corresponding set of dominant genes, and adds it to the task execution path. If the remaining energy of the unmanned platform is insufficient to execute the tasks in the set of dominant genes, it is determined whether other unexecuted tasks can be executed, if so, the executable task is added to the task execution path, otherwise, a battery replacement task location is selected from the candidate locations in the set , and added to the task execution path, and the task selection process of the above two stages is repeatedly continued. After each task selection, the elements in the set of dominant genes of each unmanned platform and the elements in the set of unexecuted tasks are updated.

[0092] A specified number of offspring are obtained, and step 52 is performed.

[0093] Step 52: Local search of the solution space is performed using multi-granularity neighborhood search.

[0094] A single neighborhood structure is easy to make the algorithm fall into a local optimum, and multiple neighborhood structures can effectively enhance the exploration ability of the algorithm, therefore, in the algorithm proposed in the present application, multi-granularity neighborhood search methods are designed from different angles, including path adjustment based on task point exchange, path adjustment based on candidate location exchange, path adjustment based on task segment exchange, and path adjustment based on task reallocation mechanism between paths.

[0095] ① Path adjustment based on task point exchange: given a solution , two tasks and are randomly selected on a task execution path, and the positions of the two tasks are exchanged, and energy supplement positions are added or adjusted for unmanned platforms that do not meet the endurance time constraint.

[0096] ② Path adjustment based on candidate location exchange: a roulette wheel selection method is used, and the candidate locations in the sets and Add a candidate location. Due to limitations on the number of transport vehicles, the collection... and Each candidate location must be removed again using a roulette wheel selection method. The paths to the removed candidate locations (starting point, ending point, or refueling location) are then adjusted, and refueling locations are adjusted or added for unmanned platforms that do not meet the endurance constraints.

[0097] ③ Path adjustment based on task fragment swapping: given a solution Randomly select two task execution paths, and choose any two tasks. and The gene fragments were exchanged, and energy replenishment locations were adjusted or added for unmanned platforms that did not meet the endurance constraints.

[0098] ④ Path adjustment based on inter-path task reallocation mechanism: Considering that two paths with smaller spatial distances are more likely to undergo effective task adjustment, calculate the distance between each path and select the two closest paths. and ; on two paths and A certain percentage of tasks are randomly deleted; then a task redistribution mechanism between paths is used to repair each path.

[0099] The distance between each path can be calculated using the following formula (5.1). Where, and Representing paths and The number of tasks.

[0100] (5.1) like Figure 9 As shown, the path adjustment method based on the inter-path task reallocation mechanism is as follows: First, calculate the task execution time increment for each task in the deleted task set inserted into different positions along all paths, i.e., the difference between the distance moved before and after task insertion and the task execution time. Then, select the task with the lowest cost to insert into the corresponding position. Finally, update the insertable positions, the remaining tasks in the deleted task set, and their corresponding task execution time increments, and continue selecting the next deleted task with the lowest task execution time increment to insert into the corresponding position. For example... Figure 9 As shown, inserting task number 4 into position 3 of the path of unmanned platform II yields the minimum time increment. After completing the insertion of all deleted tasks, adjust or add energy replenishment positions for unmanned platforms that do not meet the endurance time constraint.

[0101] Step 53: Update the population and the best individuals using a biased population evaluation mechanism while maintaining population diversity.

[0102] Step 54: judge whether the algorithm end condition is met, if not, return to step 51; if yes, output the multi-agent cluster operation energy supply and task path planning scheme, that is, the site selection path planning scheme of the transport vehicle-unmanned platform.

[0103] So far, the present process ends.

[0104] The specific improvements of the energy supply and task path optimization method for the mobile platform cluster operation based on the differences in the overall planning scene and coupling relationship are as follows: A hierarchical heuristic-assisted population initialization rule is designed, the task distribution, the unmanned platform capability and the endurance constraint are comprehensively considered, the transfer cost and the task matching degree are introduced, and the rapid generation of the high-quality initial population is realized.

[0105] A crossover operator with vehicle site selection exploration and superior gene rearrangement is proposed, the excellent parent genes are selected and rearranged, and the new gene sequence and site selection scheme are explored, and the search efficiency of the algorithm for the complex solution space is improved.

[0106] A multi-granularity neighborhood search structure is adopted, the cooperation of the global and local search is realized, and the exploration ability of the algorithm for the solution space is effectively improved.

[0107] The above specific embodiments only describe the design principles of the present application, and the shapes and names of the components in the description can be different and are not limited. Therefore, the person skilled in the art of the present application can modify or replace the technical solutions described in the foregoing embodiments; and these modifications and replacements do not deviate from the purpose and technical solutions of the present application, and should all belong to the protection scope of the present application.

Claims

1. A method for energy replenishment and task path optimization in mobile platform cluster operations, characterized in that, include: Step 1: Determine the candidate location, transport vehicle, unmanned platform, and mission information; Step 2: Model the task planning problem, with planning constraints including: The starting points of the transport vehicles constitute the set of route start points. ; The destinations of the transport vehicles that return after completing various tasks form the path termination points. This set of elements also serves as the optional locations for the unmanned platform to perform battery swapping tasks; the set of path start and end points. and the set of path termination points Elements in the list are selected from candidate locations; the set of path start locations. and the set of path termination points The elements in the text are the same or different; The unmanned platform must not consume all its energy while driving, and it must be fully charged after each energy replenishment. Step 3: Encode the information from the carrier layer, platform layer, and task layer into individuals; where the carrier layer information is the set of transport vehicle stopping locations, including the set of route start and end points. and the set of path termination points Platform-level information includes the matching relationship between the unmanned platform and the starting and ending points of the path; task-level information includes the matching relationship between the unmanned platform and various tasks, including battery swapping tasks, as well as the task execution sequence. Step 4: Initialize individuals in the order of carrier layer, platform layer, and task layer to generate the initial population; Step 5: Perform population optimization and output a collaborative planning scheme for energy replenishment and task path for cluster operations.

2. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 1, characterized in that, In step one, the information for determining the candidate location, transport vehicle, unmanned platform, and mission is as follows: Determine the location and number of candidate locations, the number and loading capacity of transport vehicles, the number of unmanned platforms, the mission execution capability matrix and endurance time vector, and the location and requirements of the mission. Elements in the task execution capability matrix of unmanned platforms for Type of unmanned platform performs tasks per unit time The amount completed.

3. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 1, characterized in that, In step two, the overall optimization objective of the task planning problem is to minimize the cluster job completion time.

4. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 1, characterized in that, In step two, the planning constraints further include: Loading capacity constraints for transport vehicles: A transport vehicle can carry a maximum of [number] vehicles. An unmanned platform; Unmanned platform Arrive at the mission Time spent The following conditions must be met: In the formula, For type unmanned platform The speed of travel; For unmanned platforms In the region and The driving distance between; M The total number of tasks. L The total number of candidate locations. This represents the total number of unmanned platforms. The task requirements cannot be broken down, and each unmanned platform cannot travel back and forth to the task location; they must leave after the task is completed. Each transport vehicle can park independently, meaning that at most one transport vehicle can park at each candidate location.

5. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 1, characterized in that, Step four, which initializes individuals in the order of carrier layer, platform layer, and task layer to generate the initial population, specifically includes: Step 41: Initialize the carrier layer and determine the starting path set for the unmanned platform. and path termination set For each task, determine the nearest candidate location, and for each candidate location, count the number of times that candidate location is the nearest candidate point for the task. The number of times Reflects the density of task distribution around the candidate location; based on the number of times Determine the global selection probability of candidate locations ,frequency The higher the probability, the greater the global selection probability. The larger the value; based on the global selection probability. The starting set of paths is determined by using a round-robin method. and path termination set Elements in; Step 42: Perform platform-level initialization and decide on the allocation of candidate locations for unmanned platforms, including the allocation of path start and end locations for unmanned platforms. Based on the set of starting and ending points of the path determined in task and step 41. The distance between each element is used to group all tasks, and each task group corresponds to a path start point. The total time required for each unmanned platform to be assigned to each task group to execute the corresponding task is calculated. The unmanned platform-task group matching is performed according to the task assignment scheme with the minimum total time required. The path start point corresponding to the task group is used as the path start point for matching the unmanned platform. Based on the mission execution status of each unmanned platform, the endurance constraints of the unmanned platform, and the loading capacity constraints of the transport vehicle, the distance greedy principle is used to determine the path termination location matched by the unmanned platform. Step 43: Perform task layer initialization, and determine the matching tasks and task execution paths for the unmanned platform: A weighted ranking value for tasks is constructed using the transfer cost of unmanned platforms and the task execution time. When matching tasks, random numbers are generated. If random number Less than the set first threshold If a candidate task is randomly selected from the task group as a pending follow-up task, otherwise the task with the smallest weighted ranking value is selected as the pending follow-up task. Calculate whether the remaining energy after completing the pending follow-up task can reach the nearest transport vehicle. If it can, the pending follow-up task is confirmed as a follow-up task. If not, determine to add a battery swapping task and return to the nearest transport vehicle to perform the battery swapping operation. For each specified follow-up task, update the weighted ranking value of the remaining tasks and then perform follow-up task allocation until all tasks are allocated and the task execution path is obtained.

6. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 5, characterized in that, In step 43, the weighted ranking value for constructing the task using the unmanned platform transfer cost and task execution time is: Let the population size be To set intervals generate Weight coefficients For the first When initializing an individual, select the first... Weight coefficients As a weighting coefficient for the cost of relocating unmanned platforms, As a weighting coefficient for task execution time; then the first task in the task group x Candidate tasks Weighted sort value for: ,in For candidate tasks The cost of transfer, For candidate tasks Task execution time.

7. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 5, characterized in that, Step five utilizes the cultural gene algorithm for population optimization. During population optimization, individual crossover involves crossover selection at three levels: carrier layer, platform layer, and task layer. Carrier layer crossover selection: for parent individuals and The set of starting and ending points of the offspring individuals is determined through cross-selection. and the set of path termination points The method is: to determine the or Recorded as ; Sure At that time, the parental individual The set of starting and ending points of the paths of the various unmanned platforms involved is denoted as . Parental individuals The set of starting and ending points of the paths of the various unmanned platforms involved is denoted as . ;Sure At that time, the parental individual The set of path termination points for all the unmanned platforms involved is denoted as Parental individuals The set of path termination points for all the unmanned platforms involved is denoted as ; and The set of common elements in is , and The union of the unique elements of the two is ;if The number of elements is the total number of transport vehicles. P Then the set ;like The element in is less than P Then the set , where set exist The selection method used in the middle is the round betting method. This represents taking the union of sets; Platform-level cross-selection: Combining the allocation of unmanned platforms and candidate locations in the parent generation, the set of path start locations for offspring individuals is derived. Each candidate location element in the middle corresponds to the unmanned platform number in the parent generation, and forms a set. ;in, Candidate locations p The corresponding set of unmanned platform IDs; set up for The number of unmanned platforms in China The maximum load capacity of the transport vehicle on the unmanned platform; according to , , Candidate locations are processed in order of priority. for Candidate locations, offspring from candidate locations The unmanned platform that departed had the same serial number as its parent. for Candidate locations, based on mission execution time The unmanned platforms were sorted and selected from the top ones. The unmanned platform with the shortest execution time and no conflicts between multiple candidate locations is assigned to the current candidate location of the offspring. If only one candidate location exists, it meets the requirements. First, assign unmanned platform elements that do not conflict with multiple candidate locations to the current candidate location. If there are still unmanned platforms remaining after assignment, assign all remaining unmanned platforms to the current candidate location of the offspring. If multiple locations exist, all of which meet the criteria First, allocate unmanned platform elements that do not conflict with multiple candidate locations; if there are still unmanned platforms remaining after allocation, assign the remaining unmanned platforms based on the task execution time and the loading capacity constraints of the transport vehicles. In the platform-level cross-selection process, the set is updated after each candidate location is assigned. PL Elements in; Task-level cross-selection: Based on the location assigned to the unmanned platform after platform-level cross-selection, determine the set of tasks in the parent generation that use that location as the starting or ending point of the path. The union of the two task sets is taken as the dominant gene set of the current unmanned platform. The dominant gene set is used to assign task execution sequences to the unmanned platform.

8. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 7, characterized in that, In the task-layer cross-selection process, allocating task execution sequences to the unmanned platform using a set of advantageous genes involves two stages: In the first phase, the first task to be performed by each unmanned platform is to generate random numbers. If the random number Less than the set second threshold Then, unmanned platforms are randomly selected sequentially, and a task is chosen from the set of advantageous genes corresponding to the unmanned platform and added to the execution path of that unmanned platform as the first task; if the random number Not less than the second threshold Then, calculate the execution cost of the optional tasks in the set of advantageous genes corresponding to each unmanned platform, and add the task with the minimum execution cost to the task execution path of the corresponding unmanned platform as the first task; In the second phase, the subsequent tasks for each unmanned platform are selected: a "first-come, first-served" strategy is adopted, allowing the unmanned platform that first completes its current task and enters an active state to select a subsequent task from its corresponding set of advantageous genes and add it to the task execution path; if the remaining energy of an unmanned platform is insufficient to execute the tasks in its set of advantageous genes, it continues to determine whether other unexecuted tasks can be executed. If so, the executable tasks are added to the task execution path; otherwise, the subsequent tasks are selected based on the set of offspring path termination locations. The candidate locations are selected and the battery swapping task locations are added to the task execution path, and the task selection process of the first and second stages is repeated. After each task selection, the elements in the set of advantages of each unmanned platform and the elements in the set of unexecuted tasks are updated.

9. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 1, characterized in that, Step five utilizes the cultural gene algorithm for population optimization. In the local search, multi-granularity neighborhood search is employed to conduct a local search of the solution space, including path adjustment based on task point exchange, path adjustment based on candidate location exchange, path adjustment based on task fragment exchange, and path adjustment based on inter-path task reallocation mechanism.

10. The energy replenishment and task path optimization method for mobile platform cluster operations as described in claim 9, characterized in that, The path adjustment based on the inter-path task reallocation mechanism includes: Calculate the distance between paths corresponding to each task execution sequence, and select the two paths with the shortest distance. and ; Two paths and Randomly delete a set percentage of tasks; Calculate the task execution time increment for each deleted task in the deleted task set when inserted into different positions in all current paths; select the deleted task with the smallest time increment and insert it into the corresponding position. Update the insertion position, the remaining tasks in the deleted task set, and the corresponding task execution time increments, and continue to select the next deleted task with the smallest task execution time increment to insert into the corresponding position; After completing the insertion operation of all deleted tasks, adjust or add energy replenishment locations for unmanned platforms that do not meet the endurance constraints.

Images