Heterogeneous carrier cluster online task allocation NHHO algorithm and system
By improving the NHHO algorithm and combining the spiral search of the Harris Eagle and Whale algorithms, the problems of the HHO algorithm being prone to getting trapped in local optima and having poor search performance in heterogeneous carrier cluster task allocation are solved, and more efficient task allocation and resource utilization are achieved.
Patent Information
- Application Number
- CN202410953559.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-16
AI Technical Summary
The HHO algorithm is prone to getting trapped in local optima, has poor performance in the early stages of the search, has an insignificant convergence effect of the fitness function, struggles to cope with dynamic environmental changes, lacks comprehensive consideration of task priorities and target threats, and is difficult to determine payload configuration and usage timing, resulting in low task allocation efficiency and insufficient resource utilization in heterogeneous carrier clusters.
An improved NHHO algorithm is adopted, which combines the Harris Eagle algorithm with a spiral update mechanism and contract net protocol to enhance global search capabilities. The spiral search method of the whale algorithm is used to improve search performance and dynamically adjust task allocation to adapt to real-time environmental changes.
It significantly improves task allocation efficiency, enhances the algorithm's global search capability, reduces the occurrence of local optima, improves the accuracy and flexibility of task allocation, and ensures the rational use of resources.
Smart Images

Figure CN119011507A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of task allocation, and in particular relates to an NHHO algorithm and system for online task allocation of heterogeneous carrier clusters. Background Art
[0002] The task allocation of heterogeneous carrier clusters is to assign appropriate operating tasks to each carrier based on the number, type, designated target and different payloads of heterogeneous carriers in the cluster. Among them, the focus of the task allocation of heterogeneous carrier clusters is to plan a scientific action plan for each unmanned carrier based on the actual situation of the mission scope environment, combined with factors such as the performance, fuel consumption, task priority and target threat of the unmanned carrier. At the same time, in order to ensure that the heterogeneous carrier cluster achieves the best performance in the mission scope environment, the task allocation needs to reasonably formulate the payload configuration and usage time point.
[0003] In the context of heterogeneous carrier cluster task allocation, existing technologies mainly face the following problems and defects:
[0004] 1) The HHO algorithm is prone to fall into local optimal solutions:
[0005] Problem: The HHO (Harris Hawks Optimization) algorithm is prone to falling into a local optimal solution during task allocation, which makes it impossible to find a global optimal solution. This problem is particularly prominent in complex task allocation environments, affecting the overall performance of heterogeneous carrier clusters.
[0006] Impact: This results in inefficient task allocation, insufficient resource utilization, and inability to fully leverage the advantages of heterogeneous carrier clusters.
[0007] 2) Poor search performance in the early stages of the search:
[0008] Problem: The HHO algorithm has poor search performance in the early stages of the search, and it is difficult to quickly find high-quality solutions. This makes the efficiency and effect of task allocation low at the beginning of the task, affecting the overall progress of the task.
[0009] Impact: This results in poor initial task allocation results, requiring multiple adjustments and optimizations, which prolongs the task planning time.
[0010] 3) The fitness function convergence effect under the algorithm program is not obvious:
[0011] Problem: Compared with genetic algorithm and particle swarm algorithm, the convergence effect of fitness function of HHO algorithm is not obvious when the program is running. This means that it is difficult for HHO algorithm to significantly improve the quality of solution during the iteration process, which affects the superiority of the final task allocation solution.
[0012] Impact: The fitness function converges slowly, which results in a long optimization process, increased computational costs, and unsatisfactory optimization results.
[0013] 4) Failure to fully consider the dynamic changes in the task scope environment:
[0014] Problem: In the process of task allocation, existing technologies often fail to fully consider the dynamic changes in the task scope environment and lack the ability to adapt to real-time changes. This poses a challenge to the task execution of heterogeneous carrier clusters in complex environments.
[0015] Impact: The task allocation plan cannot adapt to environmental changes in a timely manner, affecting the flexibility and effectiveness of task execution.
[0016] 5) Insufficient consideration of mission priorities and target threats:
[0017] Problem: Existing technologies often fail to give adequate consideration to task priorities and target threats when allocating tasks, resulting in the inability to fully optimize task allocation plans.
[0018] Impact: As a result, high-priority tasks and high-threat targets do not receive sufficient attention, affecting the overall safety and effectiveness of mission execution.
[0019] 6) Difficulty in reasonably formulating load configuration and usage time points:
[0020] Problem: It is difficult to reasonably formulate load configuration and usage time points during task allocation. Existing technologies have limited support and optimization capabilities in this regard.
[0021] Impact: Improper use of payloads leads to resource waste or mission failure, affecting the overall performance of the heterogeneous carrier cluster.
[0022] In the field of task allocation for heterogeneous carrier clusters, existing technologies mainly face problems such as the HHO algorithm being prone to fall into local optimality, poor performance in the initial stage of search, unclear convergence of fitness function, difficulty in coping with dynamic environmental changes, insufficient comprehensive consideration of task priorities and target threats, and difficulty in setting load configuration and usage time points. These problems and defects limit the overall performance of heterogeneous carrier clusters, and there is an urgent need for improvement and enhancement in algorithm optimization and system design. Summary of the invention
[0023] In view of the problems existing in the prior art, the present invention provides an NHHO algorithm for online task allocation in heterogeneous carrier clusters.
[0024] The present invention is implemented as follows: an NHHO algorithm for online task allocation of heterogeneous carrier clusters includes:
[0025] Step 1: Input the heterogeneous carrier cluster and the relevant data of the three types of targets, and determine the initialization parameters to be set, including population size, maximum number of iterations, switching probability, jump intensity, etc.
[0026] Step 2, initialize the population according to the adversarial environment;
[0027] Step 3: Apply the Harris Eagle algorithm to introduce the spiral update mechanism to calculate the standard function value of each heterogeneous carrier cluster position, determine the fitness value of the individual, and obtain the current optimal function value and corresponding position;
[0028] Step 4, update the specific location of the 3-category target and the enemy's 1-category target;
[0029] Step 5: Dynamically allocate tasks based on heterogeneous carrier clusters and Class 1 target discovery conditions;
[0030] Step 6, determine whether the effective distance reaches the standard, if not, return to step 3;
[0031] Step 7, attacking Class 3 targets and Class 1 targets;
[0032] Step 8: If the maximum number of iterations is not reached or the target number of completions is insufficient, return to step 3; otherwise, proceed to the next step;
[0033] Step 9: Output the fitness and required time of task allocation.
[0034] Another object of the present invention is to provide a heterogeneous carrier cluster online task allocation system comprising:
[0035] The initialization module is used to input the heterogeneous carrier cluster and the relevant data of the three types of targets, determine the initialization parameters to be set, including the population size, the maximum number of iterations, the switching probability, the jump intensity, etc.; initialize the population according to the adversarial environment;
[0036] The calculation module is used to apply the Harris Eagle algorithm to introduce the spiral update mechanism calculation, calculate the standard function value of each heterogeneous carrier cluster position, determine the individual fitness value, obtain the current optimal function value and the corresponding position; update the specific position of the three types of target positions and the enemy type 1 target;
[0037] Allocation module, used for dynamic task allocation based on heterogeneous carrier clusters and Class 1 target discovery;
[0038] The judgment module is used to judge whether the action distance reaches the standard.
[0039] The output module is used to output the fitness and required time of task allocation.
[0040] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the NHHO algorithm for online task allocation of heterogeneous carrier clusters.
[0041] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the NHHO algorithm for online task allocation of heterogeneous carrier clusters.
[0042] Another object of the present invention is to provide an information data processing terminal, which is used to implement the heterogeneous carrier cluster online task allocation system.
[0043] In combination with the above technical solutions and the technical problems solved, please analyze the advantages and positive effects of the technical solutions to be protected by the present invention from the following aspects:
[0044] First, the present invention improves the HHO algorithm, and uses the contract network protocol in the online task allocation method to improve the Harris Eagle algorithm to prevent the algorithm from reaching the local optimum too early, thereby improving the convergence degree, and introduces the spiral search method of the whale algorithm to enhance the search performance of the algorithm in the early stage of the search. For different initial conditions defined in Chapter 3, the program is run using the improved algorithm, and the results are analyzed in depth. The results of this analysis clearly demonstrate the feasibility and effectiveness of the improved algorithm.
[0045] Second, the technical solution of the present invention solves the technical problems that people have been eager to solve but have never been able to solve successfully:
[0046] From the experimental results, the NHHO algorithm and system for online task allocation of heterogeneous carrier clusters proposed in the present invention are suitable for online task allocation of heterogeneous carrier clusters. The NHHO algorithm significantly improves the efficiency of task allocation and effectively guides the population to evolve towards the direction of the global optimal solution, thereby enhancing the global search capability of the algorithm. In addition, the improved measures adopted improve the optimization performance of the algorithm in the initial stage and reduce the probability of falling into the local optimal solution, which proves that the improved algorithm has a substantial improvement over the traditional algorithm. Therefore, the present invention solves the problem of online task allocation of heterogeneous carrier clusters to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 It is a flow chart of the NHHO algorithm for online task allocation of heterogeneous carrier clusters provided by an embodiment of the present invention.
[0048] Figure 2It is a structural block diagram of an online task allocation system for a heterogeneous carrier cluster provided by an embodiment of the present invention.
[0049] Figure 3 It is a structural diagram of the Harris Hawk algorithm provided by an embodiment of the present invention.
[0050] Figure 4 This is a comparison chart of different algorithms provided by an embodiment of the present invention, with the number of carriers and targets unchanged, the population size being 20, and 200 iterations.
[0051] Figure 5 This is a comparison chart of different algorithms provided by an embodiment of the present invention, with the number of carriers and targets remaining unchanged, the population size being 20, and 500 iterations.
[0052] Figure 6 This is a comparison chart of different algorithms provided by an embodiment of the present invention, with the number of carriers and targets unchanged, the population size being 20, and iterating 1000 times.
[0053] Figure 7 This is a comparison chart of different algorithms provided by an embodiment of the present invention, with the number of carriers and targets unchanged, the population size being 20, and 200 iterations.
[0054] Figure 8 This is a comparison chart of different algorithms provided by an embodiment of the present invention, with the number of carriers and targets remaining unchanged, the population size being 20, and 500 iterations.
[0055] Fig. 9 This is a comparison chart of different algorithms provided by an embodiment of the present invention, with the number of carriers and targets unchanged, the population size being 20, and iterating 1000 times. DETAILED DESCRIPTION
[0056] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0057] like Figure 1 As shown, an NHHO algorithm for online task allocation of heterogeneous carrier clusters provided by an embodiment of the present invention includes the following steps:
[0058] S101, inputting the heterogeneous carrier cluster and the relevant data of the three types of targets, and determining the initialization parameters to be set, including the population size, the maximum number of iterations, the switching probability, the jump intensity, etc.;
[0059] S102, initializing the population according to the adversarial environment;
[0060] S103, applying the Harris Eagle algorithm to introduce a spiral update mechanism to calculate the standard function value of each heterogeneous carrier cluster position, determine the fitness value of the individual, and obtain the current optimal function value and the corresponding position;
[0061] S104, updating the specific positions of the three types of targets and the enemy type 1 target;
[0062] S105, dynamically allocating tasks according to the heterogeneous carrier cluster and the discovery of Class 1 targets;
[0063] S106, judging whether the action distance reaches the standard, if not, returning to S103;
[0064] S107, attacking Category 3 targets and Category 1 targets;
[0065] S108, if the maximum number of iterations is not reached or the target number of completions is insufficient, return to S103, otherwise, proceed to the next step;
[0066] S109, outputting the fitness and required time of task allocation.
[0067] This paper proposes an NHHO algorithm for online task allocation of heterogeneous carrier clusters. By combining the Harris Hawk algorithm and introducing a spiral update mechanism, the task allocation of heterogeneous carrier clusters is optimized to improve the efficiency and adaptability of the system. The specific working principle is as follows:
[0068] Step 1: Input parameter initialization
[0069] Input data of heterogeneous carrier clusters, including relevant data of three types of targets (friendly forces, enemy forces, resource points, etc.).
[0070] Determine initialization parameters, such as population size, maximum number of iterations, switching probability, jump strength, etc.
[0071] Relevant parameters are input by the user or system preset to initialize the basic configurations of the algorithm and provide a basis for subsequent calculations.
[0072] Step 2: Population initialization
[0073] Initialize the population according to the adversarial environment.
[0074] Generate an initial population containing a variety of heterogeneous carrier positions and states.
[0075] In the initial population, each individual represents a task allocation scheme.
[0076] Step 3: Calculate the fitness value
[0077] Harris Eagle algorithm is applied to introduce spiral update mechanism to calculate the standard function value of each heterogeneous carrier cluster position and determine the fitness value of the individual.
[0078] The hunting behavior of Harris's hawk is simulated through the spiral updating mechanism, and the position of each individual and the corresponding target function value are calculated.
[0079] Determine the fitness value of each individual and obtain the current optimal fitness value and corresponding position.
[0080] Step 4: Update the target location
[0081] Update the specific locations of Category 3 targets and enemy Category 1 targets.
[0082] Based on the current environment and enemy and friendly situations, the target position is adjusted dynamically to reflect the actual battlefield situation.
[0083] Step 5: Dynamic Task Allocation
[0084] Dynamic task allocation is performed based on the discovery of heterogeneous carrier clusters and Class 1 targets.
[0085] Tasks are dynamically allocated based on the location and status of the carrier and target to ensure optimal resource utilization and efficient task completion.
[0086] Step 6: Determine the range of action
[0087] Determine whether the action distance meets the standard. If not, return to step 3 to continue iteration.
[0088] Based on the calculated distance between the carrier and the target, it is determined whether the standard is met and whether the task allocation needs to be further optimized.
[0089] Step 7: Targeted Attack
[0090] Attack Category 3 targets and Category 1 targets.
[0091] Once the mission assignment is determined and the range standard is reached, the carrier attacks the target and performs the mission.
[0092] Step 8: Iteration judgment
[0093] Determine whether the maximum number of iterations or the target number of completions has been reached. If not, return to step 3 to continue iterating.
[0094] Determine whether the current number of iterations has reached the preset maximum value, or whether the target completion number has reached the expected value. If not, continue iterative optimization.
[0095] Step 9: Output the results
[0096] Output the fitness value and required time of the task assignment.
[0097] Calculate and output the fitness value and required time of the final task allocation plan as evaluation indicators for task completion.
[0098] Specific operation examples:
[0099] Assume that there is a heterogeneous carrier swarm including drones, ground robots, and maritime patrol boats, targeting enemy vehicles, enemy soldiers, and resource points.
[0100] 1. Input parameter initialization:
[0101] Enter the carrier and target data.
[0102] Set the population size to 50, the maximum number of iterations to 1000, the switching probability to 0.3, and the jump strength to 1.5.
[0103] 2. Population initialization:
[0104] Generate an initial population containing 50 task allocation schemes.
[0105] 3. Calculate the fitness value:
[0106] The spiral update mechanism is used to simulate the hunting behavior of Harris's hawk and calculate the fitness value.
[0107] 4. Update the target location:
[0108] Dynamically adjust target positions to reflect real-time battlefield conditions.
[0109] 5. Dynamic task allocation:
[0110] Tasks are dynamically assigned based on the discovery of carriers and targets.
[0111] 6. Determine the effective distance:
[0112] Check whether the distance between the carrier and the target meets the standard.
[0113] 7. Target Attack:
[0114] Attack the target of the determined mission.
[0115] 8. Iteration judgment:
[0116] Determine whether the maximum number of iterations or the target number of completions has been reached. If not, continue iterating.
[0117] 9. Output results:
[0118] Output the final task allocation fitness value and required time.
[0119] The algorithm combines the Harris Eagle algorithm and the spiral update mechanism to achieve efficient optimization of heterogeneous carrier cluster task allocation, improving the flexibility and adaptability of the system. By dynamically adjusting task allocation and updating target positions in real time, the efficiency and accuracy of task completion are effectively improved.
[0120] like Figure 2 As shown, an online task allocation system for heterogeneous carrier clusters provided by an embodiment of the present invention includes:
[0121] The initialization module is used to input the heterogeneous carrier cluster and the relevant data of the three types of targets, determine the initialization parameters to be set, including the population size, the maximum number of iterations, the switching probability, the jump intensity, etc.; initialize the population according to the adversarial environment;
[0122] The calculation module is used to apply the Harris Eagle algorithm to introduce the spiral update mechanism calculation, calculate the standard function value of each heterogeneous carrier cluster position, determine the individual fitness value, obtain the current optimal function value and the corresponding position; update the specific position of the three types of target positions and the enemy type 1 target;
[0123] Allocation module, used for dynamic task allocation based on heterogeneous carrier clusters and Class 1 target discovery;
[0124] The judgment module is used to judge whether the action distance reaches the standard.
[0125] The output module is used to output the fitness and required time of task allocation.
[0126] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the NHHO algorithm for online task allocation of heterogeneous carrier clusters.
[0127] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the NHHO algorithm for online task allocation of heterogeneous carrier clusters.
[0128] Another object of the present invention is to provide an information data processing terminal, which is used to implement the heterogeneous carrier cluster online task allocation system.
[0129] The present invention is specifically implemented:
[0130] 1. The focus of heterogeneous carrier cluster task allocation is to plan a scientific action plan for each unmanned carrier based on the actual situation of the mission scope environment, combined with factors such as the performance, fuel consumption, task priority and target threat of the unmanned carrier. At the same time, in order to ensure that the heterogeneous carrier cluster achieves the best performance in the mission scope environment, task allocation needs to reasonably formulate load configuration and use time points.
[0131] 2 Modeling of Task Allocation Problem in Heterogeneous Carrier Clusters
[0132] Assume that our side is composed of a heterogeneous carrier cluster consisting of K unmanned carriers to perform a mission, and deploy N Class 1 targets, M Class 2 targets, and A Class 3 targets within the mission scope. According to factors such as the type of target, threat level, and execution difficulty, the threat level of the target can be determined, and this level is usually quantified by the target threat coefficient.
[0133] §2.2.1 Model parameter design
[0134] Air carrier U = {U1, U2, ..., Un} (attribute quadruple<UA,PosA,SDA,ZCJLA> , respectively represent the air carrier number, air carrier position, speed, and detection distance);
[0135] Surface vehicle U = {U1, U2, ..., Un} (attribute quadruple<UB,PosB,SDB,GJJLB> , respectively represent the number of the surface carrier, the position of the surface carrier, the speed, and the range of action);
[0136] 3 types of targets U = {U1, U2, ..., Un} (attribute quadruple<UC,PosC,JZC,WXCDC> , respectively representing the 3 types of target numbers, 3 types of target locations, mission value, and threat level);
[0137] 1 type of target U={U1,U2,...,Un}(attribute quadruple<UD,PosD,JZD,WXCDD> , respectively represent the number of category 1 target, the location of category 1 target, the mission value, and the threat level);
[0138] All the targets are targets to be hit, and the locations of Category 3 targets are known, but the locations of Category 1 targets are unknown. Category 2 targets are similar to Category 1 targets, so they are not set separately. Only the mission value and threat level are changed in the parameter settings of Category 1 targets.
[0139] §2.2.2 Task costs and benefits
[0140] When allocating air carrier coordination and surface carrier missions, the goal is to improve mission execution efficiency and reduce our losses. Therefore, this paper uses the benefits of unmanned carrier missions, the mission risks assumed, and the flight length as the criteria for measuring the effectiveness of mission execution. Heterogeneous carrier clusters need to weigh costs and benefits. The costs that unmanned carriers need to bear include the costs of potential threats to mission carriers and the additional flight and time costs consumed during mission execution.
[0141] The target threat cost is calculated as follows: When unmanned carrier i is performing a mission, the probability of target j damaging it is WXCDj, so the survival probability of carrier i can be expressed as Pij=1-WXCDj (Pij>0 is required to continue the mission). After performing n missions continuously (here n≥2), the cumulative threat cost T1 caused by the target to the carrier can be expressed as:
[0142]
[0143] Voyage cost: The voyage cost reflects the various resources consumed by the unmanned carrier during the execution of the specified mission, including but not limited to energy, time, materials and other related costs. In order to more accurately evaluate this cost, this paper uses the cumulative time required for the unmanned carrier to complete the specified mission as an indicator to measure the entire voyage cost. The voyage cost represented by the symbol T2 refers to this overall time cost. The voyage cost T2 is expressed as:
[0144]
[0145] In the formula, TMAX represents the maximum time allowed for a single carrier to perform a task, while TMIN represents the minimum time required for a single carrier to perform a task. The requirement of this article is to ensure that all unmanned carriers can reach the mission site and complete the assigned tasks, that is, the time for each unmanned carrier to perform a task should be between TMAX and TMIN.
[0146] Reward parameter: refers to the setting of the heterogeneous carrier cluster according to the mission value of the target and the required navigation time after hitting the target, which is the direction for the algorithm iteration. Reward parameter T3 is expressed as:
[0147]
[0148] In the formula, TMAX represents the maximum time allowed for a single carrier to perform any task; the smaller the reward parameter, the better the completion of this model.
[0149] Flight time cost: To prevent the heterogeneous carrier cluster from falling into the local optimum and consuming too much overall time, the flight time cost T4 is set. T4 is expressed as:
[0150]
[0151] Where TZ is the total time to complete all tasks, and T' is a constant.
[0152] §2.2.3 Task execution constraints
[0153] In order to further improve the authenticity and complexity of the model, the task execution conditions are constrained:
[0154] All air carriers can attack Category 1 and Category 2 targets without suffering any damage, but they cannot attack Category 3 targets:
[0155]
[0156] JZA is the attack performance of the airborne carrier. The paper assumes that all airborne carriers have the same performance.
[0157] Depending on the scope of the mission, all unmanned vehicles are at different altitudes and their positions can overlap:
[0158]
[0159] WZ is the overall position of the mission range, and the position of the unmanned carrier in this article is within the entire mission range.
[0160] The location of Category 3 targets can be detected by satellites and other means in the early stage, but the location of Category 1 and Category 2 targets cannot be detected:
[0161]
[0162] WZC is a known parameter, and WZD is represented by a random number.
[0163] Logical constraints require that before attacking the three types of targets, the heterogeneous carrier cluster must first complete the reconnaissance mission and then execute the strike mission:
[0164] T B iC ≥T A iC (2.10)
[0165] TB is the time for the surface carrier to attack the three types of targets i, and TAS is the time for the air carrier to attack the three types of targets i.
[0166] §2.2.4 Fitness function
[0167] The optimization objective function of heterogeneous carrier cluster task allocation is affected by multiple factors, including target threat cost, voyage cost, reward parameter and navigation time cost. A weighted method is used to balance these four costs and benefits. The importance of each influencing factor is reflected by the size of the corresponding weighted factor, where W1 = 0.2, W2 = 0.3, W3 = 0.2, W4 = 0.3;
[0168] W1+W2+W3+W4=1(2.11)
[0169] The overall cost-benefit in this article is:
[0170]
[0171] Where Z is the overall cost benefit, and n is the integration of all unmanned heterogeneous carrier clusters.
[0172] §2.3 Conclusion
[0173] The present invention firstly expounds four kinds of heterogeneous carrier cluster task execution applications according to the task scope and environmental background assumptions; then, a heterogeneous carrier cluster task allocation model is given, including model parameters, task costs and benefits, task execution constraints, fitness function, etc.; finally, based on unmanned carrier parameters and task execution methods, a heterogeneous carrier cluster task allocation model is constructed, laying the foundation for the development of heterogeneous carrier cluster task allocation methods.
[0174] 3. Task allocation method for heterogeneous carrier clusters based on HHO algorithm
[0175] The task allocation problem belongs to the majority decision problem. Common solutions include centralized solution algorithm and distributed solution algorithm. Aiming at the complex coupling restrictions and uncertainty characteristics of the task allocation problem of heterogeneous carrier clusters in adversarial environments, this paper proposes a task allocation method for heterogeneous carrier clusters based on the HHO algorithm. The effectiveness of this method is verified through simulation analysis.
[0176] like Figure 3 As shown in §3.1 Harris Hawk Algorithm
[0177] §3.1.1 Algorithm Rules
[0178] Before each iteration, it is necessary to check whether the eagle group has crossed the boundary and make necessary adjustments. At the same time, the position and fitness value of the prey are updated.
[0179] During the searching phase, hawks use two different searching methods, depending on whether they have already spotted prey.
[0180] As the siege continues, the prey's ability to escape will gradually decrease.
[0181] There is a threshold of prey energy, and once this threshold is exceeded, the Harris Hawk will move from the search phase to the swarming phase.
[0182] Before each raid, the prey escaped from the encirclement.
[0183] Harris's hawks adopt four different hunting strategies based on their assessment of the physical state of the prey trapped in the siege and its escape potential.
[0184] Each round of prey will eventually be captured, so each iteration will produce a new prey, and this position is occupied by the prey with the best fitness in the new generation population.
[0185] §3.1.2 Algorithm Features
[0186] The Harris Hawk algorithm is a calculation method for the Harris Hawk to surround its prey. It is suitable for this article in the following aspects:
[0187] Our unmanned carriers are regarded as Harris Hawk population A and Harris Hawk population B respectively, and the three types of targets are regarded as prey.
[0188] Category 3 targets do not have fixed positions, but adjust their directions to more suitable locations based on the situation.
[0189] Each unmanned carrier in the heterogeneous carrier cluster will perform task assignments based on the situation and will also change task assignments based on changes in mission objectives.
[0190] §3.2 Task allocation process for heterogeneous carrier clusters
[0191] The specific steps of the algorithm are:
[0192] Step 1: Input the heterogeneous carrier cluster and the relevant data of the three types of targets, and determine the initialization parameters to be set, including population size, maximum number of iterations, switching probability, jump intensity, etc.
[0193] Step 2: Initialize the population according to the adversarial environment;
[0194] Step 3: Apply the Harris Eagle algorithm to calculate the standard function value of each heterogeneous carrier cluster position, determine the fitness value of the individual, and obtain the current optimal function value and corresponding position;
[0195] Step 4: Update the specific location of the 3 types of targets and the enemy's 1 type of targets;
[0196] Step 5: Determine whether the effective distance reaches the standard, if not, return to step 3;
[0197] Step 6: Attack Category 3 targets and Category 1 targets;
[0198] Step 8: If the maximum number of iterations is not reached or the task is not completed, return to step 3, otherwise, proceed to the next step;
[0199] Step 9: Output the task allocation fitness and required time.
[0200] §3.3 Simulation Analysis
[0201] In order to evaluate the performance of the improved algorithm, simulation experiments were carried out using Harris Hawk Optimization (HHO), Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). In view of the randomness of the algorithm, each algorithm was simulated multiple times, and the average of the optimal fitness values of 20 times was selected as the comparison benchmark.
[0202] §3.3.1 Simulation environment settings
[0203] During a mission execution operation, we found Category 3 targets and Category 1 targets. In response to this threat, we dispatched 6 air carriers and 6 surface carriers to carry out the mission. The targets include 5 Category 3 targets, 10 Category 1 targets and 5 Category 2 targets. The locations of Category 3 targets are roughly known, and they are fixed or slightly moving targets; the locations of other targets are unknown and need to be confirmed by air carriers. Surface carriers are responsible for destroying Category 3 targets based on the target information provided by air carriers, but cannot directly attack Category 1 targets. If the surface carrier triggers a Category 1 target, it will clear the Category 1 target but will be damaged. Due to load limitations, air carriers cannot effectively attack Category 3 targets, but can effectively attack Category 1 targets and Category 2 targets. See Table 3.1 for details of relevant target parameters.
[0204] Table 3.1 Target parameters
[0205] Target Number Location (km) Mission Value Threat Level 1 (46,108) 0.6 0.6 2 (59,38) 0.8 0.8 3 (101,66) 0.7 0.7 4 (2,18) 0.8 0.8 5 (151,34) 0.7 0.7 6 (rand(200),rand(200)) 0.3 0.3 7 (rand(200),rand(200)) 0.3 0.3 8 (rand(200),rand(200)) 0.3 0.3 9 (rand(200),rand(200)) 0.3 0.3 10 (rand(200),rand(200)) 0.3 0.3 11 (rand(200),rand(200)) 0.3 0.3 12 (rand(200),rand(200)) 0.3 0.3 13 (rand(200),rand(200)) 0.3 0.3 14 (rand(200),rand(200)) 0.3 0.3 15 (rand(200),rand(200)) 0.3 0.3 16 (rand(200),rand(200)) 0.4 0.4 17 (rand(200),rand(200)) 0.4 0.4 18 (rand(200),rand(200)) 0.4 0.4 19 (rand(200),rand(200)) 0.4 0.4 20 (rand(200),rand(200)) 0.4 0.4
[0206] In the above table, 1-5 represents Category 3 targets, 6-15 represents Category 1 targets, and 16-20 represents Category 2 targets.
[0207] The attribute parameters of aerial carriers and surface carriers are shown in Tables 3.2 and 3.3.
[0208] Table 3.2 Air carrier parameters
[0209]
[0210] Table 3.3 Surface carrier parameters
[0211]
[0212]
[0213] In Tables 3.2 and 3.3 above, in order to prevent the position of Category 1 target from overlapping with the position of the surface carrier and directly damaging the surface carrier, the positions of the aerial carrier and the surface carrier are made to overlap, and the aerial carrier is released first to operate on Category 1 target.
[0214] §3.3.2 Simulation Analysis
[0215] The above model was simulated using genetic algorithm (GA), particle swarm algorithm (PSO) and Harris Hawk (HHO) respectively. The results were compared with 20 iterations of 200, 500 and 1000 with a population of 20. Figure 4 , Figure 5 and Figure 6 As shown:
[0216] By running the program 200 times, 500 times, and 1000 times respectively, the simulation results of each case show that the fitness value of the HHO algorithm is better than that of the GA algorithm and the PSO algorithm. However, it can be clearly observed from the chart that with the increase of the number of iterations, the performance difference between the HHO algorithm and the GA algorithm gradually decreases. This indicates that the HHO algorithm has a local optimal trap, the so-called "premature phenomenon".
[0217] The present invention firstly explains the principle and process of the HHO algorithm; secondly, performs simulation analysis on the HHO algorithm with different initial conditions; finally, analyzes and explains the program running results, thereby demonstrating the feasibility of the HHO algorithm in solving the task allocation problem of heterogeneous carrier clusters.
[0218] 4. Online task allocation method for heterogeneous carrier clusters based on NHHO algorithm
[0219] In solving the problem of online task allocation of heterogeneous carrier clusters, the advantages of the HHO algorithm over other classical algorithms are mainly reflected in two aspects: 1. The performance advantage of the HHO algorithm; 2. Good adaptability to task allocation problems. However, the HHO algorithm is prone to fall into the disadvantages of local optimal solutions and poor search performance in the early stage of the search, and the fitness function convergence effect under the operation of the algorithm program is not obvious compared with the genetic algorithm and the particle swarm algorithm. Therefore, the present invention intends to improve the HHO algorithm and conduct research on the task allocation method based on the NHHO algorithm.
[0220] §4.1 Improvement Basis
[0221] §4.1.1 Update mechanism based on the contract network protocol algorithm
[0222] Considering the real-time changes in the task scope, this paper uses the Contract Net Protocol (CNP) theory to perform online task allocation for heterogeneous carrier clusters. This allocation method redistributes tasks based on real-time status feedback. There are mainly three situations:
[0223] (1) When a new task (discovering Category 1 or Category 2 targets) appears during the task execution, online task allocation is used to redistribute tasks among heterogeneous carrier clusters.
[0224] (2) When an unmanned vehicle is unable to continue its mission due to limited capabilities or destruction, other unmanned vehicles of the same type will take over and reallocate the mission.
[0225] (3) If the assigned tasks are moved or changed (the tasks are completed), the heterogeneous carrier cluster will reallocate the tasks.
[0226] This paper adopts the improved contract network protocol algorithm to construct a centralized heterogeneous carrier cluster task allocation model. At the same time, considering the actual task scope, the heterogeneous carrier cluster is used as the tenderer, bidder and executor of the task.
[0227] After introducing the contract network protocol algorithm method, calculate whether the task requirements are met, and the judgment function formula is as follows:
[0228]
[0229] In the formula, cand(i,j) represents whether the air carrier can reach the target, POSA-POSC represents the range of the air carrier to reach the target type 1 target, and POSA-POSD represents the range of the air carrier to reach the target type 3. Therefore, when the range of the air carrier from the target type 1 target is less than the range of the target type 3, it returns 1, otherwise it returns 0. Therefore, as long as the air carrier can meet the range requirements, it can bid for the target task.
[0230] Then, through the contract network algorithm, the bidder with the shortest time to reach the target is selected to perform the task.
[0231] §4.1.2 Spiral update mechanism combining the idea of whale algorithm
[0232] In order to improve the search method of the HHO algorithm population and enhance the search performance of the algorithm in the early stage of the search, the spiral search method of the Whale Optimization Algorithm (WOA) is introduced into the HHO algorithm. The Whale Algorithm simulates the predation behavior of humpback whales. It will spit out a bubble net along the spiral line to complete the attack on the prey. The algorithm mainly includes three stages: surrounding the prey, bubble net attack, and searching for prey. In the bubble net attack stage, the whale spits out a bubble net along the spiral line and gradually approaches the prey. In order to simulate this process in the WOA algorithm, a spiral motion equation is constructed. The faster convergence speed of the WOA algorithm benefits from this spiral update mechanism.
[0233] This paper introduces the spiral update mechanism of the WOA algorithm into the HHO algorithm, which improves the convergence performance of the HHO algorithm. At the same time, after the introduction of the spiral mechanism of the WOA algorithm, the HHO algorithm can update the position in the spiral mechanism to determine the distance between the unmanned carrier and the target when new Class 1 and Class 2 targets are discovered, and analyze the fitness value through the position update formula to obtain the most fitness value.
[0234] The position update formula of the HHO algorithm after the introduction of the spiral mechanism of the WOA algorithm is as follows:
[0235]
[0236] Where b is the spiral constant and L is a random number between [0,1].
[0237] §4.2 Task allocation process based on improved Harris Hawk algorithm
[0238] The specific steps of the algorithm are:
[0239] Step 1: Input the heterogeneous carrier cluster and the relevant data of the three types of targets, and determine the initialization parameters to be set, including population size, maximum number of iterations, switching probability, jump intensity, etc.
[0240] Step 2: Initialize the population according to the adversarial environment;
[0241] Step 3: Apply the Harris Eagle algorithm to introduce the spiral update mechanism to calculate the standard function value of each heterogeneous carrier cluster position, determine the fitness value of the individual, and obtain the current optimal function value and corresponding position;
[0242] Step 4: Update the specific location of the 3 types of targets and the enemy's 1 type of targets;
[0243] Step 5: Dynamically allocate tasks based on heterogeneous carrier clusters and Class 1 target discovery;
[0244] Step 6: Determine whether the action distance reaches the standard, if not, return to step 3;
[0245] Step 7: Attack Category 3 targets and Category 1 targets;
[0246] Step 8: If the maximum number of iterations is not reached or the target number of completions is insufficient, return to step 3, otherwise, proceed to the next step;
[0247] Step 9: Output the fitness and required time of task allocation.
[0248] Simulation analysis: Setting the target and unmanned carrier parameters the same as in Section 3.3.1, setting the number of iterations (evolutions) to be the same as the initial population number in Chapter 3, the convergence effect of the improved optimization algorithm compared with that before the improvement is as follows:
[0249] The basic parameters are shown in Table 3.1, Table 3.2 and Table 3.3. The population size is 20, and the New Harris Hawks Optimization (NHHO) algorithm is used for 200, 500 and 1000 iterations respectively, and compared with HHO, GA and PSO. Figure 7 , Figure 8 and Fig. 9 As shown:
[0250] pass Figure 7 , Figure 8 and Fig. 9 It can be seen that compared with HHO, PSO and GA, the fitness function convergence speed of NHHO algorithm is obviously faster than that of HHO, PSO and GA algorithms, and the optimal fitness also decreases rapidly with the increase of iteration number.
[0251] In order to further illustrate that the NHHO algorithm is better than HHO, PSO and GA in task allocation in adversarial environments, the running time and optimal function value of the algorithm population and iteration number will be changed respectively. The basic parameters are the optimal fitness and running time of different algorithms in Table 3.1, Table 3.2 and Table 3.3. They are run 20 times respectively, and the average values are shown in Table 4.1, Table 4.2, Table 4.3, Table 4.4 and Table 4.5:
[0252] Table 4.1 Comparison of experimental data
[0253]
[0254] Table 4.2 Comparison of experimental data
[0255]
[0256] Table 4.3 Comparison of experimental data
[0257]
[0258] Table 4.4 Experimental data comparison
[0259]
[0260]
[0261] Table 4.5 Comparison of experimental data
[0262]
[0263] After running the program 20 times in a row and calculating the average running time and the optimal function value, the simulation results of the improved algorithm show that in the tests of the above scenarios, the optimization performance of the objective function value exceeds that of the classical algorithm before the improvement. At the same time, these running results clearly indicate that the NHHO algorithm significantly improves the efficiency of task allocation and effectively guides the population to evolve towards the global optimal solution, thereby enhancing the global search ability of the algorithm. In addition, the improved measures adopted improve the optimization performance of the algorithm in the initial stage and reduce the probability of falling into the local optimal solution, which proves that the improved algorithm has a substantial improvement compared with the traditional algorithm. Nevertheless, the simulation experiment also shows a shortcoming of the NHHO algorithm: the time required for the improved task allocation has increased. This is because the online task allocation mechanism is used in the NHHO algorithm, which increases the amount of data calculation and the complexity of the algorithm, and thus increases the running time accordingly.
[0264] This paper first improves the HHO algorithm, and uses the contract network protocol in the online task allocation method to improve the Harris Eagle algorithm to prevent the algorithm from reaching local optimality too early, thereby improving the degree of convergence, and introduces the spiral search method of the whale algorithm to enhance the search performance of the algorithm in the early stage of the search. For different initial conditions defined in Chapter 3, the program is run using the improved algorithm, and the results are analyzed in depth. The results of this analysis clearly demonstrate the feasibility and effectiveness of the improved algorithm
[0265] The present invention proposes an NHHO algorithm and system, which are applied to heterogeneous carrier cluster task allocation. When a heterogeneous carrier cluster performs a sea strike mission, the algorithm and system can quickly and accurately pre-allocate tasks and online task allocation for each carrier in the cluster. The objective function of task allocation is significantly better than the classic algorithms for task allocation—genetic algorithm, particle swarm algorithm, and classic Harris Eagle algorithm. Therefore, the present invention can effectively solve the problem of online task allocation for heterogeneous carrier clusters.
[0266] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.
[0267] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.
Claims
1. An NHHO algorithm for online task allocation in heterogeneous carrier clusters, characterized in that: The following steps are involved: a. Input the relevant data of heterogeneous carrier clusters and three types of targets, and determine the initialization parameters, including population size, maximum number of iterations, switching probability, and jump intensity; b. Initialize the population according to the adversarial environment; c. Apply the Harris Eagle algorithm to introduce the spiral update mechanism calculation to determine the standard function value of each heterogeneous carrier cluster position, thereby obtaining the individual fitness value and determining the current optimal function value and corresponding position; d. Dynamically allocate tasks based on heterogeneous carrier clusters and a class of target discovery situations; e. Determine whether the conditions for task completion are met. If not, repeat steps c to d; f. Attack Category III targets and Category I targets; g. Output the fitness and time required for task allocation.
2. The NHHO algorithm for online task allocation of heterogeneous carrier clusters according to claim 1, characterized in that: In step c, by introducing a spiral update mechanism, the Harris Hawk algorithm can simulate the hunting behavior of an eagle and update the positions of heterogeneous carrier clusters in a spiral manner, thereby optimizing task allocation.
3. The NHHO algorithm for online task allocation of heterogeneous carrier clusters according to claim 1 or 2, characterized in that: In the step d, the dynamic task allocation is based on the current position of the heterogeneous carrier cluster and the discovery of a class of targets, and the optimal allocation of tasks is achieved by evaluating the fitness value of each heterogeneous carrier.
4. A heterogeneous carrier cluster online task allocation system implementing the heterogeneous carrier cluster online task allocation NHHO algorithm as claimed in claim 1, characterized in that: The heterogeneous carrier cluster online task allocation system includes: The initialization module is used to input the heterogeneous carrier cluster and the relevant data of the three types of targets, determine the initialization parameters to be set, including the population size, the maximum number of iterations, the switching probability, the jump intensity, etc.; initialize the population according to the adversarial environment; The calculation module is used to apply the Harris Eagle algorithm to introduce the spiral update mechanism calculation, calculate the standard function value of each heterogeneous carrier cluster position, determine the individual fitness value, obtain the current optimal function value and the corresponding position; update the specific position of the three types of target positions and the enemy type 1 target; Allocation module, used for dynamic task allocation based on heterogeneous carrier clusters and Class 1 target discovery; The judgment module is used to judge whether the action distance reaches the standard. The output module is used to output the fitness and required time of task allocation.
5. A computer device, characterized in that: The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the NHHO algorithm for online task allocation of heterogeneous carrier clusters as claimed in claim 1.
6. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the NHHO algorithm for online task allocation of heterogeneous carrier clusters as claimed in claim 1.
7. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the heterogeneous carrier cluster online task allocation system as described in claim 2.
Citation Information
Patent Citations
Engineering optimization method of Harris hawks algorithm based on multi-strategy enhancement
CN110728001A
RBF (Radial Basis Function) neural network optimization method based on improved Harlisia eagle algorithm
CN113240069A
Cooperative attack task allocation method for heterogeneous unmanned aerial vehicle group
CN115829230A
Improved Harris eagle algorithm for cloud computing task scheduling
CN116579370A
Heterogeneous unmanned aerial vehicle cooperative search path planning method, electronic equipment and storage medium
CN116859722A
