A task allocation method

Abstract

The application provides a task allocation method applied to a UAV platform based on sequential ordering collaborative bidding, which comprises the following steps: dividing the UAV platform and disaster relief sites into a first point set and a second point set; representing the contact between the first point set and the second point set by a bipartite graph; allocating target tasks between the first UAV platform and the first site based on a preset bidding algorithm; the first UAV platform is any UAV platform in the first point set; the first site is any disaster relief site in the second point set; determining a bidding priority parameter of the target task; generating a bidding sequence of the UAV platform based on the bidding priority parameter and a preset roulette method; adopting a preset bid bidding algorithm to bid for the target task in the order of the bidding sequence to determine the target task with the highest bid; and performing target value decay and tolerance constraint processing on the target task with the highest bid to obtain a target UAV platform allocated with multiple target tasks.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and more particularly to a task allocation method. Background Technology

[0002] A single drone has limited payload capacity. In disasters, there are usually multiple locations that require rescue. A single drone has low coverage and cannot complete disaster relief work in a timely manner. Summary of the Invention

[0003] This invention provides a task allocation method to solve technical problems such as the low coverage of a single drone, which makes it impossible to complete disaster relief work in a timely manner.

[0004] This invention provides a task allocation method applied to a drone platform based on sequential ranking and collaborative bidding. The method includes:

[0005] The unmanned aerial vehicle platform and the disaster relief location are divided into a first set of points and a second set of points; the relationship between the first set of points and the second set of points is represented by a bipartite graph;

[0006] The target task between the first drone platform and the first location is assigned based on a preset bidding algorithm; the first drone platform is any drone platform in the first point set; the first location is any disaster relief location in the second point set.

[0007] Determine the bidding priority parameters for the target task;

[0008] The bidding sequence for the drone platform is generated based on the bidding priority parameters and the preset roulette method.

[0009] The target task is bid on using a preset bidding algorithm in the order of the bidding sequence to obtain the target task with the highest bid.

[0010] The target task with the highest bid is subjected to target value decay and tolerance constraint processing to obtain the target UAV platform assigned multiple target tasks.

[0011] In some implementations, determining the bidding priority parameter for the target task includes:

[0012] Obtain the value parameters of the target task and the corresponding first weighting factor, as well as the total number of resources of the first UAV platform and the corresponding second weighting factor;

[0013] The bidding priority parameter of the target task is determined based on the value parameter, the first weighting factor, the total number of resources, and the second weighting factor.

[0014] In some implementations, obtaining the value parameters of the target task includes:

[0015] Obtain the importance parameter of the target task and the success rate parameter of the first resource of the first UAV platform in executing the target task;

[0016] The importance parameter and the success rate parameter are used as the value parameter.

[0017] In some implementations, generating the bidding sequence for the drone platform based on the bidding priority parameter and a preset roulette wheel betting method includes:

[0018] The bidding priority parameter is converted to obtain the probability parameter of the first drone platform being selected;

[0019] Determine the probability interval of the probability parameter distribution;

[0020] Random numbers for the first UAV platform are generated based on the probability parameters and the probability interval;

[0021] The bidding sequence for the drone platform is determined based on the roulette wheel betting method and the random number.

[0022] In some implementations, the step of using a preset bidding algorithm to bid on the target task according to the order of the auction sequence to obtain the target task with the highest bid includes:

[0023] Obtain the current highest bid and the dynamically decayed task value for the target task;

[0024] The bidding revenue of the first resource of the first UAV platform for the target task is determined based on the current highest bid and the dynamically decayed task value.

[0025] The target task with the highest bid is determined based on the bidding revenue and the current highest bid for the target task.

[0026] In some implementations, the target value decay includes value decay within the same round and value decay between different rounds; the step of applying target value decay and tolerance constraint processing to the target task with the highest bid to obtain a target UAV platform assigned multiple target tasks includes:

[0027] The highest-priced target task is subjected to value decay processing within the same round to obtain the current first value of the highest-priced target task.

[0028] The first value is subjected to value decay processing across different rounds to obtain the current second value of the target task with the highest bid.

[0029] Based on the second value, a tolerance constraint is applied to obtain the target UAV platform assigned to multiple target tasks.

[0030] In some implementations, the step of performing value decay processing on the highest-priced target task within the same round to obtain the current first value of the highest-priced target task includes:

[0031] Obtain the initial weight value and preset decay index of the target task with the highest bid;

[0032] The first value is determined based on the initial weight value and the decay index.

[0033] In some implementations, the step of applying value decay processing to the first value across different rounds to obtain the current second value of the target task with the highest bid includes:

[0034] Obtain the weight of the high-value tasks frequently assigned by the first platform corresponding to the first value and the number of times the price is reset to zero;

[0035] The weight of the high-value task and the number of times the bid is reset are subjected to value decay processing across different rounds to obtain the current second value of the target task with the highest bid.

[0036] In some implementations, the process of applying tolerance constraints based on the second value to obtain a target UAV platform for assigning multi-objective tasks includes:

[0037] The tolerance level for the target drone platform in each round of bidding is determined based on the second value.

[0038] The target UAV platform for assigning multi-target tasks is determined based on the tolerance level.

[0039] In some embodiments, the method further includes:

[0040] A preset weighted fusion model is used to perform multi-objective collaborative optimization on the target UAV platform that is assigned multi-objective tasks, so as to obtain the high-value tasks assigned by the target UAV platform.

[0041] A multi-objective evaluation is performed on the high-value tasks assigned to the target UAV platform to obtain evaluation results; the evaluation results characterize the quality of the task assignments of the target UAV platform.

[0042] This invention also provides a task allocation device, the device comprising:

[0043] The acquisition unit is used to acquire first data collected by the mobile robot; the first data includes at least spatial data, image data, and motion data.

[0044] A partitioning unit is used to divide the unmanned aerial vehicle platform and the disaster relief location into a first point set and a second point set; the relationship between the first point set and the second point set is represented by a bipartite graph;

[0045] The allocation unit is used to allocate target tasks between a first drone platform and a first location based on a preset bidding algorithm; the first drone platform is any drone platform in the first set of points; the first location is any disaster relief location in the second set of points.

[0046] A determining unit is used to determine the bidding priority parameters of the target task;

[0047] A generation unit is used to generate the bidding sequence for the drone platform based on the bidding priority parameters and a preset roulette wheel betting method.

[0048] The bidding unit is used to bid on the target task according to the order of the auction sequence using a preset bidding algorithm, and obtain the target task with the highest bid.

[0049] The processing unit is used to perform target value decay and tolerance constraint processing on the target task with the highest bid to obtain a target UAV platform with multiple target tasks.

[0050] This invention provides a task allocation device, the device comprising: a processor and a memory for storing a computer program capable of running on the processor, wherein the processor, when running the computer program, performs the steps of any of the methods described above.

[0051] This invention provides a storage medium storing a computer program; when the computer program is executed by a processor, it implements the steps of any of the methods described above.

[0052] This invention provides a task allocation method, which includes: applying a drone platform to a sequentially ordered collaborative bidding process; the method includes: dividing the drone platform and the disaster relief location into a first set of points and a second set of points; representing the relationship between the first set of points and the second set of points using a bipartite graph; allocating target tasks between the first drone platform and the first location based on a preset bidding algorithm; the first drone platform being any drone platform in the first set of points; the first location being any disaster relief location in the second set of points; determining the bidding priority parameter of the target task; generating a bidding sequence for the drone platform based on the bidding priority parameter and a preset roulette wheel method; using a preset bidding algorithm to bid on the target task according to the order of the bidding sequence, obtaining the target task with the highest bid; and applying target value decay and tolerance constraints to the target task with the highest bid to obtain a target drone platform allocated multiple target tasks. The algorithm divides the drone platform and the disaster relief location into a first set and a second set. The relationships between the first and second sets are represented by a bipartite graph. A pre-defined auction algorithm is used to allocate target tasks between the first drone platform and the first location. The auction priority parameters for the target tasks are determined. An auction sequence for the drone platforms is generated based on the auction priority parameters and a pre-defined roulette wheel algorithm. A pre-defined bidding algorithm is used to bid on the target tasks according to the auction sequence, obtaining the target task with the highest bid. The target task with the highest bid is then subject to target value decay and tolerance constraints to obtain the target drone platform allocated multiple target tasks. This effectively improves resource allocation efficiency, enhances the real-time performance of the algorithm, and solves problems such as low coverage of single drones and inability to complete disaster relief work in a timely manner through the coordinated optimization of target value decay and tolerance constraints. Attached Figure Description

[0053] Figure 1 A flowchart illustrating a task allocation method provided in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram illustrating the relationship between two point sets represented by a bipartite graph in an embodiment of the present invention;

[0055] Figure 3 This is the convergence curve for the case where the number of platforms is 4 and the task objectives are 5 in this embodiment of the application.

[0056] Figure 4 This is a schematic diagram illustrating the target allocation when the number of platforms is 4 and the number of task objectives is 5, as shown in the embodiments of this application.

[0057] Figure 5 This is the convergence curve for the case where the number of platforms is 4 and the task objectives are 30 in this embodiment of the application.

[0058] Figure 6 This is a schematic diagram illustrating the target allocation when the number of platforms is 4 and the number of task objectives is 30, as shown in the embodiments of this application.

[0059] Figure 7 This is a schematic diagram of the structure of a task allocation device provided in an embodiment of the present invention;

[0060] Figure 8 This is a schematic diagram of a hardware structure of a task allocation device according to an embodiment of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0062] The specific technical features described in the various embodiments in the detailed implementation can be combined in various ways without contradiction. For example, different implementation methods can be formed by combining different specific technical features. In order to avoid unnecessary repetition, the various possible combinations of the specific technical features in this invention will not be described separately.

[0063] It should also be noted that, in order to avoid obscuring the present invention with unnecessary details, only the structures and / or processing steps closely related to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0064] Additionally, it should be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. In the following description, the terms "first," "second," etc., are used merely to distinguish different objects and do not indicate any similarity or connection between them. It should be understood that the directional descriptions such as "above," "below," "inside," and "outside" refer to the orientation under normal use conditions.

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the specific technical solutions of the invention will be further described in detail below with reference to the accompanying drawings of the embodiments of the present invention. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0066] Drone technology has developed rapidly over the past decade and has been widely used for purposes such as aerial photography, emergency information gathering, and cargo transportation. Recently, the application of drones in disaster relief has expanded to include search and rescue missions and the delivery of supplies. Drones are compact and maneuverable, easily traversing complex geographical environments such as mountains and rivers, effectively bypassing adverse conditions such as ground traffic congestion and flooding, and quickly reaching areas inaccessible by traditional means of transportation. This allows them to reach disaster sites and complete missions in a relatively short time, playing a vital role in diverse disaster scenarios and complex terrain conditions.

[0067] A single drone has limited payload capacity. In disasters, there are usually multiple locations that require rescue. A single drone has low coverage and cannot complete disaster relief work in a timely manner. Usually, multiple drones need to cooperate and coordinate to complete disaster relief work.

[0068] In addition, most current task allocation methods based on auction algorithms do not consider the reasonable allocation of drone platform resources to high- and low-value tasks, thus affecting resource utilization and task completion rate, and the real-time performance of the algorithms still has much room for improvement.

[0069] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0070] This invention provides a task allocation method, such as... Figure 1 As shown, Figure 1 This is a flowchart illustrating a task allocation method provided in an embodiment of the present invention; applied to a drone platform based on sequential ranking and collaborative bidding, the method includes:

[0071] Step S101: Divide the drone platform and the disaster relief location into a first set of points and a second set of points; the relationship between the first set of points and the second set of points is represented by a bipartite graph.

[0072] Step S102: Based on a preset auction algorithm, assign target tasks between the first drone platform and the first location; the first drone platform is any drone platform in the first point set; the first location is any disaster relief location in the second point set.

[0073] Step S103: Determine the bidding priority parameters for the target task.

[0074] Step S104: Generate the bidding sequence for the drone platform based on the bidding priority parameters and the preset roulette method.

[0075] Step S105: Using a preset bidding algorithm, bids are submitted for the target task according to the order of the bidding sequence to obtain the target task with the highest bid.

[0076] Step S106: Perform target value decay and tolerance constraint processing on the target task with the highest bid to obtain the target UAV platform with multiple target tasks.

[0077] In this embodiment, the task allocation method can be determined according to the actual situation and is not limited here. As an example, the task allocation method can be a drone platform task allocation method based on sequential ranking and collaborative bidding.

[0078] The drone platform can be determined based on actual circumstances and is not limited here. As an example, the drone platform may include multiple drones.

[0079] In step S101, the first point set can be denoted as point set A; point set A can be understood as a set of m (m is an integer greater than or equal to 2) unmanned aerial vehicle (UAV) platforms; the second point set can be denoted as point set B; point set B can be understood as a set of n (n is an integer greater than or equal to 2) disaster relief locations. The i-th UAV platform can be denoted as... The j-th location requiring disaster relief can be denoted as... As an example, m equals 2, n equals 3; the point set A includes , Point set B includes , , .

[0080] In practical applications, the unmanned aerial vehicle (UAV) platforms and the disaster relief locations are divided into two sets of points. The relationships between these two sets can be represented by a dual graph. In the dual graph, point set A represents the set of m UAV platforms, and point set B represents the set of n disaster relief locations. If the unmanned aerial vehicle platforms... Capable of performing tasks If there is a connection between their abilities, then there exists a link between them.

[0081] In step S102, the preset auction algorithm can be determined according to the actual situation, and is not limited here. As an example, target allocation based on the auction algorithm can be understood as a distributed resource allocation method. Its core idea is to dynamically allocate tasks or targets to the optimal execution entity by simulating the price competition mechanism in the auction market.

[0082] The first drone platform is any drone platform in the first set of points; the first drone platform can be denoted as... The first location is any location in the second set of locations awaiting disaster relief; the first location can be denoted as... If there is no platform Capable of performing tasks If there is a connection between their abilities, then there exists a link between them.

[0083] In step S103, the specific determination process for determining the bidding priority parameter of the target task can be determined according to the actual situation and is not limited here. As an example, determining the bidding priority parameter of the target task may include: obtaining the value parameter of the target task and the corresponding first weight factor, as well as the total number of resources of the first UAV platform and the corresponding second weight factor; and determining the bidding priority parameter of the target task based on the value parameter, the first weight factor, the total number of resources, and the second weight factor.

[0084] In step S104, the specific generation process of generating the bidding sequence for the drone platform based on the bidding priority parameter and the preset roulette wheel betting method can be determined according to the actual situation and is not limited here. As an example, generating the bidding sequence for the drone platform based on the bidding priority parameter and the preset roulette wheel betting method may include: converting the bidding priority parameter to obtain the probability parameter of the first drone platform being selected; determining the probability interval of the probability parameter distribution; generating a random number for the first drone platform based on the probability parameter and the probability interval; and determining the bidding sequence for the drone platform according to the roulette wheel betting method and the random number.

[0085] In step S105, the specific processing procedure for applying target value decay and tolerance constraint to the target task with the highest bid to obtain the target UAV platform allocated multiple target tasks can be determined according to the actual situation and is not limited here. As an example, the target value decay includes value decay within the same round and value decay between different rounds; applying target value decay and tolerance constraint to the target task with the highest bid to obtain the target UAV platform allocated multiple target tasks may include: applying value decay processing within the same round to the target task with the highest bid to obtain the current first value of the target task with the highest bid; applying value decay processing between different rounds to the first value to obtain the current second value of the target task with the highest bid; and applying tolerance constraint processing based on the second value to obtain the target UAV platform allocated multiple target tasks.

[0086] This invention provides a task allocation method. It divides the drone platform and the disaster relief location into a first set and a second set of points. The relationship between the first and second sets of points is represented by a bipartite graph. Based on a preset auction algorithm, target tasks are allocated between the first drone platform and the first location. The auction priority parameters of the target tasks are determined. An auction sequence of drone platforms is generated based on the auction priority parameters and a preset roulette wheel algorithm. A preset bidding algorithm is used to bid on the target tasks according to the order of the auction sequence, obtaining the target task with the highest bid. The target task with the highest bid is subjected to target value decay and tolerance constraints, resulting in a target drone platform allocated multiple target tasks. This method effectively improves resource allocation efficiency, enhances the real-time performance of the algorithm, and solves problems such as low coverage of single drones and inability to complete disaster relief work in a timely manner through the synergistic optimization of target value decay and tolerance constraints.

[0087] In some embodiments, determining the bidding priority parameter for the target task includes:

[0088] Obtain the value parameters of the target task and the corresponding first weighting factor, as well as the total number of resources of the first UAV platform and the corresponding second weighting factor;

[0089] The bidding priority parameter of the target task is determined based on the value parameter, the first weighting factor, the total number of resources, and the second weighting factor.

[0090] In this embodiment, the specific acquisition process for obtaining the value parameters of the target task can be determined according to actual circumstances and is not limited here. As an example, obtaining the value parameters of the target task may include: obtaining the importance parameter of the target task and the success rate parameter of the first resource of the first UAV platform executing the target task; using the importance parameter and the success rate parameter as the value parameters. The importance parameter of the target task can be denoted as... The success rate parameter of the first resource of the first UAV platform executing the target task can be denoted as: .

[0091] The first weighting factor, the total number of resources, and the second weighting factor can be determined according to actual circumstances, and are not limited here. As an example, the first weighting factor can be denoted as... The second weighting factor can be denoted as... The total number of resources can be denoted as... ; Understanding as a platform The total number of resources possessed.

[0092] The specific determination process for determining the bidding priority parameter of the target task based on the value parameter, the first weighting factor, the total number of resources, and the second weighting factor can be determined according to actual circumstances and is not limited here. As an example, determining the bidding priority parameter of the target task based on the value parameter, the first weighting factor, the total number of resources, and the second weighting factor can be achieved by determining the bidding priority parameter of the target task based on the value parameter, the first weighting factor, the total number of resources, and the second weighting factor using a preset algorithm; wherein, the preset algorithm can be determined according to actual circumstances and is not limited here. As an example, the preset algorithm can refer to... .

[0093] In practical applications, the first weighting factor and the second weighting factor can also be understood as adjustment coefficients; the priority calculation formula is the basis for the auction sequence generation rule in this algorithm. The priority calculation formula in this algorithm... in, For the task Its value, that is, its importance. For the platform The Individual resource execution tasks The success rate For the platform Total number of resources possessed , This is an adjustment coefficient, and its specific function is to... Weighted assessment, highlighting the ability to complete high-value tasks, through... Weighted allocation, which prioritizes platforms with sufficient resources, should be determined based on the actual situation. The specific steps for generating the rules are as follows: First, calculate the energy efficiency of a single resource task, using the following formula: Second, sum the platform's total energy efficiency; the total energy efficiency of a single platform is the sum of the energy efficiencies of its individual resources. Third, calculate the resource quantity score, i.e. Fourth, calculate the priority based on the overall calculation, that is... .

[0094] In some embodiments, obtaining the value parameters of the target task includes:

[0095] Obtain the importance parameter of the target task and the success rate parameter of the first resource of the first UAV platform in executing the target task;

[0096] The importance parameter and the success rate parameter are used as the value parameter.

[0097] In this embodiment, the importance parameter of the target task and the success rate parameter of the first resource of the first UAV platform executing the target task can both be determined according to the actual situation, and are not limited here. As an example, the importance parameter of the target task can be denoted as... The success rate parameter of the first resource of the first UAV platform executing the target task can be denoted as: .

[0098] In practical applications, For the task The importance of For the platform The Individual resource execution tasks The success rate For the platform The total number of resources possessed.

[0099] In some embodiments, generating the bidding sequence for the drone platform based on the bidding priority parameter and a preset roulette wheel betting method includes:

[0100] The bidding priority parameter is converted to obtain the probability parameter of the first drone platform being selected;

[0101] Determine the probability interval of the probability parameter distribution;

[0102] Random numbers for the first UAV platform are generated based on the probability parameters and the probability interval;

[0103] The bidding sequence for the drone platform is determined based on the roulette wheel betting method and the random number.

[0104] In this embodiment, the specific conversion process for the auction priority parameter to obtain the probability parameter of the first drone platform being selected can be determined according to the actual situation and is not limited here. As an example, the probability parameter can be denoted as... The conversion of the auction priority parameter to obtain the probability parameter of the first drone platform being selected can be achieved by converting the auction priority parameter using a preset algorithm. The preset algorithm can be determined based on actual circumstances and is not limited here. As an example, the preset algorithm can refer to... in, It is the first The probability of an unmanned platform being selected.

[0105] The specific process for determining the probability interval of the probability parameter distribution can be determined according to the actual situation and is not limited here. As an example, the probability interval can be [0,1].

[0106] The specific generation process for generating random numbers for the first UAV platform based on the probability parameters and the probability interval can be determined according to actual circumstances and is not limited here. As an example, the random number can be denoted as r.

[0107] The specific determination process for determining the bidding sequence of the drone platform based on the roulette wheel betting method and the random number can be determined according to the actual situation and is not limited here. As an example, determining the bidding sequence of the drone platform based on the roulette wheel betting method and the random number can be done by generating the bidding sequence of the drone platform based on the roulette wheel betting method and the random number.

[0108] In practical applications, a roulette wheel selection method is used to generate a dynamic bidding sequence. The priority of each platform has already been generated, and this priority is used to determine the order of the different platforms. The reason for using the roulette wheel selection method is that platforms with higher priority have a higher probability of being allocated the opportunity to execute tasks, while also ensuring that other platforms also have a chance to obtain priority. The priority of each platform is converted into a probability using the following formula: in, It is the first The probability of each unmanned platform being selected is calculated. After generating the probability for each unmanned platform, a probability interval [0,1] is generated, and the probabilities of each platform are distributed within this interval. A random number r is then generated to determine which platform is selected. In each iteration, the roulette wheel method needs to be executed N times (N being the total number of unmanned platforms) to generate a complete auction sequence.

[0109] In some embodiments, the step of using a preset bidding algorithm to bid on the target task according to the order of the auction sequence to obtain the target task with the highest bid includes:

[0110] Obtain the current highest bid and the dynamically decayed task value for the target task;

[0111] The bidding revenue of the first resource of the first UAV platform for the target task is determined based on the current highest bid and the dynamically decayed task value.

[0112] The target task with the highest bid is determined based on the bidding revenue and the current highest bid for the target task.

[0113] In this embodiment, the current highest bid for the target task, the dynamically decayed task value, the bidding revenue, and the target task with the highest bid can all be determined based on actual circumstances, and are not limited here. As an example, the current highest bid for the target task can be denoted as... The dynamically decayed task value can be denoted as... The bidding proceeds can be recorded as... The target task with the highest bid can be denoted as... .

[0114] The specific process for determining the bidding revenue of the first resource of the first UAV platform for the target task based on the current highest bid and the dynamically decayed task value can be determined according to actual circumstances and is not limited here. As an example, determining the bidding revenue of the first resource of the first UAV platform for the target task based on the current highest bid and the dynamically decayed task value can be achieved by using a preset algorithm to determine the bidding revenue of the first resource of the first UAV platform for the target task based on the current highest bid and the dynamically decayed task value. The preset algorithm can be determined according to actual circumstances and is not limited here. As an example, the preset algorithm can refer to... .

[0115] The specific process for determining the highest-priced target task based on the bidding revenue and the current highest bid for the target task can be determined according to actual circumstances and is not limited here. As an example, determining the highest-priced target task based on the bidding revenue and the current highest bid for the target task can be achieved by using a preset algorithm based on the bidding revenue and the current highest bid for the target task; wherein, the preset algorithm can be determined according to actual circumstances and is not limited here. As an example, the preset algorithm can refer to... .

[0116] In practical applications, after determining the bidding sequence, the unmanned platforms need to bid on tasks in order. Each unmanned platform has multiple resources, therefore each platform needs to calculate the expected revenue for each task using its unallocated resources. The formula for calculating the bidding revenue is: in, For the platform The Resource to task The bidding revenue, The current dynamically decayed task value, Indicates task The highest bid is the amount the platform needs to subtract when calculating the net profit from bidding on a task. Therefore, the higher the price, the lower the net profit. Calculate all The largest will be selected. As And obtain the value excluding the maximum value. Maximum value of the remaining array The final step involves allocating specific resources of the unmanned platform to the task with the highest potential return. After calculating the bidding returns and determining the bidding tasks, the unmanned platform needs to submit its bid. The platform's bid price is shown in the formula below: ;in, The winning bid price, This is used to break ties, and through fine-tuning, ensures a single winner, preventing the algorithm from getting stuck in an infinite loop. Meanwhile, It will be set as the highest bid in the current round after the bidding ends.

[0117] In some embodiments, the target value decay includes value decay within the same round and value decay between different rounds; the step of applying target value decay and tolerance constraint processing to the target task with the highest bid to obtain a target UAV platform assigned multiple target tasks includes:

[0118] The highest-priced target task is subjected to value decay processing within the same round to obtain the current first value of the highest-priced target task.

[0119] The first value is subjected to value decay processing across different rounds to obtain the current second value of the target task with the highest bid.

[0120] Based on the second value, a tolerance constraint is applied to obtain the target UAV platform assigned to multiple target tasks.

[0121] In this embodiment, the specific attenuation process for performing value attenuation processing on the highest-priced target task within the same round to obtain the current first value of the highest-priced target task can be determined according to actual circumstances and is not limited here. As an example, performing value attenuation processing on the highest-priced target task within the same round to obtain the current first value of the highest-priced target task may include: obtaining the initial weight value and a preset attenuation index of the highest-priced target task; and determining the first value based on the initial weight value and the attenuation index.

[0122] The specific attenuation process for applying the value attenuation process across different rounds to the first value to obtain the current second value of the target task with the highest bid can be determined based on actual circumstances and is not limited here. As an example, applying the value attenuation process across different rounds to the first value to obtain the current second value of the target task with the highest bid may include: obtaining the weight and the number of times the bid for a high-value task frequently allocated by the first platform corresponding to the first value; applying the value attenuation process across different rounds to the weight and the number of times the bid for a high-value task is reset to zero to obtain the current second value of the target task with the highest bid.

[0123] The specific tolerance constraint process for obtaining the target UAV platform for multi-objective tasks based on the second value can be determined according to the actual situation and is not limited here. As an example, the tolerance constraint process for obtaining the target UAV platform for multi-objective tasks based on the second value may include: determining the tolerance of the target UAV platform in each round of bidding based on the second value; and determining the target UAV platform for multi-objective tasks based on the tolerance.

[0124] In some embodiments, the step of performing value decay processing on the target task with the highest bid within the same round to obtain the current first value of the target task with the highest bid includes:

[0125] Obtain the initial weight value and preset decay index of the target task with the highest bid;

[0126] The first value is determined based on the initial weight value and the decay index.

[0127] In this embodiment, the initial weight value and preset decay index of the target task with the highest bid can be determined according to the actual situation, and are not limited here. As an example, the initial weight value of the target task with the highest bid can be understood as the original weight, which can be denoted as... The preset decay index can be understood as the task value decreasing exponentially with the number of bids. If the value decreases after the first bid... The second time will be wait.

[0128] The specific determination process for determining the first value based on the initial weight and the decay index can be determined according to actual circumstances and is not limited here. As an example, the first value can be denoted as... The first value can be determined based on the initial weight value and the decay index using a preset algorithm. The preset algorithm can be determined according to actual circumstances and is not limited here. As an example, the preset algorithm can refer to... .

[0129] In practical applications, value decay within the same round is to prevent multiple platforms from concentrating their bidding on the same high-value task in the same round, leading to uneven resource allocation. Specifically, during the bidding phase, when a platform bids for a task... After a bid is submitted, the value of the task will decrease according to the following formula: in, For the task The current value, whose initial value is the original weight. Using this method, the value of the task decreases exponentially with the number of bids; if the initial bid is... The second time will be This reduces the platform's subsequent willingness to select this task, leading them to focus on other tasks and promote diversity in task allocation.

[0130] In some embodiments, the step of performing value decay processing on the first value across different rounds to obtain the current second value of the target task with the highest bid includes:

[0131] Obtain the weight of the high-value tasks frequently assigned by the first platform corresponding to the first value and the number of times the price is reset to zero;

[0132] The weight of the high-value task and the number of times the bid is reset are subjected to value decay processing across different rounds to obtain the current second value of the target task with the highest bid.

[0133] In this embodiment, the weight of the high-value task and the number of times the bid is reset to zero are subjected to value decay processing between different rounds. The specific decay processing process in the current second value of the target task with the highest bid can be determined according to the actual situation and is not limited here.

[0134] In practical applications, value decay between different rounds refers to the reduction of the weight of frequently allocated high-value tasks across different main loop iterations, i.e., different bidding rounds. This is achieved through a tolerance mechanism, which prevents low-value tasks from remaining unallocated for extended periods. If the task set is not fully covered, meaning there are unallocated tasks, the number of times bids are reset to zero in subsequent rounds is reduced, thereby limiting the over-allocation of high-value tasks. Although the original weights... While not directly attenuating, limiting the number of times tasks can be reset reduces the incentive to bid on high-value tasks, thus indirectly balancing task priorities.

[0135] In some embodiments, the tolerance constraint processing based on the second value to obtain the target UAV platform assigned multiple target tasks includes:

[0136] The tolerance level for the target drone platform in each round of bidding is determined based on the second value.

[0137] The target UAV platform for assigning multi-target tasks is determined based on the tolerance level.

[0138] In this embodiment, the specific process for determining the tolerance level of the target drone platform in each round of bidding based on the second value can be determined according to actual circumstances and is not limited here. As an example, the tolerance level can be denoted as... .

[0139] In practical applications, the bid-clearing mechanism can increase bidding enthusiasm for high-value tasks and promote collaborative execution of the same task across multiple platforms and resources, thereby improving the task success rate. However, if task allocation is biased towards high-value tasks, low-value tasks may not be allocated at all. The tolerance mechanism balances the allocation priority of high-value and low-value tasks by dynamically limiting the number of bid clearings, ensuring that all tasks have a chance to be bid on. A tolerance level exists in each round of bidding. This indicates the number of times the remaining bids are reset to zero. During the bidding phase, if platform resources are not allocated and... Then the price of the highest-value task will be reset to zero. ;in, This refers to the number corresponding to the highest-value task at present. When Afterwards, the benefits It will be simplified to At this point, the net profit of all platforms depends solely on their own success rate. The attractiveness of high-value tasks is maximized, thus significantly enhancing the overall appeal of the task. This is achieved by... If the initial bid is zeroed out, the platform can then bid at a lower cost, encouraging more resources to participate and fostering collaborative execution. If an iteration contains uncovered tasks, It will decrease by 1, that is: The number of zeroing attempts available in subsequent iterations decreases, and the algorithm gradually shifts from "exploring high-value tasks" to "covering all tasks" through dynamic balancing.

[0140] In some embodiments, the method further includes:

[0141] A preset weighted fusion model is used to perform multi-objective collaborative optimization on the target UAV platform that is assigned multi-objective tasks, so as to obtain the high-value tasks assigned by the target UAV platform.

[0142] A multi-objective evaluation is performed on the high-value tasks assigned to the target UAV platform to obtain evaluation results; the evaluation results characterize the quality of the task assignments of the target UAV platform.

[0143] In this embodiment, the weighted fusion model can be determined according to the actual situation, and is not limited here. As an example, the weighted fusion model can be a third-order weighted fusion model.

[0144] The process of using a preset weighted fusion model to perform multi-objective collaborative optimization on the target UAV platform, which is assigned multiple objective tasks, to obtain high-value tasks assigned to the target UAV platform can be determined according to actual circumstances and is not limited here. As an example, the process of using a preset weighted fusion model to perform multi-objective collaborative optimization on the target UAV platform, which is assigned multiple objective tasks, to obtain high-value tasks assigned to the target UAV platform can be achieved by using a preset algorithm and a preset weighted fusion model to perform multi-objective collaborative optimization on the target UAV platform, which is assigned multiple objective tasks. The preset algorithm can be determined according to actual circumstances and is not limited here. As an example, the preset algorithm can refer to... ; .

[0145] In practical applications, the multi-objective collaborative optimization mechanism uses a third-order weighted fusion model in its evaluation system, which includes the following core indicators: in, For task rewards, represent the total expected value of completed tasks. This represents the total number of tasks. For the number of unmanned platforms, It is a binary variable, representing the platform. resources Assign to task By using a probability product model, the cumulative success probability of multi-platform resource collaborative execution of tasks is quantified, thus avoiding the risk of task failure due to the failure of a single resource. in, The total cost of platform resources consumed. For the platform Use resources The cost reflects the economic constraints of resources. in, As a penalty item, Let it be a binary variable representing the task. Whether the task has been assigned is indicated by 1 (assigned) or 0 (unassigned). If a task is not assigned, its value is factored into a penalty term, forcing the algorithm to adjust its allocation strategy. This penalty coefficient strengthens the coverage of critical tasks. The core purpose of the penalty function is to force the algorithm to cover as many tasks as possible, avoiding situations where some tasks remain completely unassigned. Initially, the algorithm prioritizes high-value tasks due to their higher rewards, but by penalizing unassigned low-value tasks, it gradually shifts resources towards the remaining tasks. Therefore, the penalty term prevents local optima, where the algorithm only assigns high-value tasks.

[0146] The specific evaluation process for obtaining the evaluation result by performing a multi-objective evaluation of the high-value tasks assigned to the target UAV platform can be determined according to the actual situation and is not limited here. As an example, the multi-objective evaluation of the high-value tasks assigned to the target UAV platform can be performed using a preset algorithm to obtain the evaluation result. The preset algorithm can be determined according to the actual situation and is not limited here. As an example, the preset algorithm can refer to...

[0147] In practical applications, It is a comprehensive evaluation value for multi-objective optimization, used to quantify the overall merits of task allocation schemes. It unifies three conflicting objectives—task completion benefits, resource consumption costs, and task coverage penalties—into a single optimizable index, as shown in the following formula: ; For the target item, As a weighting coefficient, it is used to balance the trade-off between the benefits of task completion and the execution costs. At this time, the algorithm focuses more on task completion. At this time, the algorithm tends to save resources. This is the penalty coefficient, which is modified according to the actual situation to adjust the weight of the penalty item.

[0148] In practical applications, a specific task allocation method can be a drone platform task allocation method based on sequential ranking and collaborative bidding. Through the collaborative design of a dynamic target value decay mechanism, a bid clearing strategy, and a tolerance mechanism, combined with a priority-driven bidding sequence generation method, it solves the problems of insufficient allocation of high-value task resources, incomplete coverage of low-value tasks, and poor real-time performance in traditional auction algorithms. This achieves efficient many-to-many matching of tasks and resources in highly dynamic environments. Specifically, the dual value decay mechanism suppresses task clustering and balances task priority weights, respectively; the bid clearing strategy strengthens the resource attraction capacity of high-value tasks through price reset; and the tolerance mechanism dynamically constrains the number of clearings to ensure global coverage of the task set, forming a collaborative optimization paradigm that prioritizes the allocation of high-value tasks and guarantees full task coverage.

[0149] The UAV platform task allocation method based on sequential collaborative bidding proposed in this application consists of four parts: a dynamic generation method for bidding priorities, a target value decay and bid clearing mechanism, target value decay and tolerance constraints, and a multi-objective optimization evaluation system. Specifically, it may include the following steps:

[0150] 1.1 Target allocation based on auction algorithm.

[0151] Target allocation based on auction algorithms is a distributed resource allocation method. Its core idea is to dynamically allocate tasks or targets to the optimal execution entity by simulating the price competition mechanism in the auction market.

[0152] The target assignment problem can be described using bipartite graph theory. The drone platform and the disaster relief locations are divided into two sets of points, and the relationships between these two sets can be represented by a bipartite graph. For example... Figure 2 As shown, Figure 2 This is a schematic diagram illustrating the relationship between two point sets in an embodiment of the present invention, represented by a bipartite graph. In the bipartite graph, point set A is a set of m unmanned aerial vehicle (UAV) platforms, and point B is a set of n disaster relief locations. If the UAV platforms... Capable of performing tasks There is a connection between their capabilities, and which platform prioritizes executing tasks, and which task to execute, needs to be determined by further algorithm design.

[0153] 1.2 Method for dynamically generating bidding priorities.

[0154] When a disaster occurs, the urgency varies from location to location, thus requiring different priorities to be assigned. This mechanism dynamically determines the bidding order for each drone platform. The system calculates priority based on each platform's resource availability and mission success rate; platforms with stronger capabilities and higher efficiency will receive priority bidding rights, and random selection will be used to increase flexibility.

[0155] 1.2.1 Priority calculation.

[0156] The priority calculation formula forms the basis of the auction sequence generation rules in this algorithm. The priority calculation formula in this algorithm is shown below:

[0157]

[0158] in, For the task Its value, that is, its importance. For the platform The Individual resource execution tasks The success rate For the platform Total number of resources possessed , This is an adjustment coefficient, and its specific function is to... Weighted assessment, highlighting the ability to complete high-value tasks, through... Weighting, which ensures priority for platforms with sufficient resources, should be determined based on the actual situation. The specific steps for generating rules are as follows:

[0159] (1) Calculate the energy efficiency of a single resource task, the specific formula is as follows:

[0160]

[0161] (2) The total energy efficiency of the platform is summed up. The total energy efficiency of a single platform is the sum of the energy efficiency of its individual resources.

[0162] (3) Calculate the resource quantity score, i.e. .

[0163] (4) Calculate the priority of the comprehensive calculation, that is:

[0164]

[0165] 1.2.2 Dynamic auction sequence generation.

[0166] To generate a dynamic bidding sequence, a roulette wheel selection method is required. The priority of each platform was generated in section 2.1.2, and this priority is used to determine the order of different platforms. The reason for using the roulette wheel selection method is that platforms with higher priority have a higher probability of being allocated the opportunity to execute tasks, while also ensuring that other platforms also have a chance to obtain priority. The priority of each platform is converted into a probability using the following formula:

[0167]

[0168] in, It is the first The probability of each drone platform being selected is calculated. After generating the probability for each drone platform, a probability interval [0,1] is generated, and the probabilities of each platform are distributed within this interval. A random number r is then generated to determine which platform is selected. In each iteration, the roulette wheel algorithm needs to be executed N times (N being the total number of drone platforms) to generate a complete auction sequence.

[0169] 1.2.3 Bidding algorithm for pricing.

[0170] After the bidding sequence is determined, the drone platforms need to bid on the tasks in order. Each drone platform has multiple resources, therefore each platform needs to calculate the expected revenue for each task using its unallocated resources. The formula for calculating the bidding revenue is:

[0171]

[0172] in, For the platform The Resource to task The bidding revenue, The current dynamically decayed task value, Indicates task The highest bid is the amount the platform needs to subtract when calculating the net profit from bidding on a task. Therefore, the higher the price, the lower the net profit. Calculate all The largest will be selected. As And obtain the value excluding the maximum value. Maximum value of the remaining array Ultimately, specific resources for the drone platform are allocated to bidding for the missions with the highest returns.

[0173] After calculating the bidding revenue and determining the bidding tasks, the drone platform needs to submit a bid. The platform's bid price is shown in the formula:

[0174]

[0175] in, The winning bid price, This is used to break ties, and through fine-tuning, ensures a single winner, preventing the algorithm from getting stuck in an infinite loop. Meanwhile, It will be set as the highest bid in the current round after the bidding ends.

[0176] 1.3 Target value decay and tolerance constraints.

[0177] 1.3.1 Target value decay.

[0178] The target value decay mechanism is divided into two levels: within the same round and between different rounds. It aims to dynamically adjust the attractiveness of tasks, avoid excessive concentration of resources and ignore low-value tasks, and ultimately ensure that every task can be executed.

[0179] (1) Value decay within the same round.

[0180] Value decay within the same round is designed to prevent multiple platforms from concentrating their bidding on the same high-value task in the same round, leading to uneven resource allocation. Specifically, during the bidding phase, when a platform bids for a task... After a bid is submitted, the value of the task will decrease according to the following formula:

[0181]

[0182] in, For the task The current value, whose initial value is the original weight. Using this method, the value of the task decreases exponentially with the number of bids; if the initial bid is... The second time will be This reduces the platform's subsequent willingness to select this task, leading them to focus on other tasks and promote diversity in task allocation.

[0183] (2) Value decay between different rounds.

[0184] Value decay between different rounds refers to the reduction of the weight of frequently allocated high-value tasks across different main loop iterations, i.e., different bidding rounds. This is achieved through a tolerance mechanism, which prevents low-value tasks from remaining unallocated for extended periods. If the task set is not fully covered, meaning there are unallocated tasks, the number of times bids are reset to zero in subsequent rounds is reduced, thereby limiting the over-allocation of high-value tasks. Although the original weights... While not directly attenuating, limiting the number of times tasks can be reset reduces the incentive to bid on high-value tasks, thus indirectly balancing task priorities.

[0185] 1.3.2 Collaborative bidding process under tolerance constraints.

[0186] The zero-bid mechanism can enhance the bidding enthusiasm for high-value tasks and promote the collaborative execution of the same task across multiple platforms and resources, thereby improving the task success rate.

[0187] In task allocation, an overemphasis on high-value tasks may result in low-value tasks remaining unassigned. The tolerance mechanism balances the allocation priority between high-value and low-value tasks by dynamically limiting the number of times "bids are reset," ensuring that all tasks have a chance to be bid on.

[0188] There is a tolerance level in each round of bidding. This indicates the number of times the remaining bids are reset to zero. During the bidding phase, if platform resources are not allocated and... Then the price of the highest-value task will be reset to zero, that is:

[0189]

[0190] in, This refers to the number corresponding to the highest-value task at present. When Then, the returns in formula (5) It will be simplified to At this point, the net profit of all platforms depends solely on their own success rate. The attractiveness of high-value tasks is maximized, thus significantly enhancing the overall appeal of the task. This is achieved by... If the initial bid is reduced to zero, the platform can then bid at a lower cost, encouraging more resources to participate and forming a collaborative execution.

[0191] If there are uncovered tasks in a certain iteration It will decrease by 1, that is:

[0192]

[0193] The number of zeroing attempts available in subsequent iterations decreases, and the algorithm gradually shifts from "exploring high-value tasks" to "covering all tasks" through dynamic balancing.

[0194] 1.4 Multi-objective optimization evaluation system.

[0195] 1.4.1 Multi-objective collaborative optimization mechanism.

[0196] The evaluation system adopts a third-order weighted fusion model, which includes the following core indicators:

[0197]

[0198] in, For task rewards, represent the total expected value of completed tasks. This represents the total number of tasks. For the number of drone platforms, It is a binary variable, representing the platform. resources Assign to task By using a probability product model, the cumulative success probability of multi-platform resource collaborative execution of tasks is quantified, thus avoiding the risk of task failure due to the failure of a single resource.

[0199]

[0200] in, The total cost of platform resources consumed. For the platform Use resources The cost reflects the economic constraints of resources.

[0201]

[0202] in, As a penalty item, Let it be a binary variable representing the task. Whether the task has been assigned is indicated by 1 (assigned) or 0 (unassigned). If a task is not assigned, its value is factored into a penalty term, forcing the algorithm to adjust its allocation strategy. This penalty coefficient strengthens the coverage of critical tasks. The core purpose of the penalty function is to force the algorithm to cover as many tasks as possible, avoiding situations where some tasks remain completely unassigned. Initially, the algorithm prioritizes high-value tasks due to their higher rewards, but by penalizing unassigned low-value tasks, it gradually shifts resources towards the remaining tasks. Therefore, the penalty term prevents local optima, where the algorithm only assigns high-value tasks.

[0203] 1.4.2 Multi-objective optimization and comprehensive evaluation.

[0204] It is a comprehensive evaluation value for multi-objective optimization, used to quantify the overall merits of task allocation schemes. It unifies three conflicting objectives—task completion benefits, resource consumption costs, and task coverage penalties—into a single optimizable index, as shown in the following formula:

[0205]

[0206] For the target item, As a weighting coefficient, it is used to balance the trade-off between the benefits of task completion and the execution costs. At this time, the algorithm focuses more on task completion. At this time, the algorithm tends to save resources. This is the penalty coefficient, which is modified according to the actual situation to adjust the weight of the penalty item.

[0207] Figure 4 This is a schematic diagram illustrating the target allocation when the number of platforms is 4 and the number of task objectives is 5, as shown in the embodiments of this application. Figure 6 This is a schematic diagram illustrating the target allocation when the number of platforms is 4 and the number of task objectives is 30, as shown in the embodiments of this application; Figure 4 and Figure 6 The figures illustrate the effectiveness of the proposed method, showing the target allocation when the number of platforms is 4 and the task objectives are 5 and 30 respectively. Figure 3 and Figure 5 The rapid convergence of the convergence curve indicates that the algorithm can provide a feasible solution for the UAV platform within a finite time, while the point set graph shows that the platform can achieve reasonable allocation of resources.

[0208] This invention proposes a task allocation method for UAV platforms based on sequential ranking and collaborative bidding. It adopts a priority-driven dynamic bidding sequence generation method, which effectively improves resource allocation efficiency and enhances the real-time performance of the algorithm. Through the collaborative optimization of target value decay and tolerance constraints, it solves the problems of insufficient resource allocation for high-value tasks and incomplete coverage of low-value tasks in traditional auction algorithms.

[0209] Based on the same inventive concept as described above Figure 7 This is a schematic diagram of a task allocation device provided in an embodiment of the present invention, as shown below. Figure 7 As shown, the device 500, set on a drone platform for sequential collaborative bidding, includes:

[0210] The partitioning unit 501 is used to divide the UAV platform and the disaster relief location into a first point set and a second point set; the relationship between the first point set and the second point set is represented by a bipartite graph.

[0211] The allocation unit 502 is used to allocate target tasks between a first drone platform and a first location based on a preset bidding algorithm; the first drone platform is any drone platform in the first point set; the first location is any disaster relief location in the second point set.

[0212] The determining unit 503 is used to determine the bidding priority parameters of the target task;

[0213] The generation unit 504 is used to generate the bidding sequence of the drone platform based on the bidding priority parameters and the preset roulette method;

[0214] Bidding unit 505 is used to bid on the target task according to the order of the bidding sequence using a preset bidding algorithm, and obtain the target task with the highest bid.

[0215] Processing unit 506 is used to perform target value decay and tolerance constraint processing on the target task with the highest bid to obtain a target UAV platform with multiple target tasks.

[0216] In some embodiments, the determining unit 503 is further configured to obtain the value parameter of the target task and the corresponding first weight factor, as well as the total number of resources of the first UAV platform and the corresponding second weight factor; and determine the bidding priority parameter of the target task based on the value parameter, the first weight factor, the total number of resources and the second weight factor.

[0217] In some embodiments, the determining unit 503 is further configured to obtain the importance parameter of the target task and the success rate parameter of the first resource of the first UAV platform executing the target task; and use the importance parameter and the success rate parameter as the value parameter.

[0218] In some embodiments, the generation unit 504 is further configured to: convert the auction priority parameter to obtain a probability parameter for the first drone platform to be selected; determine a probability interval of the probability parameter distribution; generate a random number for the first drone platform based on the probability parameter and the probability interval; and determine the auction sequence for the drone platform according to the roulette wheel method and the random number.

[0219] In some embodiments, the bidding unit 505 is further configured to obtain the current highest bid and the dynamically decayed task value of the target task; determine the bidding revenue of the first resource of the first UAV platform for the target task based on the current highest bid and the dynamically decayed task value of the target task; and determine the target task with the highest bid based on the bidding revenue and the current highest bid of the target task.

[0220] In some embodiments, the target value decay includes value decay within the same round and value decay between different rounds; the processing unit 506 is further configured to perform value decay processing within the same round on the target task with the highest bid to obtain the current first value of the target task with the highest bid; perform value decay processing between different rounds on the first value to obtain the current second value of the target task with the highest bid; and perform tolerance constraint processing based on the second value to obtain the target UAV platform assigned multiple target tasks.

[0221] In some embodiments, the processing unit 506 is further configured to obtain the initial weight value and a preset decay index of the target task with the highest bid; and determine the first value based on the initial weight value and the decay index.

[0222] In some embodiments, the processing unit 506 is further configured to obtain the weight and the number of times the price of a high-value task frequently allocated by the first platform corresponding to the first value; and to perform value decay processing on the weight and the number of times the price of the high-value task is reset between different rounds to obtain the current second value of the target task with the highest price.

[0223] In some embodiments, the processing unit 506 is further configured to determine the tolerance level of the target drone platform in each round of bidding based on the second value; and to determine the target drone platform for assigning multi-target tasks based on the tolerance level.

[0224] In some embodiments, the apparatus 500 further includes an evaluation unit; wherein,

[0225] The processing unit 506 is further configured to perform multi-objective collaborative optimization processing on the target UAV platform that is assigned multi-objective tasks using a preset weighted fusion model, so as to obtain the high-value tasks assigned by the target UAV platform.

[0226] The evaluation unit is used to perform multi-objective evaluation of the high-value tasks assigned to the target UAV platform and obtain evaluation results; the evaluation results characterize the quality of the task assignment of the target UAV platform.

[0227] It should be noted that the task allocation device provided in the embodiments of the present invention and the configuration method provided in the aforementioned embodiments of the present invention belong to the same inventive concept. The meanings of the terms appearing here have been explained in detail above and will not be repeated here.

[0228] This invention also provides a storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0229] This invention also provides a task allocation device, which includes a processor and a memory for storing a computer program that can run on the processor, wherein when the processor runs the computer program, it executes the steps of the method embodiments described above stored in the memory.

[0230] Figure 8 This is a schematic diagram of a hardware structure of a task allocation device according to an embodiment of the present invention. The task allocation device 600 includes at least one processor 601 and a memory 602. Optionally, the task allocation device 600 may further include at least one communication interface 603. The various components in the task allocation device 600 are coupled together through a bus system 604. It can be understood that the bus system 604 is used to realize the connection and communication between these components. In addition to a data bus, the bus system 604 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in... Figure 8 The general designated all buses as Bus System 604.

[0231] It is understood that memory 602 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Sync Link Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 602 described in this embodiment of the invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0232] The memory 602 in this embodiment of the invention is used to store various types of data to support the operation of the task allocation device 600. Examples of such data include any computer program for operation on the task allocation device 600, and programs implementing the methods of this embodiment of the invention may be included in the memory 602.

[0233] The methods disclosed in the above embodiments of the present invention can be applied to or implemented by processor 601. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of the present invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in a memory. The processor reads information from the memory and, in conjunction with its hardware, completes the steps of the aforementioned method.

[0234] In an exemplary embodiment, the task allocation device 600 may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the methods described above.

[0235] In the several embodiments provided by this invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms. The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units; some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs. In addition, all functional units in the various embodiments of this invention can be integrated into one processing module, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated units can be implemented in hardware or in the form of hardware plus software functional units.

[0236] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A method of task allocation, characterized by, The method is applied to a UAV platform based on sequential ranking cooperative bidding, and the method comprises the following steps: Dividing the UAV platform and disaster relief sites into a first point set and a second point set; the relationship between the first point set and the second point set is represented by a bipartite graph; Assigning a target task between a first UAV platform and a first site based on a preset bidding algorithm; the first UAV platform is any UAV platform in the first point set; the first site is any disaster relief site in the second point set; Determining a bidding priority parameter of the target task; Generating a bidding sequence of the UAV platform based on the bidding priority parameter and a preset roulette method; Bidding for the target task in the order of the bidding sequence by using a preset bid bidding algorithm, and obtaining a target task with the highest bid; Processing the target value attenuation and tolerance constraint of the target task with the highest bid to obtain a target UAV platform assigned with multiple target tasks; The method of generating the bidding sequence of the UAV platform based on the bidding priority parameter and the preset roulette method comprises the following steps: Converting the bidding priority parameter to obtain a probability parameter of the first UAV platform being selected; Determining a probability interval of the probability parameter; Generating a random number of the first UAV platform based on the probability parameter and the probability interval; Determining the bidding sequence of the UAV platform according to the roulette method and the random number.

2. The task allocation method according to claim 1, characterized in that, The method of determining the bidding priority parameter of the target task comprises the following steps: Obtaining a value parameter of the target task and a corresponding first weight factor, and a total number of resources of the first UAV platform and a corresponding second weight factor; Determining the bidding priority parameter of the target task based on the value parameter, the first weight factor, the total number of resources and the second weight factor.

3. The task allocation method according to claim 2, wherein, The method of obtaining the value parameter of the target task comprises the following steps: Obtaining an importance parameter of the target task and a success rate parameter of a first resource of the first UAV platform in executing the target task; Taking the importance parameter and the success rate parameter as the value parameter.

4. The task allocation method according to claim 1, characterized by, The method of bidding for the target task in the order of the bidding sequence by using a preset bid bidding algorithm to obtain a target task with the highest bid comprises the following steps: Obtaining a current highest bid of the target task and a task value after dynamic attenuation; Determining a bidding income of a first resource of the first UAV platform for the target task based on the current highest bid of the target task and the task value after dynamic attenuation; Determining the target task with the highest bid according to the bidding income and the current highest bid of the target task.

5. The task allocation method according to claim 1, wherein, The target value attenuation comprises value attenuation within the same round and value attenuation between different rounds; the method of processing the target value attenuation and tolerance constraint of the target task with the highest bid to obtain a target UAV platform assigned with multiple target tasks comprises the following steps: Processing the value attenuation within the same round of the target task with the highest bid to obtain a first value of the target task with the highest bid at present; The first value is subjected to value decay processing between different rounds to obtain a second value of the target task with the highest bid at present; Based on the second value, tolerance constraint processing is performed to obtain a target unmanned aerial vehicle platform for allocating multiple target tasks.

6. The task allocation method according to claim 5, wherein, The value decay processing of the target task with the highest bid in the same round obtains a first value of the target task with the highest bid at present, which comprises: Obtaining the initial value of the weight of the target task with the highest bid and the preset decay index; Determine the first value according to the initial value of the weight and the decay index.

7. The task allocation method according to claim 5, wherein, The value decay processing of the first value between different rounds obtains a second value of the target task with the highest bid at present, which comprises: Obtaining the weight of the high-value task frequently allocated by the first platform corresponding to the first value and the number of bid clearing; The weight of the high-value task and the number of bid clearing are subjected to value decay processing between different rounds to obtain a second value of the target task with the highest bid at present.

8. The task allocation method of claim 5, wherein, Based on the second value, tolerance constraint processing is performed to obtain a target unmanned aerial vehicle platform for allocating multiple target tasks, which comprises: Based on the second value, the tolerance of the target unmanned aerial vehicle platform existing in each round of auction is determined; According to the tolerance, the target unmanned aerial vehicle platform for allocating multiple target tasks is determined.

9. The task allocation method according to any one of claims 1 to 8, characterized in that, The method further comprises: Using a preset weighted fusion model to perform multi-target collaborative optimization processing on the target unmanned aerial vehicle platform for allocating multiple target tasks to obtain a high-value task allocated by the target unmanned aerial vehicle platform; Multi-target evaluation is performed on the high-value task allocated by the target unmanned aerial vehicle platform to obtain an evaluation result; the evaluation result represents the degree of advantage or disadvantage of the task allocation of the target unmanned aerial vehicle platform.

Images