A multi-AUV task allocation method based on consensus bundling algorithm

Abstract

The present invention discloses a multi-AUV task allocation method based on a consensus bundling algorithm, including: step S1, obtaining a task list and an AUV list to be allocated, constructing a data structure required by the algorithm, and initializing various parameters; step S2, considering the communication distance and the path loss of underwater acoustic communication during the communication process, and improving the communication topology network matrix in the algorithm; step S3, formulating corresponding solutions for tasks requiring multi-AUV cooperation and long-distance communication tasks; step S4, processing all tasks based on the improved algorithm. Aiming at the weak communication conditions underwater, the present invention uses underwater acoustic communication loss as an element of the communication topology network adjacency matrix, and discards low-quality and high-loss communication processes by setting a communication quality threshold, thereby improving the communication quality of the system and reducing the communication load and cost.

Description

Technical Field

[0001] The present application belongs to the technical field of multi-agent distributed task planning, and specifically relates to a multi-AUV task allocation method based on a consensus bundling algorithm. Background Art

[0002] AUV (Autonomous Underwater Vehicle) is a new type of autonomous underwater vehicle that has received extensive attention and application in the field of marine engineering in recent years. A single AUV is easily restricted by factors such as operating efficiency, energy consumption, intelligence and environment, and cannot complete complex tasks independently. The multi-AUV system proposed on this basis can not only meet the operational requirements of complex tasks, but also take into account the optimization of multiple attributes such as improving operating efficiency and reducing operating costs. The tasks performed by AUVs are usually path tracking, hovering, exploration and search, etc. According to the needs of the task, they can be divided into single AUV execution tasks and multi-AUV execution tasks.

[0003] Task allocation is usually divided into two methods: centralized allocation and distributed allocation according to the communication mode of multiple AUVs. The centralized method requires the system's only central controller to complete the best matching calculation of all tasks and AUVs through analysis and processing of global multi-source heterogeneous sensor data information. Even if it can obtain the global optimal allocation plan, the low robustness and redundant computing load lead to high latency and low security of the centralized allocation method. The distributed method completes real-time information interaction through the controllers attached to each AUV, so the consistency of data information is greatly improved. In addition, the failure of a single AUV will not affect the communication paralysis of the entire system, and the system has strong robustness.

[0004] In ocean scenarios, due to the presence of dynamic and complex ocean currents, reflections from the surface and bottom of the water, and absorption by seawater, the conditions for underwater acoustic communication cannot be favorable, the quality of underwater acoustic communication will be compromised during signal propagation, and the real-time performance of underwater acoustic communication will be reduced. Distributed methods can provide the system with greater robustness in complex environments. Each AUV can make independent decisions, and the allocation plan is determined by communication negotiation between AUVs.

[0005] The Consensus-Based Bundle Algorithm (CBBA) is a distributed task allocation algorithm that uses a market-based auction strategy as a distributed task selection mechanism and a local communication-based consensus strategy as a conflict mediation solution, combining the advantages of the two to quickly achieve a conflict-free allocation solution. However, the traditional CBBA algorithm roughly divides the communication network between AUVs into two states: communicable and non-communicable, based on the communication network consensus strategy as a conflict resolution mechanism, ignoring the degradation of communication quality caused by the underwater acoustic communication propagation process. When facing long-distance tasks that exceed the communication distance of the AUV, the existence of such tasks will be ignored because the connectivity of the communication network cannot be guaranteed, and the system requirements cannot be met normally. Summary of the invention

[0006] The present invention discloses a multi-AUV task allocation method based on a consensus bundling algorithm, which can effectively solve the technical problems involved in the background technology.

[0007] To achieve the above object, the technical solution of the present invention is:

[0008] A multi-AUV task allocation method based on a consensus bundling algorithm comprises the following steps:

[0009] Step S1, obtain the task list and AUV list to be assigned, build the data structure required by the algorithm, and initialize various parameters;

[0010] Step S2, considering the communication distance and path loss of underwater acoustic communication during the communication process, and improving the communication topology network matrix in the algorithm;

[0011] Step S3, formulate corresponding solutions for tasks requiring multi-AUV cooperation and long-distance communication tasks;

[0012] Step S4, processing all tasks based on the improved algorithm.

[0013] As a preferred improvement of the present invention, in step S1, obtaining the task list J and AUV list I to be assigned specifically includes:

[0014] Step S11, define N t Tasks and N u AUVs, each AUV can be assigned a maximum of L t The number of AUVs required for each task is L. u , then the task list J = {1,2,…,N t}, AUV list I = {1,2,…,N u};

[0015] Step S12, constructing an optimization objective function, which is expressed by the following formula:

[0016]

[0017] Among them, x ij is a binary decision variable, c ij is the profit function;

[0018] Step S13, defining constraint conditions, including:

[0019] C1:

[0020] C2:

[0021] C3:

[0022] C4:

[0023] Where C1 represents the binary decision variable x ij C2 represents the maximum number of tasks that can be assigned to a single AUV, C3 represents that a single task cannot be assigned to multiple AUVs, and C4 represents the termination condition for task assignment, that is, for the decision variable x ij The sum in the direction of the two subscripts should be the number of tasks N t The smaller value of the sum of the number of missions that all AUVs can undertake.

[0024] As a preferred improvement of the present invention, in step S1, the data structure required for constructing the algorithm includes the data structure required for constructing the algorithm in the task package construction phase and the conflict resolution phase, specifically:

[0025] During the mission package construction phase, each AUV will carry four vectors, namely, mission package b i , execution path list p i 、List of successful bidders i And the winning bidder's bid list y i ,in,

[0026] Mission Pack represents the set of tasks assigned to AUVi, b i Internal tasks are arranged in the order in which they are added to the task package;

[0027] Execution path list Represents the execution path list of AUVi, p i The tasks within are arranged in the order of execution of AUVi;

[0028] List of successful bidders Represents the set of winning bidder numbers for all current tasks;

[0029] Winning bidders list Represents the bid set of all current task winners;

[0030] In the conflict resolution phase, the three vectors that need to reach consensus include the list of winners z i , the winning bidder's bid list y i And the newly added timestamp list s i ,s i Indicates the time when AUVi last updated data information during the communication process, as shown in the following formula:

[0031]

[0032]

[0033] In the formula, τ r is the reception time of the communication message, g ik is the element size of the i-th row and k-th column in the communication matrix G, which indicates the quality of communication between AUVi and AUVk. The larger the value, the more communication loss. Indicates that from g im Select the s pointed to by m corresponding to the minimum value mk .

[0034] As a preferred improvement of the present invention, in step S2, the communication topology network matrix in the improved algorithm specifically includes:

[0035] Step S21, calculating the speed of sound of underwater acoustic communication, and on this basis calculating the communication distance according to the local communication time;

[0036] Step S22, calculating the path loss of underwater acoustic communication according to the communication distance;

[0037] Step S23, using the communication distance and the path loss as decision factors to calculate the size of the communication network matrix elements.

[0038] As a preferred improvement of the present invention, in step S3, the corresponding solutions for tasks requiring multi-AUV cooperation and long-distance communication tasks are formulated, specifically including the following steps:

[0039] Step S31: for multiple AUV tasks, obtain the number of AUVs required for each task, L u ;

[0040] Step S32: L u >1 task copy L u -1 copy, together with the task itself, a total of L u The same task is also involved in subsequent decision-making allocation;

[0041] Step S33, for the long-distance communication task, obtain the task coordinates and AUV coordinates in the spatial rectangular coordinate system, and select the AUV with the shortest distance to execute;

[0042] Step S34, construct a virtual communication relay task corresponding to the long-distance communication task and participate in task allocation.

[0043] As a preferred improvement of the present invention, step S4 specifically includes the following steps:

[0044] Step S41, at the beginning of the t-th iteration, the communication network adjacency matrix is ​​updated, and the data structure obtained in the t-1 iteration is used as the starting data of this iteration;

[0045] Step S42, define AUVi according to the execution path p i The benefit of the job, if task j is assigned to p i If the task j is in the nth position, the marginal benefit of task j is calculated;

[0046] Step S43, define the effective task vector h ij Determine the validity of the task. If the task is valid, the auction is successful, and the qualified tasks and their corresponding placement locations are obtained;

[0047] Step S44, updating the corresponding data structure;

[0048] Step S45, after entering the conflict resolution phase, when AUVi receives the communication data of another AUVk, i.e., z k ,y k and k When these three vectors are compared, i and z k 、s i and k To determine which AUV carries the latest mission information, and perform relevant operations on the information receiver's task allocation data based on the comparison results;

[0049] Step S46, setting a communication quality threshold K according to the importance of the task. When the communication adjacency matrix elements corresponding to the two communicating AUVs exceed the threshold, no data information communication will be performed between the two AUVs.

[0050] Step S47: If after step S45, the list of winners of AUVi is i and the winning bidder's bid list y i If a change occurs, AUVi checks its own task package b i Is there an influence on z i and iIf there is a changed task, release the task and all tasks in the subsequent task package;

[0051] Step S48, returning to the task package construction phase, repeating steps S41 to S48 for the unassigned tasks until all tasks are assigned.

[0052] The beneficial effects of the present invention are as follows:

[0053] 1. The multi-AUV task allocation method proposed in the present invention, on the basis of adapting to weak underwater communication conditions, considers the acoustic loss to improve the adjacency matrix of the AUV communication topology network, and concretely describes the communication status and communication conditions between AUVs. By setting the communication quality threshold, some low-quality and high-loss communication processes are discarded, the quality of local communication of the system is improved, the communication load and cost are reduced, and at the same time, the dynamic allocation and conflict resolution of tasks are realized by regularly updating the communication network matrix;

[0054] 2. The present invention improves the profit function, parameterizes the start execution time of the task, and realizes an ideal or nearly ideal execution process by reducing the execution profit that deviates from the standard execution time under the premise of completing all tasks;

[0055] 3. The present invention realizes the coordination of multiple AUVs by duplicating tasks and participating in algorithm allocation, and at the same time constrains the start execution time of multiple AUV tasks, so that multiple AUVs can execute the task simultaneously or approximately simultaneously. For tasks requiring long-distance communication, by constructing a virtual communication relay task and participating in the allocation method, a suitable AUV is selected as the communication relay for executing the AUV, which reduces the potential risk of being out of the communication network when executing long-distance tasks and avoids unnecessary losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of an embodiment of the present invention;

[0057] Figure 2 An algorithm flow chart provided for an embodiment of the present invention;

[0058] Figure 3 A local communication network diagram for setting a communication quality threshold provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present invention will be described clearly and completely below in combination with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0061] In addition, in the present invention, descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0062] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0063] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0064] See also Figure 1-3 As shown, the present invention proposes a multi-AUV task allocation method based on a consensus bundling algorithm, comprising the following steps:

[0065] Step S1, obtain the task list and AUV list to be assigned, build the data structure required by the algorithm, and initialize various parameters;

[0066] In step S1, the task list J and AUV list I to be assigned are obtained. The goal of task assignment is to find a match between a task and an AUV without conflict so as to maximize the global reward. Specifically, the following steps are included:

[0067] Step S11: Assume that there are N t Tasks and N u AUVs, each AUV can be assigned a maximum of L t The number of AUVs required for each task is L. u, then the task list J = {1,2,…,N t}, AUV list I = {1,2,…,N u};

[0068] Step S12, constructing an optimization objective function, the objective function is the sum of the local benefit functions, expressed by the following formula:

[0069]

[0070] In the formula, x ij is a binary decision variable, c ij is the profit function;

[0071] Step S13, defining constraint conditions, including:

[0072] C1:

[0073] C2:

[0074] C3:

[0075] C4:

[0076] Where C1 represents the binary decision variable x ij C2 represents the maximum number of tasks that can be assigned to a single AUV, C3 represents that a single task cannot be assigned to multiple AUVs, and C4 represents the termination condition for task assignment, that is, for the decision variable x ij The sum in the direction of the two subscripts should be the number of tasks N t The smaller value of the sum of the number of missions that all AUVs can undertake.

[0077] The CBBA algorithm is divided into two stages: task package construction and conflict resolution. Therefore, the data structures required by the algorithm need to be constructed in these two stages, specifically:

[0078] (1) During the mission package construction phase, each AUV will carry four vectors, namely, mission package b i , execution path list p i 、List of successful bidders i And the winning bidder's bid list y i ,in,

[0079] Mission Pack represents the set of tasks assigned to AUVi, b i Internal tasks are arranged in the order in which they are added to the task package;

[0080] Execution path list Represents the execution path list of AUVi, pi The tasks within are arranged in the order of execution of AUVi;

[0081] List of successful bidders Represents the set of winning bidder numbers for all current tasks;

[0082] Winning bidders list represents the bid set of all current task winners. The price calculation method is the same as c in the objective function. ij related;

[0083] (2) In the conflict resolution phase, each AUV needs to share its own task data structure with each other through local communication, filter out invalid tasks and conflicting tasks, and finally achieve conflict-free task allocation. The screening of task validity will be further explained in step S4. The three vectors that need to reach consensus in this phase are the list of winners z i , the winning bidder's bid list y i And the newly added timestamp list s i ,s i Indicates the time when AUVi last updated data information during the communication process, as shown in the following formula:

[0084]

[0085]

[0086] In the formula, τ r is the reception time of the communication message, g ik is the element size of the i-th row and k-th column in the communication matrix G, which indicates the quality of communication between AUVi and AUVk. The larger the value, the more communication loss. Indicates that from g im Select the s pointed to by m corresponding to the minimum value mk .

[0087] Step S2, considering the communication distance and the path loss of underwater acoustic communication during the communication process, improving the communication topology network matrix in the algorithm; specifically including the following steps:

[0088] Step S21, the speed of sound in underwater acoustic communication is affected by the ocean environment parameters. Even a slight change will have a profound impact on communication data transmission. c The calculation of is as follows:

[0089] v c =1449.2+4.6T-0.055T 2 +(1.34+0.01T)(S-35)+0.016Z

[0090] In the formula, T is the seawater temperature, S is the seawater salinity, and Z is the seawater depth. Then, based on this, the communication distance d, that is, the distance between AUVs, is calculated according to the local communication time, as shown in the following formula:

[0091]

[0092] In the formula, t s is the time when a certain AUV sends information, t r is the information receiving time of the same AUV, and Δt is the time consumed for information processing.

[0093] Step S22, due to the existence of diffusion and absorption, sound waves inevitably produce loss in the transmission path, and the calculation of the underwater acoustic communication path loss is shown in the following formula;

[0094] w=10lgd n +f(α)

[0095] Wherein, w is the total path loss, n is the expansion coefficient, which is related to the shape of the expansion surface. When the expansion is spherical, n = 2, and when the expansion is cylindrical, n = 1.5. f(α) is the attenuation loss, which is related to the frequency f and the absorption coefficient α.

[0096] According to the Thorp empirical formula, the expression of the attenuation coefficient is as follows:

[0097]

[0098] Where f is the sound wave frequency and α is the absorption coefficient.

[0099] Step S23, let G(τ) be the symmetric adjacency matrix corresponding to the multi-AUV system communication network at time τ, and use the communication distance and path loss as decision factors for calculating the matrix elements. The calculation formula is as follows:

[0100]

[0101] In the formula, g ik is the element in the i-th row and k-th column of the communication adjacency matrix. It should be noted that when the two AUVs cannot communicate directly, that is, when the communication radius is exceeded, g ik Assign a value of ∞.

[0102] Step S3, formulate corresponding solutions for tasks requiring multi-AUV cooperation and long-distance communication tasks, specifically including the following steps:

[0103] Step S31: For a multi-AUV mission, first obtain the number of AUVs required for each mission, L u ;

[0104] Step S32: L u >1 task copy Lu -1 copy, together with the task itself, a total of L u The same task participates in the subsequent decision allocation at the same time. In order to ensure that the task and its replicas are operated at the same time, it is necessary to constrain the estimated operation time and minimize its variance, as shown in the following formula:

[0105]

[0106] In the formula, Indicates that AUVi follows its operation path p i Execute to the estimated start time of task j, represents the corresponding variance, Represents the AUV set that performs the multi-AUV mission in the current iteration;

[0107] Step S33, for the long-distance communication task, first obtain the task coordinates and AUV coordinates in the spatial rectangular coordinate system, as shown in the following formula, and select the AUV with the shortest distance to execute:

[0108]

[0109] In the formula, (X i ,Y i ,Z i ) is the coordinate of AUVi in the three-dimensional rectangular coordinate system, (X t ,Y t ,Z t ) is the coordinate of the long-distance communication task in the three-dimensional rectangular coordinate system, is the distance between the two;

[0110] Step S34, construct a virtual communication relay task corresponding to the long-distance communication task and participate in task allocation.

[0111] Step S4, processing all tasks based on the improved algorithm.

[0112] In step S4, the execution process of the improved CBBA algorithm is divided into the following sub-steps:

[0113] Step S41, at the beginning of the t-th iteration, update the communication network adjacency matrix, and use the data structure obtained in the t-1 iteration as the starting data for this iteration:

[0114] Step S42, define AUVi according to the execution path p i The income of the operation is As shown below:

[0115]

[0116] If task j is assigned to be placed in p i The marginal benefit of task j is calculated as follows:

[0117]

[0118] In the formula, Indicates that in p i The total revenue obtained by the AUV after placing task j at the nth position; j represents the time discount factor of task j, and its value can be determined according to the importance and complexity of the task in the actual operation. j ∈(0,1); represents the fixed income of executing task j; Indicates that AUVi follows its operation path p i The estimated start time of execution until task j; τ j represents the standard start execution time of task j; cost ij It represents the operation cost of AUVi to perform task j.

[0119] Step S43, define the effective task vector h ij Determine the effectiveness of the task, that is, by comparing the marginal benefit c corresponding to the current task ij and bid y ij The size of the task is used to determine whether it meets the allocation condition for maximum benefit:

[0120]

[0121] If the task is valid, the auction is successful, and the task J that meets the conditions is obtained. i And its corresponding placement

[0122] J i = argmax j (c ij ·h ij )

[0123]

[0124] In the formula, argmax(·) is a function that takes the set of points of the independent variable corresponding to the maximum value of the function in the brackets.

[0125] Step S44, update the corresponding data structure:

[0126]

[0127] In the formula, Indicated in J is inserted at the end of the vectori , Indicated in Vector J is placed in the position i .

[0128] Step S45, after entering the conflict resolution phase, when AUVi receives the communication data of another AUVk, i.e., z k ,y k and k When these three vectors are compared, z i and z k 、s i and k To determine which AUV carries the latest mission information, and perform relevant operations on the information receiver's task allocation data based on the comparison results;

[0129] There are three types of operations: update, let y ij =y kj , z ij =z kj ; Reset, let y ij =0, Leave, let y ij =y ij , z ij =z ij , the specific operation selection rules are shown in Table 1 below:

[0130] Table 1. Processing rules after AUVi communicates with AUVk for task j

[0131]

[0132] Step S46, setting a communication quality threshold K according to the importance of the task, where g ik ≤K, it can be seen that the more important the task is, the higher K is, which means the larger the communication range is, that is, more other AUVs will participate in the assigned communication of the task. When the communication adjacency matrix elements g corresponding to the two communicating AUVs are ik When the threshold is exceeded, no data information communication will be carried out between the two to reduce communication costs;

[0133] In order to ensure that conflicts between AUVs can be fully resolved under limited communication conditions, the working time domain is divided and the communication topology network and its matrix are updated regularly. As the AUV work progresses, AUVs with no conflicts will gradually approach, and the communication adjacency matrix will be updated as the communication distance decreases, realizing dynamic conflict resolution.

[0134] Step S47: If after step S45, the list of winners of AUVi is i and the winning bidder's bid list yi If a change occurs, AUVi will check its own task package b i Is there an influence on z i and i If there is a changed task, it is necessary to release the task and all tasks in the subsequent task package; the specific release operation is shown in the following formula:

[0135]

[0136]

[0137]

[0138] Where b in Represents task package b i The nth element in It represents task package b i The first location in which the task needs to be released.

[0139] Step S48, returning to the task package construction phase, repeating steps S41 to S48 for the unassigned tasks until all tasks are assigned.

[0140] The beneficial effects of the present invention are as follows:

[0141] 1. The multi-AUV task allocation method proposed in the present invention, on the basis of adapting to weak underwater communication conditions, considers the acoustic loss to improve the adjacency matrix of the AUV communication topology network, and concretely describes the communication status and communication conditions between AUVs. By setting the communication quality threshold, some low-quality and high-loss communication processes are discarded, the quality of local communication of the system is improved, the communication load and cost are reduced, and at the same time, the dynamic allocation and conflict resolution of tasks are realized by regularly updating the communication network matrix;

[0142] 2. The present invention improves the profit function, parameterizes the start execution time of the task, and realizes an ideal or nearly ideal execution process by reducing the execution profit that deviates from the standard execution time under the premise of completing all tasks;

[0143] 3. The present invention realizes the coordination of multiple AUVs by duplicating tasks and participating in algorithm allocation, and at the same time constrains the start execution time of multiple AUV tasks, so that multiple AUVs can execute the task simultaneously or approximately simultaneously. For tasks requiring long-distance communication, by constructing a virtual communication relay task and participating in the allocation method, a suitable AUV is selected as the communication relay for executing the AUV, which reduces the potential risk of being out of the communication network when executing long-distance tasks and avoids unnecessary losses.

[0144] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A multi-AUV task allocation method based on consensus bundling algorithm, characterized in that: The following steps are involved: Step S1, obtain the task list and AUV list to be assigned, build the data structure required by the algorithm, and initialize various parameters; Step S2, considering the communication distance in the communication process and the path loss of underwater acoustic communication, improving the communication topology network matrix in the algorithm, the communication topology network matrix in the improved algorithm specifically includes: Step S21, calculating the speed of sound of underwater acoustic communication, and on this basis calculating the communication distance according to the local communication time; Step S22, calculating the path loss of underwater acoustic communication according to the communication distance; Step S23, using the communication distance and path loss as decision factors to calculate the size of the communication network matrix element; Step S3, formulate corresponding solutions for tasks requiring multi-AUV cooperation and long-distance communication tasks, specifically including the following steps: Step S31: for multiple AUV tasks, obtain the number of AUVs required for each task, L u ; Step S32: L u >1 task copy L u -1 copy, together with the task itself, a total of L u The same task is also involved in subsequent decision-making allocation; Step S33, for the long-distance communication task, obtain the task coordinates and AUV coordinates in the spatial rectangular coordinate system, and select the AUV with the shortest distance to execute; Step S34, constructing a virtual communication relay task corresponding to the long-distance communication task and participating in task allocation; Step S4, based on the improved algorithm, processes all tasks, specifically including the following steps: Step S41, at the beginning of the t-th iteration, the communication network adjacency matrix is ​​updated, and the data structure obtained in the t-1 iteration is used as the starting data of this iteration; Step S42, define AUVi according to the execution path p i The benefit of the job, if task j is assigned to p i If the task j is in the nth position, the marginal benefit of task j is calculated; Step S43, define the effective task vector h ij Determine the validity of the task, that is, by comparing the marginal benefit corresponding to the current task and the size of the bid of the winning bidder to determine whether the task meets the allocation conditions for maximum benefit. If the task is valid, the auction is successful, and the qualified task and its corresponding placement are obtained; Step S44, updating the corresponding data structure; Step S45, after entering the conflict resolution phase, when AUVi receives the communication data of another AUVk, i.e., z k ,y k and k When these three vectors are compared, i and z k 、s i and k To determine which AUV carries the latest mission information, and perform relevant operations on the information receiver's task allocation data based on the comparison results; Step S46, setting a communication quality threshold K according to the importance of the task. When the communication adjacency matrix elements corresponding to the two communicating AUVs exceed the threshold, no data information communication will be performed between the two AUVs. Step S47: If after step S45, the list of winners of AUVi is i and the winning bidder's bid list y i If a change occurs, AUVi checks its own task package b i Is there an influence on z i and i If there is a changed task, release the task and all tasks in the subsequent task package; Step S48, returning to the task package construction phase, repeating steps S41 to S48 for the unassigned tasks until all tasks are assigned.

2. The method according to claim 1, characterized in that In step S1, the task list J and AUV list I to be assigned are obtained, which specifically includes: Step S11, define N t Tasks and N u AUVs, each AUV can be assigned a maximum of L t Each task requires L number of AUVs to perform u , then the task list J = {1,2,…,N t }, AUV list I = {1,2,…,N u }; Step S12, constructing an optimization objective function, which is expressed by the following formula: Among them, x ij is a binary decision variable, c ij is the profit function; Step S13, defining constraint conditions, including: Where C1 represents the binary decision variable x ij C2 represents the maximum number of tasks that can be assigned to a single AUV, C3 represents that a single task cannot be assigned to multiple AUVs, and C4 represents the termination condition for task assignment, that is, for the decision variable x ij The sum in the direction of the two subscripts should be the number of tasks N t The smaller value of the sum of the number of missions that all AUVs can undertake.

3. The method according to claim 1, characterized in that In step S1, the data structure required for constructing the algorithm includes the data structure required for constructing the algorithm in the task package construction phase and the conflict resolution phase, specifically: During the mission package construction phase, each AUV will carry four vectors, namely, mission package b i , execution path list p i 、List of successful bidders i And the winning bidder's bid list y i ,in, Mission Pack represents the set of tasks assigned to AUVi, b i Internal tasks are arranged in the order in which they are added to the task package; Execution path list Represents the execution path list of AUVi, p i The tasks within are arranged in the order of execution of AUVi; List of successful bidders Represents the set of winning bidder numbers for all current tasks; Winning bidders list Represents the bid set of all current task winners; In the conflict resolution phase, the three vectors that need to reach consensus include the list of winners z i , the winning bidder's bid list y i And the newly added timestamp list s i ,s i Indicates the time when AUVi last updated data information during the communication process, as shown in the following formula: In the formula, τ r is the reception time of the communication message, g ik is the element size of the i-th row and k-th column in the communication matrix G, which indicates the quality of communication between AUVs. The larger the value, the more communication loss. Indicates that from g im Select the s pointed to by m corresponding to the minimum value mk .

Images