Cooperative task planning method based on improved parrot optimization algorithm

Abstract

The invention particularly relates to a cooperative task planning method based on an improved parrot optimization algorithm, and the method comprises the steps: obtaining the initial positions and movement speeds of all unmanned underwater vehicles and targets, determining the parameters and target functions of three layers of task planning, namely external planning, intermediate planning and internal planning, the task completion time of the unmanned underwater vehicle is planned by taking the unmanned underwater vehicle which finally departs as a reference; determining a target position and unmanned underwater vehicle task completion time according to task completion time planning, determining parameters and target functions of two-layer task planning of collaborative external planning and collaborative internal planning, performing collaborative path search planning of the three unmanned underwater vehicles, and realizing a task of collaborative discovery of a target by the three unmanned underwater vehicles; and adopting an improved parrot optimization algorithm, and carrying out setting optimization on the collaborative task planning parameters by solving the minimum value of the target function of each layer of task planning to obtain optimal task planning parameters. According to the method, collaborative task planning can be quickly realized.

Description

Technical Field

[0001] The present invention relates to the field of unmanned underwater vehicle mission planning, and in particular to a collaborative mission planning method based on an improved Parrot optimization algorithm. Background Art

[0002] Mission planning for unmanned underwater vehicles is influenced by multiple factors, including the environment, mission objectives, communications, and energy management. Traditional algorithms struggle to plan missions for moving targets. Therefore, an intelligent clustering algorithm with global optimization capabilities is employed to design optimal parameters.

[0003] In 2024, Junbo Lian et al., inspired by the relationship between parrots and their owners, proposed the Parrot Optimizer (PO). This algorithm simulates parrots' foraging, resting, fear of strangers, and communication behaviors to update candidate solutions. The PO can be effectively applied to mission planning problems, particularly those in complex terrain.

[0004] Mission planning for an unmanned underwater vehicle (UUV) requires obtaining relevant underwater terrain data, including elevation and slope information. This data serves as the foundation for mission planning. By analyzing and processing this data, a mission plan suitable for the UUV can be developed. The Parrot optimization algorithm utilizes this terrain data to find the optimal navigation path.

[0005] In actual mission planning, many factors must be considered, such as terrain steepness, irregularities, and the impact of obstacles. These factors can affect the UUV's navigation path, and therefore require thorough consideration and analysis during mission planning. The Parrot Optimization Algorithm, as an intelligent optimization algorithm, can identify the optimal mission planning solution based on these considerations.

[0006] However, the standard Parrot optimization algorithm still has some defects: although the Parrot optimization algorithm has the characteristics of strong optimization ability and fast convergence speed, it also has the disadvantages of imbalance between global exploration and local development capabilities, easy to produce local optimality, and weak global exploration ability. Therefore, in response to the above problems, some scholars have begun to conduct improvement research on the algorithm; for example, by introducing an adaptive mechanism in the population foraging stage, the probability of the algorithm falling into the local optimality is reduced, and the accuracy of the later optimization is improved. The above studies have improved the performance of the Parrot optimization algorithm to a certain extent, but the problems of decreased convergence performance in the later stage, low accuracy of global optimization results, and easy to fall into local optimality still exist. In order to obtain the best results for mission planning, it is necessary to significantly improve the Parrot optimization algorithm to obtain the optimal navigation parameter design scheme.

[0007] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0008] The present invention provides a collaborative task planning method based on an improved Parrot optimization algorithm, which is used to solve the problems of existing parameter design methods in task planning optimization, such as imbalance between global exploration and local development capabilities, easy generation of local optimality, and weak global exploration capabilities, so as to achieve faster collaborative task planning.

[0009] Other features and advantages of the present invention will become apparent from the following detailed description, or may be learned in part by practice of the present invention.

[0010] According to a first aspect of the present invention, there is provided a collaborative task planning method based on an improved Parrot optimization algorithm, the method comprising:

[0011] Obtain the initial positions and velocities of all UUVs and targets, determine the parameters and objective functions for the three-level mission planning: external planning, intermediate planning, and internal planning, and plan the time it takes for the UUV to complete its mission based on the last UUV to depart.

[0012] The target location and the unmanned underwater vehicle mission completion time are determined based on the mission completion time plan. Then, the parameters and objective functions of the collaborative external planning and collaborative internal planning two-level mission planning are determined, and the collaborative path search planning of the three unmanned underwater vehicles is carried out to achieve the task of collaborative target discovery by the three unmanned underwater vehicles.

[0013] The improved Parrot optimization algorithm is used to optimize the collaborative task planning parameters by solving the minimum value of the objective function of each layer of task planning, and the optimal task planning parameters are obtained.

[0014] The determination of the improved Parrot optimization algorithm includes:

[0015] The improved Circle chaos map is used to initialize the parrot population;

[0016] In the foraging phase of the Parrot Optimization Algorithm, a nonlinear inertia weight ω is introduced to adjust the correlation between the position update and the current position;

[0017] The tangent function is used throughout the entire process during the stop phase of the Parrot optimization algorithm;

[0018] Added sparrow alert strategy to the communication phase of the parrot optimization algorithm.

[0019] In some exemplary embodiments, determining the parameters and objective functions of the three-level task planning of external planning, intermediate planning, and internal planning specifically includes:

[0020] The external planning is an arrival time design loop. The external planning performs encounter time planning. The goal is to assign a reasonable task completion time. At this time point, the target moves to a certain position, and the unmanned underwater vehicle simultaneously reaches the detection circle corresponding to the target position, finds the target and switches to attack.

[0021] The intermediate planning is the shortest time design loop. The intermediate planning is to plan the fastest navigation time to the target detection circle according to a certain expected encounter time and the corresponding predicted target position. The goal is to find the position on the target detection circle that the unmanned underwater vehicle can reach the fastest, that is, the relative angle.

[0022] The internal planning is a path planning and design loop. The internal planning performs arrival path planning based on a specific location on a given target detection circle. The goal is to find the fastest path to a certain location on the target detection circle and the actual navigation time.

[0023] The intermediate planner searches for the position on the target detection circle based on the actual navigation time fed back by the internal planner, finds the fastest time to reach the target detection circle, and then feeds this time back to the external planner; the external planner calculates the difference between the shortest navigation time and the expected encounter time based on the shortest navigation time fed back by the intermediate planner, and adjusts the expected encounter time according to the difference until the difference meets a certain threshold requirement or the planning termination condition is met. If the threshold requirement is met, it is considered that the shortest encounter time, the position on the target detection circle and the implementation path have been found.

[0024] In some exemplary embodiments, the unmanned underwater vehicle mission completion time planning specifically includes:

[0025] Establish the objective function model of external planning - time difference model;

[0026] The purpose of mission completion time planning is to make the UUV reach the target detection circle within the specified mission time. That is, the difference between the time it takes for the UUV to move along the path to the target detection circle and the time it takes for the target to move should be as small as possible, and the encounter time should be as small as possible. Therefore, the objective function of external planning is:

[0027] J=give_time+ωΔt, Δt=|t fxq -give_time|

[0028] Among them, t fxq is the actual navigation time of the UUV to reach the target detection circle; give_time is the given estimated mission completion time, which is also the target movement time; Δt is the difference between the actual navigation time of the UUV and the estimated encounter time between the two; ω is the optimization weight;

[0029] The design variable for external planning is: give_time

[0030] The objectives of external planning are to:

[0031] give_time minJ(give_time)

[0032] The inputs passed to the intermediate planner for each optimization are: the expected task completion time give_time and the predicted target position TarPos(give_time) after give_time;

[0033] After each intermediate plan is completed, the input returned to the external planner is: the actual shortest navigation time t that the unmanned underwater vehicle can achieve corresponding to the target position detection circle under the expected task completion time fxq (give_time) and optimal path X best (t fxq (give_time));

[0034] The final output of external planning is: the optimal task completion time best_time, the actual shortest navigation time t fxq (best_time), optimal path X best (t fxq (best_time)).

[0035] In some exemplary embodiments, the unmanned underwater vehicle mission completion time planning specifically includes:

[0036] Establish the objective function model of intermediate planning - the actual navigation time model of the unmanned underwater vehicle;

[0037] The goal of the intermediate planning problem is to minimize the navigation time of the unmanned underwater vehicle on the detection circle to reach a given target position, so the objective function of the intermediate planning is:

[0038] J=t fxq

[0039] The design variables for the intermediate planning are: entry angle The goal of intermediate planning is

[0040] The input passed to the internal planner for each planning is: the target position TarPos(give_time) predicted at the current expected task completion time give_time at the entry angle The expected path end point at time:

[0041]

[0042] Where, Dis detectis the detection range of the unmanned underwater vehicle, that is, the radius of the target detection circle;

[0043] After each intermediate plan is completed, the input returned to the external planner is: the shortest navigation time to reach the desired path end point and the optimal path under the corresponding entry angle

[0044] In some exemplary embodiments, the unmanned underwater vehicle mission completion time planning specifically includes:

[0045] Establish the objective function model of internal planning - internal constraint model;

[0046] Internal planning is a path planning problem, that is, finding a path from the initial point to the given task end point The optimal path And the corresponding actual exercise time

[0047] The design variables of internal planning are: navigation path X i The goal of internal planning is minJ=minF(X i );

[0048] After each internal planning is completed, the intermediate planner is returned with the actual navigation time to reach the desired path end point. and navigation paths

[0049] In some exemplary embodiments, the UUV collaborative path search planning specifically includes:

[0050] The purpose of the collaborative external planning of the collaborative path search planning is to make the unmanned underwater vehicle reach the specified mission time best_time i (i=1,2) to reach the target detection circle, that is, the time required for the unmanned underwater vehicle to move along the path to the target detection circle and The smaller the time difference, the better, that is, the detection circle is reached at the same time, so the objective function of the collaborative external planning of the collaborative path search planning is:

[0051]

[0052] Where, is the actual navigation time of the unmanned underwater vehicle i (i=1,2) to reach the target detection circle;

[0053] The design variables of collaborative external planning are: the entry angle of the unmanned underwater vehicle i (i = 1, 2) on the target detection circle

[0054] The objectives of collaborative external planning are to:

[0055] minJ i (i=1,2)

[0056] The input passed to the collaborative internal planner for each optimization is: the current best_time i (i=1,2) The predicted target position TarPos(best_time i )(i=1,2) at the entry angle The expected path end point at time:

[0057]

[0058] After each collaborative internal planning is completed, the input returned to the collaborative external planner is: best_time i (i=1,2) corresponding to the target position detection circle entry angle The shortest practical flight time that an unmanned underwater vehicle can achieve and the optimal path

[0059] The final output of the collaborative external planning is: the optimal mission completion time of the unmanned underwater vehicle i (i = 1, 2) and the optimal path

[0060] The collaborative internal planning model of collaborative path search planning is consistent with the internal planning model described in three-level task planning.

[0061] In some exemplary embodiments, the initializing the parrot population using the improved Circle chaos map specifically includes:

[0062] The Circle chaos mapping formula is improved to make its chaos value distribution more uniform.

[0063] The improved Circle chaos mapping expression is:

[0064]

[0065] Based on the above formula, a random matrix A with N rows and M columns is generated, where N is the population size and M is the spatial dimension of the optimization variable; then the value of each element is calculated iteratively:

[0066] W i,j =W min +A i ×(W max -W min )

[0067] Among them, W i,jis the value of the jth variable specified by the i-th parrot individual, W max 、W min are the upper and lower bounds of the problem to be solved.

[0068] In some exemplary embodiments, the nonlinear inertia weight ω is introduced during the foraging phase of the parrot optimization algorithm to adjust the correlation between the position update and the current position, as shown below:

[0069]

[0070] Among them, W i t Indicates the current position of the parrot, W i t+1 Indicates the location of the successful update, represents the average position within the current population, Levy(M) is used to describe the parrot's navigation, W best Indicates the best position from initialization to the current search.

[0071] In some exemplary embodiments, the introducing of the tangent function during the dwell phase specifically includes:

[0072] Use the tangent function to get the new stop position, which is in the interval [0, π]. After determining the new stop position, continue sailing:

[0073] W i t+1 =W i t +W best Levy(M)+tan(θ) ones(1,M)

[0074] Among them, ones(1,M) is a vector of all 1s of dimension M.

[0075] In some exemplary embodiments, adding a sparrow alert strategy during the communication phase specifically includes:

[0076] Use the sparrow alert strategy to replace the original parrot communication behavior. The formula is as follows:

[0077]

[0078] During the execution of the algorithm, W best and W worst To represent the position of the best solution and the worst solution currently searched, β is a random number drawn from the standard normal distribution. It is a key parameter for step size control, ranging from -1 to 1, and is used to adjust the dynamics of the search. In the second stage, after the position of the i-th parrot is updated, its fitness value is determined by F iThe global best and worst fitness values ​​are represented by F best and F worst Definition, where ε is a constant that prevents the denominator from being zero and ensures the stability of numerical calculations.

[0079] The collaborative task planning method based on the improved Parrot optimization algorithm provided by the embodiments of the present invention has the following beneficial effects compared with the prior art:

[0080] Because existing parameter design methods for task planning optimization suffer from imbalanced global exploration and local development capabilities, prone to local optima, and weak global exploration capabilities, the present invention provides a collaborative task planning method for unmanned underwater vehicles based on an improved Parrot optimization algorithm. Unlike traditional static target planning, this method utilizes multi-level planning to achieve task planning for three unmanned underwater vehicles surrounding a dynamic target. Based on the standard Parrot optimization algorithm, this method employs an improved Circle chaos map to initialize the Parrot population; improves the Parrot optimization algorithm's foraging phase using a nonlinear inertia weight factor; improves the Parrot optimization algorithm's dwell phase using a tangent function; and improves the Parrot optimization algorithm's communication phase using a sparrow alert strategy. This collaborative task planning method based on the improved Parrot optimization algorithm exhibits faster convergence, higher parameter search efficiency, and stronger global optimization capabilities.

[0081] Specifically, to improve the performance of the basic Parrot search algorithm, particularly in terms of search speed and optimization capabilities, innovative improvements have been made to its initial population generation strategy. By introducing an improved Circle mapping, the new method ensures a more even distribution of the initial population within the search space, effectively enhancing the algorithm's ability to explore the solution space. This improvement not only increases the diversity of the population but also helps prevent the algorithm from prematurely converging to a local optimum, laying a solid foundation for finding the global optimal solution.

[0082] Specifically, coordinating the local optimization ability and global search ability of the metaheuristic algorithm is a key factor affecting the algorithm's optimization accuracy and optimization speed. In the parrot optimization algorithm, the update of the individual position is closely related to the current position, so the nonlinear inertia weight ω is introduced to regulate the correlation between the position update and the current position. In the early stages of the iteration, a smaller ω value means that the individual update is less affected by the current position, which helps the algorithm explore in a wider solution space, thereby enhancing its global search ability. As the iteration deepens, the ω value gradually increases, making the individual update more dependent on the current position, which helps to narrow the search range and concentrate on searching for the optimal solution, thereby not only enhancing the accuracy of the local search, but also accelerating the convergence rate of the algorithm. In addition, in order to strike a balance between local and global search, the leadership strategy of the Salp Swarm Algorithm was improved and integrated into the parrot optimization algorithm;

[0083] Specifically, during the dwell phase of the Parrot optimization algorithm, the tangent function can be efficiently used to penetrate the global solution and escape from the local optimal solution;

[0084] Specifically, in the communication stage of the parrot algorithm, by incorporating the sparrow's vigilance behavior, not only the algorithm's search efficiency is improved, but also its ability to explore the solution space is enhanced.

[0085] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] The accompanying drawings are incorporated into and constitute a part of this specification, illustrate embodiments consistent with the present invention, and together with the description, serve to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and it is clear that those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0087] Figure 1 Flowchart of an optimization design method for an improved parrot optimization algorithm provided in an embodiment of the present invention;

[0088] Figure 2 A flowchart of a collaborative task planning method based on an improved Parrot optimization algorithm provided in an embodiment of the present invention is provided;

[0089] Figure 3 This is a simulation process diagram of the collaborative task planning method based on the improved Parrot optimization algorithm provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0090] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0091] In view of the shortcomings and deficiencies of the existing technology, this example embodiment provides a collaborative task planning method based on an improved parrot optimization algorithm. Figure 2 As shown, the following steps may be specifically included:

[0092] Step S1, obtain the initial position, movement speed and other information of all unmanned underwater vehicles and targets, determine the parameters and objective functions of the three-level task planning of external planning, intermediate planning and internal planning. Since all unmanned underwater vehicles have different departure times, in order to achieve the task of three unmanned underwater vehicles collaboratively discovering the target, the unmanned underwater vehicle task completion time planning is based on the last unmanned underwater vehicle to depart.

[0093] For UUVs, regardless of whether they are collaborative or not, the path planning algorithm that achieves the shortest possible time to reach the target detection circle is the underlying algorithm for collaborative mission planning. This algorithm consists of an external planning loop, which is the arrival time design loop; an intermediate planning loop, which is the shortest time design loop; and an internal planning loop, which is the path planning design loop.

[0094] External planning is used to plan the encounter time. The goal is to assign a reasonable task completion time. At this time point, the target moves to a certain position, and the unmanned underwater vehicle simultaneously reaches the detection circle corresponding to the target position, finds the target and switches to attack.

[0095] The intermediate planning is to plan the fastest navigation time to the target detection circle based on a certain expected encounter time and the corresponding predicted target position. The goal is to find the position on the target detection circle where the unmanned underwater vehicle can reach the fastest, that is, the relative angle.

[0096] Internal planning is based on a specific location on a given target detection circle and performs path planning. The goal is to find the fastest path to a location on the target detection circle and the actual flight time.

[0097] The intermediate plan searches for the position on the target detection circle based on the actual navigation time fed back by the internal plan, finds the fastest time to reach the target detection circle, and then feeds this time back to the external planner.

[0098] The external planning calculates the difference between the shortest navigation time feedback from the intermediate planning and the expected encounter time, and adjusts the expected encounter time according to the difference until the difference meets a certain threshold requirement or the planning termination condition is met. If the threshold requirement is met, it is considered that the shortest encounter time, the position on the target detection circle and the implementation path have been found.

[0099] Step S1 specifically includes the following steps S11 to S14:

[0100] Step S11: extract map model information, set obstacles, and give the initial positions, movement speeds, and other information of all unmanned underwater vehicles and targets.

[0101] Extracting map model information and setting obstacles are as follows:

[0102] By extracting complete information from the TIF format map, the map is then 3D modeled based on the extracted information. Any obstacles can be set on the constructed map.

[0103] Specifically, the target initial position is set to start target =[44785,37285,6428]m, the maximum speed is assumed to be v target =70kn, the movement direction is set to direction target =-pi*17 / 22.

[0104] Specifically, this embodiment involves three unmanned underwater vehicles, and collaborative mission planning is performed for the three unmanned underwater vehicles, that is, the number of unmanned underwater vehicles n=3, the first unmanned underwater vehicles to depart are numbered 1 and 2, and the last to depart is 3.

[0105] The initial position of the unmanned underwater vehicle 1 is set to start UUV1 =[2571,2571,6428]m, speed is set to v UUV1 =80kn;

[0106] The initial position of the UUV 2 is set to start UUV2 =[3000,3000,6428]m, speed is set to v UUV2 =80kn;

[0107] The initial position of the unmanned underwater vehicle 3 is set to start UUV3 =[2142,2142,6428]m, speed is set to v UUV2 =80kn;

[0108] The up and down tilt angle range of the unmanned underwater vehicle is limited to Inside;

[0109] The azimuth range of the UUV is limited to Inside;

[0110] The maximum search radius is

[0111] The minimum search radius is

[0112] The number of search path nodes, that is, the number of waypoints that the unmanned underwater vehicle needs to navigate, is set to num=43.

[0113] Step S12: establishing an objective function model of the external planning of the task completion time planning, namely, a time difference model.

[0114] The purpose of mission completion time planning is to make the UUV reach the target detection circle within the specified mission time. That is, the difference between the time it takes for the UUV to move along the path to the target detection circle and the time it takes for the target to move should be as small as possible, and the encounter time should be as small as possible. Therefore, the objective function of external planning is:

[0115] J=give_time+ωΔt, Δt=|t fxq -give_time|

[0116] Where:

[0117] t fxq The actual navigation time of the unmanned underwater vehicle to reach the target detection circle;

[0118] give_time is the estimated task completion time given, which is also the target movement time;

[0119] Δt is the difference between the actual navigation time of the UUV and the estimated encounter time between the two;

[0120] ω is the optimization weight, which can be selected from 5 to 10 to avoid the phenomenon of large Δt deviation and small give_time.

[0121] The design variable for external planning is: give_time.

[0122] The objectives of external planning are to:

[0123] minJ(give_time)

[0124] The inputs passed to the intermediate planner for each optimization are: the expected task completion time give_time and the target position TarPos(give_time) predicted after give_time.

[0125] After each intermediate plan is completed, the input returned to the external planner is: the actual shortest navigation time t that the unmanned underwater vehicle can achieve corresponding to the target position detection circle under the expected task completion time fxq (give_time) and optimal path X best (t fxq (give_time)).

[0126] The final output of external planning is: the optimal task completion time best_time, the actual shortest navigation time t fxq (best_time), optimal path X best (t fxq (best_time)).

[0127] Step S13: establishing an objective function model of the intermediate planning of the task completion time planning, that is, an actual navigation time model of the unmanned underwater vehicle.

[0128] Since the target detection circle is a circle with the UAV detection range as its radius and the target position as its center, the UAV can reach any position on the target detection circle, that is, it can reach the target detection circle from different entry angles. Therefore, the goal of the intermediate planning problem is to minimize the time it takes for the UAV to reach the given target position on the detection circle. Therefore, the objective function of the intermediate planning is:

[0129] J=t fxq

[0130] The design variables for the intermediate planning are: entry angle The goal of intermediate planning is

[0131] The input passed to the internal planner for each planning is: the target position TarPos(give_time) predicted at the current expected task completion time give_time at the entry angle The expected path end point at time:

[0132]

[0133] Where, Dis detect is the detection range of the unmanned underwater vehicle, that is, the radius of the target detection circle.

[0134] After each intermediate plan is completed, the input returned to the external planner is: the shortest navigation time to reach the desired path end point and the optimal path under the corresponding entry angle

[0135] Step S14, establishing an internal planning objective function model for task completion time planning, namely, an internal constraint model;

[0136] Internal planning is a path planning problem, that is, finding a path from the initial point to the given task end point The optimal path And the corresponding actual exercise time

[0137] In order to make the UUV operate efficiently, the planned path needs to be optimal under certain criteria, that is, the path length should be shortened as much as possible. i It is represented as a list of n waypoints that the unmanned underwater vehicle needs to navigate. Each waypoint corresponds to a path node in the search map with coordinates P ij =(x ij ,y ij ,zij ). By expressing the Euclidean distance between three nodes as The path length cost F1 can be calculated as:

[0138]

[0139] In addition to optimality, the planned path also needs to ensure the safe operation of the UUV by guiding the UUV through threats usually caused by obstacles present in the operation space. Let K be the set of all threats, i.e. obstacles, and assume that each threat is defined in a cylinder whose projection has the center coordinate C k and radius R k For a given path segment The threat cost associated with its impact on C k The distance d k By considering the diameter D of the UUV and the hazard distance S to the collision zone, the waypoint P of the obstacle set K ij The threat cost F2 is calculated as follows:

[0140]

[0141] While the diameter D is determined by the size of the UUV, the distance S depends on several factors such as the actual application, operating environment, and positioning accuracy.

[0142] The operation of unmanned underwater vehicles usually needs to consider safety and avoid collision with ground obstacles, so it is necessary to set the minimum navigation depth h of the unmanned underwater vehicle. min and maximum navigation depth h max . Minimum navigation depth h min Determined by the terrain or obstacles, while also considering the navigation stability and obstacle avoidance capabilities of the unmanned underwater vehicle; the maximum navigation depth h max Limited by the maximum navigation depth of the unmanned underwater vehicle and relevant regulations and standards. ij The depth information can be used to calculate the navigation cost related to the navigation depth:

[0143]

[0144] Among them, h ij Indicates the navigation depth relative to the ground. ij To maintain the average depth and penalize out-of-range values. ij Summing this gives the depth cost:

[0145]

[0146] The smoothing cost evaluates the turning and climbing rate, which is crucial for generating a feasible path. ij are three consecutive path segments and Two vectors projected on the horizontal plane Oxy and The angle between is a unit vector in the z-axis direction, and the projection vector can be calculated as:

[0147]

[0148] Therefore, the turning angle is calculated as follows:

[0149]

[0150] Climb angle ψ ij Is a path segment and its projection onto the horizontal plane The angle between . It is given by the following formula:

[0151]

[0152] The smoothing cost is calculated as follows:

[0153]

[0154] Among them, a1 and a2 are the penalty coefficients of turning angle and climbing angle respectively.

[0155] By considering the path X i The relevant optimality, safety, and feasibility constraints allow the overall cost function to be defined as follows:

[0156]

[0157] Among them, b k is the weight coefficient, F1(X i ) to F4(X i ) are the costs associated with path length, safety and feasibility, smoothness, and navigation depth. The decision variable is X i , including a list P of n waypoints ij =(x ij ,y ij ,z ij ), so that P ij ∈O, where O is the operation space of the UUV. Given these definitions, the cost function F is fully determined and can be used as input to the path planning process.

[0158] The design variables of internal planning are: navigation path X iThe goal of internal planning is minJ=minF(X i ).

[0159] After each internal planning is completed, the intermediate planner is returned with the actual navigation time to reach the desired path end point. and navigation paths

[0160] Step S2: The target location and the UUV task completion time can be determined based on the task completion time planning. Then, the parameters and objective functions of the collaborative external planning and collaborative internal planning two-layer task planning are determined, and the three UUVs perform collaborative path search planning to achieve the task of collaborative target discovery by the three UUVs.

[0161] Collaborative path search planning mainly focuses on designing near the mission entry angle of the last-departing unmanned underwater vehicle, which is equivalent to constraining the target entry angle design range when planning the mission of the first-departing unmanned underwater vehicle.

[0162] To solve the coordination problem of three unmanned underwater vehicles, the first unmanned underwater vehicle that departs first directly uses collaborative path search planning to design the path when launching.

[0163] When the last unmanned underwater vehicle is launched, the mission completion time planning algorithm is first used to calculate the shortest encounter time between the last unmanned underwater vehicle and the target detection circle and the actual shortest navigation time of the last unmanned underwater vehicle.

[0164] Next, the predicted position of the target is determined based on the encounter time, and the optimization goal is to minimize the difference between the time when the first unmanned underwater vehicle reaches the detection circle and the actual shortest navigation time of the last unmanned underwater vehicle. The collaborative path search planning is performed on the unmanned underwater vehicle that departs first. That is, the collaborative path search planning takes the shortest encounter time between the last-departing unmanned underwater vehicle and the target detection circle and the actual shortest navigation time of the last-departing unmanned underwater vehicle as input, and takes minimizing the collaborative time difference as the goal, and optimizes the angle at which the first-departing unmanned underwater vehicle reaches the target detection circle as the design variable.

[0165] To simplify the problem, the last unmanned underwater vehicle to depart is numbered 3, and the first unmanned underwater vehicle to depart is numbered i (i = 1, 2).

[0166] Taking the last departing unmanned underwater vehicle as the benchmark, the mission completion time planning model is used to plan the mission of the last departing unmanned underwater vehicle, and the mission completion time best_time3 of the last departing unmanned underwater vehicle and the actual shortest navigation time are obtained. Then, based on the task completion time, the expected completion time best_time for the first unmanned underwater vehicle to plan its task can be determined. i = best_time3 (i = 1, 2) and the center position of the target detection circle TarPos (best_time i )(i=1,2).

[0167] Step S2 specifically includes the following steps S21-S22;

[0168] Step S21 , establishing an objective function model of collaborative external planning of collaborative path search planning, namely, a time difference model.

[0169] The purpose of the collaborative external planning of the collaborative path search planning is to make the unmanned underwater vehicle reach the specified mission time best_time i (i=1,2) to reach the target detection circle, that is, the time required for the unmanned underwater vehicle to move along the path to the target detection circle and The smaller the time difference, the better, that is, the detection circle is reached at the same time, so the objective function of the collaborative external planning of the collaborative path search planning is:

[0170]

[0171] Where, is the actual navigation time of the unmanned underwater vehicle i (i=1,2) to reach the target detection circle;

[0172] The design variables of collaborative external planning are: the entry angle of the unmanned underwater vehicle i (i = 1, 2) on the target detection circle

[0173] The objectives of collaborative external planning are to:

[0174] minJ i (i=1,2)

[0175] The input passed to the internal planner for each optimization is: the current best_time i (i=1,2) The predicted target position TarPos(best_time i )(i=1,2) at the entry angle The expected path end point at time:

[0176]

[0177] After each internal planning is completed, the input returned to the collaborative external planner is: best_time i (i=1,2) corresponding to the target position detection circle entry angle The shortest practical flight time that an unmanned underwater vehicle can achieve and the optimal path

[0178] The final output of the collaborative external planning is: the optimal mission completion time of the unmanned underwater vehicle i (i = 1, 2) and optimal path

[0179] Step S22, establishing an objective function model of collaborative internal planning of collaborative path search planning, namely, an internal constraint model;

[0180] The collaborative internal planning model of collaborative path search planning is consistent with the internal planning model described in three-level task planning.

[0181] The design variables of collaborative internal planning are: navigation path X i The goal of collaborative internal planning is minJ=minF(X i ).

[0182] After each collaborative internal planning is completed, the return to the collaborative external planner is: the actual navigation time to reach the desired path end point and navigation paths

[0183] Step S3: Set algorithm parameters for task completion time planning and collaborative path search planning

[0184] 1. Task completion time planning:

[0185] (1) External planning

[0186] Given an estimated completion timeframe [Time min ,Time max ], Time min =0 is the lower limit of the range, Time max =3000s is the upper limit of the range;

[0187] The dimension of the planning problem is dim Time =1;

[0188] The number of algorithm populations is pop Time =20;

[0189] The number of algorithm iterations is Iteration Time =10.

[0190] (2) Intermediate planning

[0191] Given the angle range of the unmanned underwater vehicle entering the target detection circle [Theta1 min ,Theta1max ], Theta1 min =0 is the lower limit of the range, Theta1 max =2π is the upper limit of the range;

[0192] The dimension of the planning problem is dim1 Theta =1;

[0193] The number of algorithm populations is pop1 Theta =20;

[0194] The number of algorithm iterations is Iteration1 Theta =10.

[0195] (3) Internal planning

[0196] The up and down tilt angle range of the unmanned underwater vehicle is limited to [phi1 min ,phi1 max ]Inside, is the lower bound of the range, is the upper bound of the range;

[0197] The azimuth range of the UUV is limited to [psi1 min ,psi1 max ]Inside, is the lower bound of the range, is the upper bound of the range;

[0198] The search radius of the unmanned underwater vehicle is limited to [r1 min ,r1 max ]Inside, is the lower bound of the range, is the upper bound of the range;

[0199] The dimension of the planning problem is dim1 = 200;

[0200] The number of algorithm populations is pop1=50;

[0201] The number of algorithm iterations is Iteration1=30.

[0202] 2. Collaborative path search planning:

[0203] (1) Collaborative external planning

[0204] Given the angle range of the unmanned underwater vehicle entering the target detection circle [Theta2 min ,Theta2 max ], Theta2 min =0 is the lower limit of the range, Theta2 max =2π is the upper limit of the range;

[0205] The dimension of the planning problem is dim2 Theta =1;

[0206] The number of algorithm populations is pop2 Theta =50;

[0207] The number of algorithm iterations is Iteration2 Theta =30.

[0208] (2) Collaborative internal planning

[0209] The up and down tilt angle range of the unmanned underwater vehicle is limited to [phi2 min ,phi2 max ]Inside, is the lower bound of the range, is the upper bound of the range;

[0210] The azimuth range of the UUV is limited to [psi2 min ,psi2 max ]Inside, is the lower bound of the range, is the upper bound of the range;

[0211] The search radius of the unmanned underwater vehicle is limited to [r2 min ,r2 max ]Inside, is the lower bound of the range, is the upper bound of the range;

[0212] The dimension of the planning problem is dim2 = 200;

[0213] The number of algorithm populations is pop2=100;

[0214] The number of algorithm iterations is Iteration2=100.

[0215] Step S4, using the improved Parrot optimization algorithm to optimize the collaborative task planning parameters by solving the minimum value of the objective function to obtain the optimal task planning parameters;

[0216] Step S4 specifically includes the following steps S41-S44;

[0217] Step S41, initialize the parrot population status

[0218] Initialize the parrot population using the improved Circle Chaos Map, including:

[0219] To improve the efficiency and optimization capabilities of the Parrot Search algorithm, we have made innovative improvements to the initial population generation strategy. By introducing an improved Circle mapping method, we ensure a more even distribution of the population within the search space, enhancing the algorithm's exploration capabilities. This not only increases population diversity but also prevents the algorithm from prematurely converging, laying the foundation for finding the global optimal solution.

[0220] The original Circle chaos map expression is:

[0221]

[0222] However, the chaotic value distribution of the Circle chaotic mapping is uneven, and the values ​​between [0.2, 0.6] are relatively dense. Therefore, the Circle chaotic mapping formula is improved to make its chaotic value distribution more uniform.

[0223] The improved Circle chaos mapping expression is:

[0224]

[0225] The above formula implements the improved Circle Chaos Map. Specifically, it generates a random matrix A with N rows and M columns, where N is the population size and M is the spatial dimension of the optimization variable. It then iterates and calculates the value of each element.

[0226] W i,j =W min +A i ×(W max -W min )

[0227] The above formula uses the improved Circle chaos map to initialize the parrot population, where W i,j is the value of the jth variable specified by the i-th parrot individual, W max 、W min are the upper and lower bounds of the problem to be solved.

[0228] Step S42: Introducing a nonlinear inertia weight factor during the foraging phase

[0229] In order to control the degree of correlation between the parrot's position and its original position during the position update process, a nonlinear inertia weight factor ω is introduced. The method for determining this factor is as follows:

[0230]

[0231] Among them, t is the current iteration number and T is the maximum iteration number.

[0232] In the early iterations of the algorithm, a smaller ω value means that individual position updates are less affected by the current position, which encourages the algorithm to explore a wider solution space and thus enhances its global search capabilities. As the iterations progress, the ω value gradually increases, making individual updates more dependent on the current position. This helps narrow the search range and focus resources on promising areas, not only enhancing the algorithm's search efficiency within the local area but also accelerating the algorithm's pace of finding the optimal solution.

[0233] Through the improvement, the parrot's position update formula has been optimized, as shown below:

[0234]

[0235] In the formula, W i t Indicates the current position of the parrot, and W i t+1 Indicates a successfully updated location. W represents the average position within the current population, and Levy(M) is used to describe the parrot's navigation. best Indicates the best position from initialization to the current search.

[0236] Step S43: Introducing tangent function in the dwell phase

[0237] Use the tangent function to get the new stop position, which is in the interval [0, π]. After determining the new stop position, continue sailing:

[0238] W i t+1 =W i t +W best Levy(M)+tan(θ) ones(1,M)

[0239] Among them, ones(1,M) is a vector of all 1s of dimension M.

[0240] Step S44: Fear of strangers

[0241] Birds in general exhibit a natural fear of strangers, and parrots are no exception. As described below, their behavior to avoid strangers and seek a safe environment with their owners is as follows:

[0242]

[0243] Step S45: Introducing the sparrow alert strategy during the communication phase

[0244] To improve the Parrot algorithm's optimization efficiency, especially in its later stages, we introduced a sparrow vigilance strategy. This strategy enables the Parrot algorithm to approach the optimal solution more quickly. When perceiving a threat, marginal individuals quickly adjust their positions and move toward favorable areas; while central individuals engage in random exploration to approach superior individuals. This vigilance mechanism, incorporating the behavioral characteristics of sparrows, accelerates convergence and improves search performance.

[0245] Use this strategy to replace the original parrot communication behavior. The formula is as follows:

[0246]

[0247] During the execution of the algorithm, W best and W worst To represent the positions of the best and worst solutions currently searched. β is a random number drawn from a standard normal distribution. It is a key parameter for step size control and ranges from -1 to 1. It is used to adjust the dynamics of the search. In the second stage, after the position of the i-th parrot is updated, its fitness value is determined by F i The global best and worst fitness values ​​are represented by F best and F worst Definition, where ε is a constant that prevents the denominator from being zero and ensures the stability of numerical calculations.

[0248] Step S46: Repeat the above steps

[0249] Task completion time plan:

[0250] For UUVs, regardless of whether they are collaborative or not, the path planning algorithm that achieves the shortest possible time to reach the target detection circle is the underlying algorithm for collaborative mission planning. This algorithm consists of an external planning loop that optimizes the arrival time, an intermediate planning loop that optimizes the shortest possible path planning, and an internal planning loop that optimizes the path planning.

[0251] External planning is used to plan the encounter time. The goal is to assign a reasonable task completion time. At this time point, the target moves to a certain position, and the unmanned underwater vehicle simultaneously reaches the detection circle corresponding to the target position, finds the target and switches to attack.

[0252] The intermediate planning is to plan the fastest navigation time to the target detection circle based on a certain expected encounter time and the corresponding predicted target position. The goal is to find the position on the target detection circle where the unmanned underwater vehicle can reach the fastest, that is, the relative angle.

[0253] Internal planning is based on a specific location on a given target detection circle and performs path planning. The goal is to find the fastest path to a location on the target detection circle and the actual flight time.

[0254] The intermediate plan searches for the position on the target detection circle based on the actual navigation time fed back by the internal plan, finds the fastest time to reach the target detection circle, and then feeds this time back to the external planner.

[0255] The external planning calculates the difference between the shortest navigation time feedback from the intermediate planning and the expected encounter time, and adjusts the expected encounter time according to the difference until the difference meets a certain threshold requirement or the planning termination condition is met. If the threshold requirement is met, it is considered that the shortest encounter time, the position on the target detection circle and the implementation path have been found.

[0256] Collaborative path search planning:

[0257] Collaborative path search planning mainly focuses on designing near the mission entry angle of the last-departing unmanned underwater vehicle, which is equivalent to constraining the target entry angle design range when planning the mission of the first-departing unmanned underwater vehicle.

[0258] To solve the coordination problem of three unmanned underwater vehicles, the first unmanned underwater vehicle that departs first directly uses collaborative path search planning to design the path when launching.

[0259] When the last unmanned underwater vehicle is launched, the mission completion time planning algorithm is first used to calculate the shortest encounter time between the last unmanned underwater vehicle and the target detection circle and the actual shortest navigation time of the last unmanned underwater vehicle.

[0260] Next, the predicted position of the target is determined based on the encounter time, and the optimization goal is to minimize the difference between the time when the first unmanned underwater vehicle reaches the detection circle and the actual shortest navigation time of the last unmanned underwater vehicle. The collaborative path search planning is performed on the unmanned underwater vehicle that departs first. That is, the collaborative path search planning takes the shortest encounter time between the last-departing unmanned underwater vehicle and the target detection circle and the actual shortest navigation time of the last-departing unmanned underwater vehicle as input, and takes minimizing the collaborative time difference as the goal, and optimizes the angle at which the first-departing unmanned underwater vehicle reaches the target detection circle as the design variable.

[0261] To simplify the problem, the last unmanned underwater vehicle to depart is numbered 3, and the first unmanned underwater vehicle to depart is numbered i (i = 1, 2).

[0262] Taking the last departing unmanned underwater vehicle as the benchmark, the mission completion time planning model is used to plan the mission of the last departing unmanned underwater vehicle, and the mission completion time best_time3 of the last departing unmanned underwater vehicle and the actual shortest navigation time are obtained. Then, based on the task completion time, the expected completion time best_time for the first unmanned underwater vehicle to plan its task can be determined. i =best_time n (i=1,2) and the center position of the target detection circle TarPos(best_time i )(i=1,2).

[0263] Step S5: Output collaborative task planning results:

[0264] 1. Task completion time planning output

[0265] The optimal task completion time best_time3 = 2005.0s;

[0266] The time difference between the unmanned underwater vehicle 3 and the target is Δt3 = 12.6410s;

[0267] 3. Optimal positioning angle of unmanned underwater vehicle

[0268] Actual shortest time for unmanned underwater vehicle 3 to be deployed

[0269] Unmanned underwater vehicle 3 voyage3 = 8857 m;

[0270] Optimal path for unmanned underwater vehicles 3

[0271] 2. Collaborative path search planning output

[0272] Optimal coordinated positioning time of unmanned underwater vehicle 1

[0273] Optimal positioning angle of unmanned underwater vehicle 1

[0274] The time difference between UUV 1 and UUV 2 is Δt1 = 0.1745 s;

[0275] Unmanned underwater vehicle 1 voyage1 = 8033 m;

[0276] Optimal path for unmanned underwater vehicle 1

[0277] Optimal coordinated positioning time of unmanned underwater vehicle 2

[0278] Optimal positioning angle of unmanned underwater vehicle 2

[0279] The time difference between UAV 2 and UAV 2 is Δt2 = 6.3120 s;

[0280] Unmanned underwater vehicle 2 voyage2 = 7439 m;

[0281] Optimal path for unmanned underwater vehicle 2

[0282] A collaborative mission planning method based on the improved Parrot optimization algorithm enables an unmanned underwater vehicle to reach the target's encircling circle in the shortest possible time. This method, based on the standard Parrot optimization algorithm, employs an improved Circle chaos map, a nonlinear inertia weight factor, a tangent function, and a sparrow alert strategy to improve the Parrot optimization algorithm. This increases the diversity of individuals, enabling the Parrot optimization algorithm to have better global development capabilities in the early stages of iteration and excellent local exploration capabilities in the later stages of iteration, thereby improving the algorithm's convergence speed.

[0283] Furthermore, the above-described figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention and are not intended to be limiting. It is readily understood that the processes illustrated in the above-described figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0284] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the invention herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the invention being indicated by the claims.

[0285] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings and that various modifications and variations can be made without departing from the scope thereof, which is limited only by the appended claims.

Claims

1. A collaborative task planning method based on an improved Parrot optimization algorithm, characterized in that: The method comprises: Obtain the initial positions and velocities of all UUVs and targets, determine the parameters and objective functions for the three-level mission planning: external planning, intermediate planning, and internal planning, and plan the time it takes for the UUV to complete its mission based on the last UUV to depart. The target location and the unmanned underwater vehicle mission completion time are determined based on the mission completion time plan. Then, the parameters and objective functions of the collaborative external planning and collaborative internal planning two-level mission planning are determined, and the collaborative path search planning of the three unmanned underwater vehicles is carried out to achieve the task of collaborative target discovery by the three unmanned underwater vehicles. The improved Parrot optimization algorithm is used to optimize the collaborative task planning parameters by solving the minimum value of the objective function of each layer of task planning, and the optimal task planning parameters are obtained. The determination of the improved Parrot optimization algorithm includes: The improved Circle chaos map is used to initialize the parrot population; In the foraging phase of the Parrot Optimization Algorithm, a nonlinear inertia weight ω is introduced to adjust the correlation between the position update and the current position; The tangent function is used throughout the entire process during the stop phase of the Parrot optimization algorithm; Added sparrow alert strategy to the communication phase of the parrot optimization algorithm.

2. The method according to claim 1, characterized in that The parameters and objective functions of the three-level task planning of external planning, intermediate planning and internal planning are determined, specifically including: The external planning is an arrival time design loop. The external planning performs encounter time planning. The goal is to assign a reasonable task completion time. At this time point, the target moves to a certain position, and the unmanned underwater vehicle simultaneously reaches the detection circle corresponding to the target position, finds the target and switches to attack. The intermediate planning is the shortest time design loop. The intermediate planning is to plan the fastest navigation time to the target detection circle according to a certain expected encounter time and the corresponding predicted target position. The goal is to find the position on the target detection circle that the unmanned underwater vehicle can reach the fastest, that is, the relative angle. The internal planning is a path planning and design loop. The internal planning performs arrival path planning based on a specific location on a given target detection circle. The goal is to find the fastest path to a certain location on the target detection circle and the actual navigation time. The intermediate planner searches for the position on the target detection circle based on the actual navigation time fed back by the internal planner, finds the fastest time to reach the target detection circle, and then feeds this time back to the external planner; the external planner calculates the difference between the shortest navigation time and the expected encounter time based on the shortest navigation time fed back by the intermediate planner, and adjusts the expected encounter time according to the difference until the difference meets a certain threshold requirement or the planning termination condition is met. If the threshold requirement is met, it is considered that the shortest encounter time, the position on the target detection circle and the implementation path have been found.

3. The method according to claim 2, characterized in that The unmanned underwater vehicle mission completion time planning specifically includes: Establish the objective function model of external planning - time difference model; The purpose of mission completion time planning is to make the UUV reach the target detection circle within the specified mission time. That is, the difference between the time it takes for the UUV to move along the path to the target detection circle and the time it takes for the target to move should be as small as possible, and the encounter time should be as small as possible. Therefore, the objective function of external planning is: J=give_time+ωΔt,Δt=|t fxq -give_time| Among them, t fxq is the actual navigation time of the UUV to reach the target detection circle; give_time is the given estimated mission completion time, which is also the target movement time; Δt is the difference between the actual navigation time of the UUV and the estimated encounter time between the two; ω is the optimization weight; The design variable for external planning is: give_time The objectives of external planning are to: give_time minJ(give_time) The inputs passed to the intermediate planner for each optimization are: the expected task completion time give_time and the predicted target position TarPos(give_time) after give_time; After each intermediate plan is completed, the input returned to the external planner is: the actual shortest navigation time t that the unmanned underwater vehicle can achieve corresponding to the target position detection circle under the expected task completion time fxq (give_time) and optimal path X best (t fxq (give_time)); The final output of external planning is: the optimal task completion time best_time, the actual shortest navigation time t fxq (best_time), optimal path X best (t fxq (best_time)).

4. The method according to claim 2, characterized in that The unmanned underwater vehicle mission completion time planning specifically includes: Establish the objective function model of intermediate planning - the actual navigation time model of the unmanned underwater vehicle; The goal of the intermediate planning problem is to minimize the navigation time of the unmanned underwater vehicle on the detection circle to reach a given target position, so the objective function of the intermediate planning is: J=t fxq The design variables for the intermediate planning are: entry angle The goal of intermediate planning is The input passed to the internal planner for each planning is: the target position TarPos(give_time) predicted at the current expected task completion time give_time at the entry angle The expected path end point at time: Where, Dis detect is the detection range of the unmanned underwater vehicle, that is, the radius of the target detection circle; After each intermediate plan is completed, the input returned to the external planner is: the shortest navigation time to reach the desired path end point and the optimal path under the corresponding entry angle 5. The method according to claim 2, characterized in that The unmanned underwater vehicle mission completion time planning specifically includes: Establish the objective function model of internal planning - internal constraint model; Internal planning is a path planning problem, that is, finding a path from the initial point to the given task end point The optimal path And the corresponding actual exercise time The design variables of internal planning are: navigation path X i The goal of internal planning is minJ=minF(X i ); After each internal planning is completed, the intermediate planner is returned with the actual navigation time to reach the desired path end point. and navigation paths 6. The method according to claim 1, characterized in that The unmanned underwater vehicle collaborative path search planning specifically includes: The purpose of the collaborative external planning of the collaborative path search planning is to make the unmanned underwater vehicle reach the specified mission time best_time i (i=1,2) to reach the target detection circle, that is, the time required for the unmanned underwater vehicle to move along the path to the target detection circle and The smaller the time difference, the better, that is, the detection circle is reached at the same time, so the objective function of the collaborative external planning of the collaborative path search planning is: Where, is the actual navigation time of the unmanned underwater vehicle i (i=1,2) to reach the target detection circle; The design variables of collaborative external planning are: the entry angle of the unmanned underwater vehicle i (i = 1, 2) on the target detection circle The objectives of collaborative external planning are to: mice i (i=1,2) The input passed to the collaborative internal planner for each optimization is: the current best_time i (i=1,2) The predicted target position TarPos(best_time i )(i=1,2) at the entry angle The expected path end point at time: After each collaborative internal planning is completed, the input returned to the collaborative external planner is: best_time i (i=1,2) corresponding to the target position detection circle entry angle The shortest practical flight time that an unmanned underwater vehicle can achieve and the optimal path The final output of the collaborative external planning is: the optimal mission completion time of the unmanned underwater vehicle i (i = 1, 2) and the optimal path The collaborative internal planning model of collaborative path search planning is consistent with the internal planning model described in three-level task planning.

7. The method according to claim 1, characterized in that The improved Circle chaotic map is used to initialize the parrot population, specifically including: The Circle chaos mapping formula is improved to make its chaos value distribution more uniform. The improved Circle chaos mapping expression is: Based on the above formula, a random matrix A with N rows and M columns is generated, where N is the population size and M is the spatial dimension of the optimization variable; then the value of each element is calculated iteratively: W i,j =W min +A i ×(W max -W min ) Among them, W i,j is the value of the jth variable specified by the i-th parrot individual, W max 、W min are the upper and lower bounds of the problem to be solved.

8. The method according to claim 1, characterized in that The nonlinear inertia weight ω is introduced in the foraging phase of the parrot optimization algorithm to adjust the correlation between the position update and the current position, as shown below: Among them, W i t Indicates the current position of the parrot, W i t+1 Indicates the location of the successful update, represents the average position within the current population, Levy(M) is used to describe the parrot's navigation, W best Indicates the best position from initialization to the current search.

9. The method according to claim 1, characterized in that The introduction of the tangent function during the dwell phase specifically includes: Use the tangent function to get the new stop position, which is in the interval [0, π]. After determining the new stop position, continue sailing: W i t+1 =W i t +W best ·Levy(M)+tan(θ)·ones(1,M) Among them, ones(1,M) is a vector of all 1s of dimension M.

10. The method according to claim 1, characterized in that The sparrow alert strategy introduced during the communication phase specifically includes: Use the sparrow alert strategy to replace the original parrot communication behavior. The formula is as follows: During the execution of the algorithm, W best and W worst To represent the position of the best solution and the worst solution currently searched, β is a random number drawn from the standard normal distribution. It is a key parameter for step size control, ranging from -1 to 1, and is used to adjust the dynamics of the search. In the second stage, after the position of the i-th parrot is updated, its fitness value is determined by F i The global best and worst fitness values ​​are represented by F best and F worst Definition, where ε is a constant that prevents the denominator from being zero and ensures the stability of numerical calculations.

Images