Method for optimizing heterogeneous cooperation of unmanned aerial vehicle and unmanned ship for encircling based on pigeon thinking

Abstract

The application discloses a pigeon type thinking based optimization unmanned aerial vehicle and unmanned ship heterogeneous cooperative hunting method, comprising the following steps: establishing a mathematical model of the unmanned aerial vehicle and the unmanned ship; designing a leader-follower model of the unmanned aerial vehicle and the unmanned ship based on a binary tree algorithm; the unmanned aerial vehicle leader and the unmanned ship leader communicate; designing a hunting control law of the unmanned ship; the leader-follower model of the unmanned aerial vehicle detects and acquires position information of a marine target to be hunted; the unmanned aerial vehicle leader sends the position information to the unmanned ship leader; the unmanned ship hunts the marine target to be hunted according to the hunting control law of the unmanned ship; and the leader-follower model is optimized based on pigeon type thinking to obtain an optimal leader, so that the heterogeneous cooperative hunting of the unmanned aerial vehicle and the unmanned ship is realized. The existing unmanned aerial vehicle and unmanned ship cooperation system cannot meet the actual environmental needs, communication problems occur frequently, and information exchange between the unmanned aerial vehicle and the unmanned ship also leads to low efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of multi-water surface ship path planning, and particularly relates to a heterogeneous cooperative hunting method for unmanned aerial vehicles and unmanned ships based on pigeon thinking optimization. BACKGROUND

[0002] With the rapid development of ship and unmanned aerial vehicle control technology, unmanned control system has become a trend to perform tasks, and unmanned systems have been used in various applications, such as surveillance, positioning, search and rescue, and tracking and surrounding. The platforms involved are homogeneous unmanned aerial vehicles, unmanned ships, etc. For example, unmanned aerial vehicles have the characteristics of fast speed and wide field of view; and unmanned ships perform better in terms of endurance and stability. In recent years, in order to achieve better cooperation and efficiency, unmanned aerial vehicles and unmanned ships have gradually been valued in civil and military fields, such as reconnaissance tasks, static or mobile target tracking and detection, geographic information collection, mobile robot planning and unmanned system control. However, the existing unmanned aerial vehicle and unmanned ship cooperation system cannot meet the requirements of actual environment and tasks, and some problems need to be solved before the unmanned aerial vehicle and unmanned ship system can be widely used in practice, for example, the communication between the unmanned aerial vehicle and the unmanned ship is prone to problems, and the information exchange between them also leads to low efficiency. In addition, the existing target tracking and surrounding of unmanned aerial vehicle and unmanned ship cooperation involves multi-sensor information fusion, target positioning, path planning and formation control, and is not suitable for the real-time performance of the whole hybrid system. SUMMARY

[0003] The present application provides a heterogeneous cooperative hunting method for unmanned aerial vehicles and unmanned ships based on pigeon thinking optimization to overcome the above technical problems.

[0004] In order to achieve the above purpose, the technical scheme of the present application is:

[0005] A heterogeneous cooperative hunting method for unmanned aerial vehicles and unmanned ships based on pigeon thinking optimization, comprising the following steps:

[0006] Step S1: establishing a mathematical model of the unmanned aerial vehicle and the unmanned ship;

[0007] Step S2: setting the leader and the follower of the unmanned aerial vehicle and the unmanned ship through the mathematical model of the unmanned aerial vehicle and the unmanned ship, and designing a leader-follower model of the unmanned aerial vehicle and the unmanned ship based on a binary tree algorithm;

[0008] Microwave two-way communication is adopted between the leader and the follower of the unmanned aerial vehicle and between the leader and the follower of the unmanned ship; and the leader of the unmanned aerial vehicle is communicatively linked with the leader of the unmanned ship;

[0009] Step S3: designing a hunting control law of the unmanned ship according to the leader-follower model of the unmanned aerial vehicle and the unmanned ship;

[0010] Step S4: optimizing the leader-follower model of the UAV and the unmanned ship based on pigeon thinking to obtain an optimal leader-follower model;

[0011] Step S5: detecting the position information of the marine target to be captured through the optimal leader-follower model of the UAV;

[0012] The position information is the relative horizontal distance between the marine target to be captured and the leader of the UAV; the leader of the UAV sends the position information to the leader of the unmanned ship, and the leader of the unmanned ship sends the position information to the follower of the unmanned ship;

[0013] The leader and the follower of the unmanned ship capture the marine target to be captured according to the capture control law of the unmanned ship.

[0014] Further, the mathematical model of the UAV in step S1 is

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] wherein, represents the speed of the UAV in the x direction; represents the speed of the UAV in the y direction; represents the speed of the UAV in the Z direction; the x direction, the y direction and the Z direction are all artificially set according to the geographic coordinate system; represents the yaw rate of the UAV; s.t. |a nu | represents the real-time acceleration of the UAV; nu represents the number of UAVs, x nu , y nu , z nu is the position of the UAV relative to the ground, v nu is the ground speed of the UAV, represents the first derivative of the ground speed of the UAV; a nu is the acceleration of the UAV, g is the gravitational acceleration, ψ nu is the heading angle of the UAV, φ nu is the roll angle of the UAV, λ nu is the rate of change of the altitude of the UAV, x nu , y nu , v nu , anu and ψ nu represent state variables and control variables of the UAV; v u min and v u max represent speed constraints of the UAV; a u max and represent maximum acceleration and roll angle of the UAV, respectively;

[0021] The mathematical model of the UAV is

[0022]

[0023]

[0024]

[0025] wherein, represents the x-direction derivative of the UAV; represents the y-direction derivative of the UAV; s.t.v g min represents the real-time speed of the UAV; x ng , y ng , v ng and ψ ng represent the position, speed and heading of the UAV, respectively; a ng represents the acceleration of the UAV, v g min and v g max represent speed constraints of the UAV.

[0026] Further, the leader-follower model of the UAV and the unmanned ship is designed based on the binary tree algorithm in step S2, specifically

[0027] Step S2.1: a hybrid multi-layer distributed system is set, which includes a plurality of UAVs and a plurality of unmanned ships;

[0028] The UAVs and the unmanned ships establish a communication topology graph based on the binary tree algorithm, and set the leaders and followers of the initial UAVs and the unmanned ships, the UAV followers will detect the marine target to be captured in the detection area, and send the position information of the marine target to be captured to the UAV leader;

[0029] Step S2.2: the UAV leader generates the final position information of the marine target to be captured according to the position information of the marine target to be captured sent by the UAV follower and the relative distance between the UAV leader and the UAV follower;

[0030] The drone leader sends the final location information of the target to be captured at sea to the unmanned vessel leader;

[0031] Step S2.3: The unmanned vessel leader sends the final location information of the target to be captured at sea to the unmanned vessel followers, so that the unmanned vessel leader and followers can provide capture data for the target at sea based on the final location information of the target.

[0032] Furthermore, the detection area of ​​the drone is obtained by collecting the coverage area of ​​the detector, and the calculation formula is as follows:

[0033]

[0034] In the formula: R S This indicates the coverage area of ​​the data acquisition detector; z nu This refers to the drone's position relative to the ground. This indicates the maximum value of the drone's built-in line-of-sight parameters.

[0035] Furthermore, the control law for the unmanned vessel described in step S3 is specifically as follows:

[0036]

[0037] in, and n g1 and n g2 The expected capture radius and the expected carrying space; n g1 and n g2 The unmanned vessel's identification number; ρ ng (t) represents the radius as it changes over time; Represents any unmanned vessel n during the encirclement and capture. g1 speed; This indicates that during the encirclement and capture, there is any unmanned vessel n. g1 Adjacent unmanned vessels n g2 velocity; ρ ng This refers to the radius during the actual encirclement and capture. x represents the expected radius after the encirclement is completed. ng Here is the x-coordinate of the unmanned surface vessel relative to the ground; x nt The x-coordinate of the target relative to the ground; y-coordinate. ng Here is the y-coordinate of the unmanned surface vessel relative to the ground; y nt The target being surrounded has the y-coordinate relative to the ground. The real-time speed of the target for the capture of the unmanned vessel; v nt (t) represents the real-time velocity of the target being pursued.

[0038] Furthermore, the relative distance between the drone and the target at sea is calculated using the following formula:

[0039]

[0040] wherein, / f dist denotes the relative distance between the UAV and the marine target to be captured; denotes the horizontal distance from the UAV to the tracked target; N u is the number of selected UAVs; n u is the model number of all UAVs.

[0041] Further, the leader-follower model is optimized based on the pigeon thinking in step S4 to obtain an optimal leader-follower model, specifically

[0042] Step S5.1: In the original pigeon thinking optimization map and compass operator, the unmanned ship or the unmanned aerial vehicle is defined as a pigeon individual in the pigeon population, the random speed and position of each pigeon individual are set, and the fitness value of each pigeon is compared to obtain the current global optimal solution;

[0043] The initial speed and position of each pigeon individual are updated, and the fitness value of the pigeon individual is compared to obtain the latest global optimal solution;

[0044] The initial position and speed of pigeon i are defined as and

[0045] And the initial position and speed state update formula of pigeon i is:

[0046]

[0047] x i (t+1)=x i (t)+v i (t+1)

[0048] Where i=1, 2, 3…N p ;N p is the population size; D and R are the size of the search space, the map and the compass factor respectively; represents the current global optimal solution; X i denotes the position of pigeon i; denotes the distance between pigeon i and pigeon 1; denotes the distance between pigeon i and pigeon d; denotes the speed difference between pigeon i and pigeon 1; denotes the speed difference between pigeon i and pigeon d; v i (t+1) indicates the speed of pigeon i at t+1; x i (t) indicates the position of pigeon i at t; e-Rt A limiting function representing velocity; x i (t+1) represents the position of pigeon i at time t+1; V i (t) represents the velocity of pigeon i at time t;

[0049] Step S5.2: Using the energy estimation function The formula for selecting a leader in a group is expressed as follows:

[0050]

[0051] in, The energy of a leader that influences their attributes; Energy values ​​representing the performance of group search; Represents the energy estimation function;

[0052] The energy values ​​of each pigeon in the pigeon group are sorted in descending order, and pigeons with energy values ​​greater than the preset leader energy threshold are selected as the candidate leader set.

[0053] The energy of a leader that influences leader attributes The calculation formula is

[0054]

[0055]

[0056] in, T0 represents the number of iterations, and Fit() represents the fitness value; This represents the estimated position of the leader at time t; This represents the estimated position of the leader at time t+1; This represents an estimate of the globally optimal solution representing the leader's central position at time t. This represents the estimated position of the center at time t;

[0057] The energy value of the group search performance The calculation formula is

[0058]

[0059] Where Np represents the individual size in pigeon-like thinking; This represents the estimated position of pigeon i at time t; This represents the estimated position of the i-th pigeon at time t+1;

[0060] Step S5.3: All individuals in the pigeon flock are energy-matched according to their fitness value Fit(), and randomly divided into N groups. d The set of candidate leaders is defined as a set of sub-intervals.

[0061] X cand ={X cand1 ,X cand2 ,...,X candNc}

[0062] Wherein, Nc represents the number of unmanned aerial vehicle or unmanned ship candidates; a random individual in set X cand is selected as a new leader;

[0063] The center X center (t) of the pigeon group is updated according to the new leader, and the sub-interval center X centeri (t) is changed to X centeri (t) position update formula:

[0064] X i (t+1)=X i (t)+rand(X center (t)-X i (t))

[0065]

[0066] Wherein, X j (t) represents the position of pigeon j; i, j represent different target pigeons; rand(), f() represent functions; S d (Xi), respectively, the neighbor set and the number of neighbors of the individual;

[0067] Step S5.4: According to the behavior control of invalid space, avoid the collision between unmanned aerial vehicle or unmanned ship individuals through Jacob matrix J ng , said Jacob matrix J ng is

[0068]

[0069] Wherein, P ns =[x ns ,y ns ]=P nr ; x ns is the x direction position of the detector relative to the ground; y ns is the y direction position of the detector relative to the ground; x ng is the x direction position of the unmanned ship relative to the ground; y ng is the y direction position of the unmanned ship relative to the ground; P ng is the vector form of the position of the unmanned ship; P ns is the vector form of the range of the detector; P nr represents the velocity feedback gain matrix;

[0070] Step S5.5 is according to Jacobian matrix J ng The total control force surrounding the task is obtained as:

[0071]

[0072]

[0073] wherein, represents the total control force surrounding; represents the repulsion force between individuals; represents the null space behavior control; u ng represents the acceleration of the unmanned ship; and K3 is a gain factor; is a potential function of the detector, which is determined by the selected detector; is the minimum distance of the detector; and K2 is a gain factor; is the maximum distance of the detector; P ng is a vector form of the position of the unmanned ship; P ns is a vector form of the range of the detector; x ns is the x-direction position of the detector relative to the ground; x ng is the x-direction position of the unmanned ship relative to the ground; y ng is the y-direction position of the unmanned ship relative to the ground; y ns is the y-direction position of the detector relative to the ground.

[0074] Beneficial effects: the application provides a heterogeneous collaborative hunting method for optimizing unmanned aerial vehicles and unmanned ships based on pigeon thinking, designs a leader-follower model for unmanned aerial vehicles and unmanned ships based on a binary tree algorithm, provides convenience for communication management and collaboration consistency within a homogeneous group, optimizes the target tracking process by using energy estimation pigeon thinking optimization (EEPI0) according to the operability and wide detection area of the unmanned aerial vehicle in target tracking, designs a control law based on airspace behavior and a switchable topological structure, realizes the collaborative hunting of the unmanned ship on the target to be surrounded by real-time communication between the unmanned aerial vehicle leader and the unmanned ship leader, and solves the problems that the existing unmanned aerial vehicle and unmanned ship cooperation system cannot meet the requirements of actual environment and tasks, the communication between the unmanned aerial vehicle and the unmanned ship often fails, and information exchange between the unmanned aerial vehicle and the unmanned ship leads to low efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0075] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are only some embodiments of the present application, and the other drawings can be obtained by those skilled in the art without creative labor under the premise of the drawings.

[0076] Figure 1 The flow chart of the pigeon thinking based optimization method for heterogeneous cooperation of unmanned aerial vehicle and unmanned ship;

[0077] Figure 2 The leader-follower model diagram of the pigeon thinking based optimization method for heterogeneous cooperation of unmanned aerial vehicle and unmanned ship;

[0078] Figure 3 The model diagram of the unmanned ship of the pigeon thinking based optimization method for heterogeneous cooperation of unmanned aerial vehicle and unmanned ship. DETAILED DESCRIPTION

[0079] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0080] The present embodiment provides a pigeon thinking based optimization method for heterogeneous cooperation of unmanned aerial vehicle and unmanned ship, as shown in the following steps: Figure 1

[0081] Step S1: establishing the mathematical model of the unmanned aerial vehicle and the unmanned ship;

[0082] Step S2: setting the leader and the follower of the unmanned aerial vehicle and the unmanned ship through the mathematical model of the unmanned aerial vehicle and the unmanned ship, and designing the leader-follower model of the unmanned aerial vehicle and the unmanned ship based on the binary tree algorithm;

[0083] The microwave two-way communication is adopted between the leader and the follower of the unmanned aerial vehicle and between the leader and the follower of the unmanned ship; and the communication link is established between the leader of the unmanned aerial vehicle and the leader of the unmanned ship;

[0084] Step S3: designing the hunting control law of the unmanned ship according to the leader-follower model of the unmanned aerial vehicle and the unmanned ship;

[0085] ​Step S4: Optimize the leader-follower model of the UAV and the unmanned ship based on pigeon thinking to obtain an optimal leader-follower model;

[0086] Step S5: Detect the position information of the sea target to be surrounded by the optimal leader-follower model of the UAV;

[0087] The position information is the relative horizontal distance between the sea target to be surrounded and the leader of the UAV; the leader of the UAV sends the position information to the leader of the unmanned ship, and the leader of the unmanned ship sends the position information to the follower of the unmanned ship;

[0088] The leader and the follower of the unmanned ship surround the sea target to be surrounded according to the surrounding control law of the unmanned ship. Through the mathematical model of the UAV and the unmanned ship, the leader and the follower of the UAV and the unmanned ship are initially set, and the leader-follower model of the UAV and the unmanned ship is designed based on the binary tree algorithm, which provides convenience for communication management and cooperation consistency in the homogeneous group, while ensuring information sharing between the heterogeneous groups of the UAV and the unmanned ship; According to the advantages of the UAV in target tracking, such as operability and wide detection area, the EEPIO (Energy Estimation Pigeon Optimization) is used to optimize the target tracking process; The control law based on airspace behavior and switchable topological structure is designed, the leader of the UAV and the leader of the unmanned ship are communicated in real time, the cooperative surrounding of the unmanned ship to the target to be surrounded is realized, and the problems that the existing UAV and unmanned ship cooperation system cannot meet the requirements of actual environment and task, the communication between the UAV and the unmanned ship often fails, and the information exchange between them leads to low efficiency are solved.

[0089] In specific embodiments, the mathematical model of the UAV established in step S1 is

[0090]

[0091]

[0092]

[0093]

[0094]

[0095] wherein, vx represents the speed of the UAV in the x direction; vy represents the speed of the UAV in the y direction; vz represents the speed of the UAV in the Z direction; the x direction, the y direction and the Z direction are artificially set according to the geographic coordinate system; represents the heading angle speed of the UAV; s.t. |anu| represents the real-time acceleration of the UAV; nu represents the number of UAVs, x nu , y nu , z nu is the position of the UAV relative to the ground, v nu is the ground speed of the UAV, represents the first-order derivative of the ground speed of the UAV; a nu is the acceleration of the UAV, g is the gravitational acceleration, ψ nu is the heading angle of the UAV, φ nu is the roll angle of the UAV, λ nu is the rate of change of the altitude of the UAV, x nu , y nu , v nu , a nu and ψ nu represent the state variables and control variables of the UAV; v u min and v u max represent the speed constraints of the UAV, a u max and respectively represent the maximum acceleration and roll angle of the UAV;

[0096] The mathematical model of the UAV is

[0097]

[0098]

[0099]

[0100] wherein, represents the x-direction derivative of the UAV; represents the y-direction derivative of the UAV; s.t. v g min represents the real-time speed of the UAV; x ng , y ng , v ng and ψ ng respectively represent the position, speed and heading of the UAV; a ng represents the acceleration of the UAV, v g min and v g max represent the speed constraints of the UAV.

[0101] In specific embodiments, the leader-follower model of the UAV and the unmanned ship is designed based on the binary tree algorithm in step S2, specifically

[0102] Step S2.1: Set up a hybrid multi-layer distributed system, which includes several drones and several unmanned vessels;

[0103] The drone and unmanned vessel establish a communication topology based on a binary tree algorithm, and set an initial leader and follower for the drone and unmanned vessel. The drone follower will detect the maritime targets to be captured in the detection area and send the location information of the maritime targets to be captured to the drone leader.

[0104] Step S2.2: The drone leader generates the final location information of the target at sea based on the location information of the target to be captured at sea sent by the drone follower and the relative distance between the drone leader and the drone follower.

[0105] The drone leader sends the final location information of the target to be captured at sea to the unmanned vessel leader;

[0106] Step S2.3: The unmanned vessel leader sends the final location information of the target to be captured at sea to the unmanned vessel followers, so that the unmanned vessel leader and followers can provide capture data for the target at sea based on the final location information of the target.

[0107] like Figure 2 As shown, a hybrid multi-layered distributed system of UAVs and unmanned surface vessels (USVs) is designed. The UAVs and USVs establish a communication topology based on a binary tree algorithm. Real-time communication between the UAVs and USVs enables target tracking and coordinated capture. Target tracking is transformed into an optimization problem to ensure coverage of detected targets. UAVs can communicate with each other (indicated by arrows). A UAV is selected as the information processing node (UAV leader) to generate control commands for the other UAVs (UAV followers). Due to limited observation range, UAVs should promptly detect and track moving targets. If a USV follower's detector detects a target to be captured, the follower transmits the detected target's location information to the UAV. The drone leader generates position information between itself and the target to be captured based on the position information received from the drone followers. The methods or approaches by which the drone acquires or generates position information with the target are existing known technologies and are not the inventive point of this application, and will not be elaborated here. At this time, the drone leader establishes communication with the unmanned vessel leader and sends the position information of the target to the unmanned vessel leader. The unmanned vessel leader then sends the received position information of the target to each unmanned vessel follower. The unmanned vessel leader and unmanned vessel followers, according to the parameters of their respective set unmanned vessel capture control laws, coordinate to capture the target, thereby achieving a highly efficient capture process.

[0108] In specific embodiments, the line-of-sight model obstacle profile of the unmanned aerial vehicle (UAV) can be simplified as a circumferential cylinder sharing the same height and base to reduce the complexity of calculation, and the detection area of the UAV is obtained by the coverage area of the acquisition probe, and the calculation formula is

[0109]

[0110] In the formula, R S represents the coverage area of the acquisition probe; z nu is the position of the UAV relative to the ground; represents the maximum value of the line-of-sight parameter of the UAV.

[0111] In specific embodiments, the leader and follower unmanned ships are set to the same pursuit radius and pursuit angle to surround the target for the surrounding task, the pursuit radius is the distance between the center of the unmanned ship and the center of the target to be surrounded, and the pursuit angle is the relative angle between the unmanned ship and the target to be surrounded in the set geographic coordinate system after the completion of the pursuit. The pursuit control law of the unmanned ship in step S3 is specifically

[0112]

[0113] wherein, and are the expected pursuit radius and expected carrying space of n g1 and n g2 , respectively; n g1 and n g2 are the numbers of the unmanned ships; p ng (t) means the radius changing with time; represents the speed of any unmanned ship n g1 during the pursuit; represents the speed of the adjacent unmanned ship n g1 to any unmanned ship n g2 during the pursuit; p ng is the actual radius during the pursuit; is the expected radius after the completion of the pursuit; x ng is the x-coordinate of the unmanned ship relative to the ground; x nt is the x-coordinate of the target to be surrounded relative to the ground; y ng is the y-coordinate of the unmanned ship relative to the ground; y nt is the y-coordinate of the target to be surrounded relative to the ground; is the real-time speed of the target to be surrounded by the unmanned ship; v nt (t) is the real-time speed of the target to be surrounded.

[0114] In specific embodiments, when the acquisition detector of the UAV discovers the target to be captured, the cooperative capture with the unmanned ship is started, and the relative distance between the UAV and the target to be captured at sea is used to describe the tracking process, and the relative distance between the UAV and the target to be captured at sea is calculated as follows:

[0115]

[0116] Wherein, f dist represents the relative distance between the UAV and the target to be captured at sea; represents the horizontal distance from the UAV to the tracked target; N u is the number of selected UAVs; n u is the model number of all UAVs.

[0117] As shown in Figure 3 , due to the limitations of sensor capability and self-energy, when the unmanned ship leader sends the target capture instruction to the unmanned ship follower, the sensor of each unmanned ship (unmanned ship) cannot cover all unmanned ship team members; at the same time, in the case of confrontation, the unmanned ship may be attacked by external force, resulting in failure of detection or communication, therefore, only a part of the unmanned ship can detect the target to be captured and share the perception information with other unmanned ships, at this time, it is necessary to reconstruct the topology or restore the communication; if the follower of the unmanned ship is detected within the sensor monitoring range, the follower is changed into the leader, and the communication with the UAV leader is established, at the same time, the position information of the target to be captured sent by the UAV leader is sent to the rest of the unmanned ships for cooperative capture of the target to be captured.

[0118] In specific embodiments, the leader-follower model is optimized to obtain the optimal leader-follower model based on the pigeon thinking in step S4, specifically

[0119] Step S5.1: In the original pigeon thinking optimized map and compass operator, the unmanned ship or UAV is defined as a pigeon individual in the pigeon population, the random speed and position of each pigeon individual are set, and the fitness value of each pigeon is compared to obtain the current global optimal solution;

[0120] The initial speed and position of each pigeon individual are updated, and the fitness value comparison of the pigeon individual is performed to obtain the latest global optimal solution;

[0121] The initial position and speed of pigeon i are defined as and

[0122] And the initial position and speed state update formula of pigeon i is:

[0123]

[0124] x i (t+1) = x i (t) + v i (t+1)

[0125] where i = 1, 2, 3…N p ; N p is the population size; D and R are the size of search space, map and compass factors, respectively; represents the current global best solution; X i denotes the position of pigeon i; denotes the distance between pigeon i and pigeon 1; denotes the distance between pigeon i and pigeon d; denotes the speed difference between pigeon i and pigeon 1; denotes the speed difference between pigeon i and pigeon d; v i (t+1) denotes the speed of pigeon i at t+1; x i (t) denotes the position of pigeon i at t; e -Rt denotes a limiting function of speed; x i (t+1) denotes the position of pigeon i at t+1; V i (t) denotes the speed of pigeon i at t;

[0126] Step S5.2: In order to evaluate the ability of development and exploration, an energy estimation function is proposed Through the energy estimation function The leader in the group is selected, and the formula is expressed as

[0127]

[0128] wherein, represents the energy of the leader affecting the leader attribute; represents the energy value of the group search performance; denotes the energy estimation function;

[0129] The energy values of each pigeon individual in the pigeon group are arranged in descending order, and the pigeon individuals greater than the preset leader energy threshold are selected as the candidate leader set;

[0130] The energy of the leader affecting the leader attribute The calculation formula is

[0131]

[0132]

[0133] wherein, T0 is the number of cycles, Fit() represents the fitness value; represents the estimated value of the position of the leader at time t; represents the estimated value of the position of the leader at time t+1; represents the estimated value of the global optimal solution of the central position of the leader at time t; represents the estimated value of the central position at time t;

[0134] The energy value of the group search performance The calculation formula is

[0135]

[0136] Wherein, Np is the individual scale in pigeon thinking; represents the estimated value of the position of pigeon i at time t; represents the estimated value of the position of the i-th pigeon at time t+1; When the leader has enough power to lead the group, it can be guaranteed that And

[0137] Step S5.3: All individuals of the pigeon group are energy adapted according to the fitness value Fit(), and are randomly divided into N d sub-intervals, and the candidate leader set is defined as

[0138] X cand ={X cand1 , X cand2 ,..., X candNc}

[0139] Wherein, Nc represents the number of UAV or unmanned ship candidates; A individual is randomly selected from the set X cand as a new leader;

[0140] According to the new leader, the center X center (t) of the pigeon group is updated to the sub-interval center X centeri (t), and the position update formula of the sub-interval center X centeri (t) is:

[0141] X i (t+1) = X i (t) + rand(X center (t) - X i (t))

[0142]

[0143] Wherein, X j(t) represents the position of pigeon j; i and j represent pigeons with different targets; rand() and f() represent functions; S d (Xi), These are the individual's neighborhood set and the number of neighbors, respectively.

[0144] Step S5.4: Based on the behavior control of the invalid space, through the Jacobian matrix J ng To avoid collisions between individual drones or unmanned surface vessels, the Jacobian matrix J... ng for

[0145]

[0146] Among them, P ns =[x ns y ns ] = P nr ;x ns The x-axis represents the detector's position relative to the ground; the y-axis represents its position in the x-direction. ns The position of the detector relative to the ground in the y-direction; x ng The x-axis represents the position of the unmanned vessel relative to the ground; the y-axis represents its position. ng P represents the y-axis position of the unmanned surface vessel relative to the ground. ng P represents the vector form of the unmanned surface vessel's position. ns P is in vector form representing the detector range; nr Represents the velocity feedback gain matrix;

[0147] Step S5.5 Based on the Jacobian matrix J ng The total control over the task of gaining encirclement is:

[0148]

[0149]

[0150] in, Indicates the overall control over the encirclement; This indicates the repulsive force between individuals; Indicates zero-space behavior control; u ng κ3 represents the acceleration of the unmanned vessel; κ3 is the gain factor. The potential function of the detector is determined by the selected detector; κ2 is the minimum distance of the detector; κ2 is the gain factor. P represents the maximum distance of the detector. ng P represents the vector form of the unmanned surface vessel's position. ns The vector form of the detector range; x ns X represents the detector's position relative to the ground in the x-direction; ng The x-axis represents the position of the unmanned vessel relative to the ground; the y-axis represents its position.ng The unmanned vessel's position relative to the ground in the y-direction; y ns Let y be the position of the detector relative to the ground.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for heterogeneous collaborative capture of unmanned aerial vehicles (UAVs) and unmanned surface vessels based on pigeon-like thinking optimization, characterized in that... Includes the following steps: Step S1: Establish mathematical models for drones and unmanned vessels; Step S2: Using mathematical models of UAVs and unmanned surface vessels (USVs), initially define the leader and follower roles for both UAVs and USVs, and design the leader-follower model for UAVs and USVs based on a binary tree algorithm, specifically as follows: Step S2.1: Set up a hybrid multi-layer distributed system, which includes several drones and several unmanned vessels; The drone and unmanned vessel establish a communication topology based on a binary tree algorithm, and set an initial leader and follower for the drone and unmanned vessel. The drone follower will detect the maritime targets to be captured in the detection area and send the location information of the maritime targets to be captured to the drone leader. Step S2.2: The drone leader generates the final location information of the target at sea based on the location information of the target to be captured at sea sent by the drone follower and the relative distance between the drone leader and the drone follower. The drone leader sends the final location information of the target to be captured at sea to the unmanned vessel leader; Step S2.3: The unmanned vessel leader sends the final location information of the target to be captured at sea to the unmanned vessel follower, so that the unmanned vessel leader and follower provide capture data for the target to be captured at sea based on the final location information of the target to be captured at sea; The drone leader and drone followers, and the unmanned vessel leader and unmanned vessel followers, use two-way microwave communication; and the drone leader and unmanned vessel leader have a communication link. Step S3: Based on the leader-follower model of the UAV and unmanned surface vessel (USV), design the encirclement control law for the USV. The encirclement control law for the USV is specifically as follows: in, and They are respectively and The expected capture radius and the expected carrying space; and The number of the unmanned vessel; It means the radius that changes over time; Indicates any unmanned boat during the encirclement. speed; Indicates the presence of any unmanned vessel during the encirclement. Adjacent unmanned boats speed; This refers to the radius during the actual encirclement and capture. ; Let x be the x-coordinate of the unmanned vessel relative to the ground. The x-coordinate of the target relative to the ground; Let y be the y-coordinate of the unmanned vessel relative to the ground; y-coordinate; The real-time speed for capturing unmanned vessels; The real-time speed of the target being surrounded; Step S4: Optimize the leader-follower model of drones and unmanned ships based on pigeon thinking to obtain the optimal leader-follower model; Step S5: Obtain the location information of the target to be captured at sea by detecting the leader-follower model of the optimal UAV; The location information is the relative horizontal distance between the target to be captured at sea and the leader of the drone; the leader of the drone sends the location information to the leader of the unmanned vessel, and the leader of the unmanned vessel sends the location information to the unmanned vessel followers; The unmanned vessel leader and followers use the unmanned vessel's encirclement control law to encircle and capture targets at sea.

2. The method for heterogeneous collaborative capture of unmanned aerial vehicles and unmanned surface vessels based on pigeon-like thinking optimization as described in claim 1, characterized in that, The mathematical model of the UAV established in step S1 is as follows: in, This indicates the speed of the drone in the x-direction; This indicates the speed of the drone in the y-direction; Indicates drone Velocity in the direction; x-direction, y-direction and The directions are all manually set based on the geographic coordinate system; This indicates the angular velocity of the drone's heading. Represents the drone's real-time acceleration; Indicates the number of drones, , , The position of the drone relative to the ground. For the ground speed of the drone, The first derivative representing the ground velocity of the drone; Let g be the acceleration of the drone, and g be the acceleration due to gravity. The heading angle of the drone. For the roll angle of the drone, The rate of change of altitude of the drone, , , , and Represents the state variables and control variables of the drone; and Represents the speed constraint of the drone. and These represent the drone's maximum acceleration and roll angle, respectively. The mathematical model of the unmanned vessel in question is as follows: in, This represents the x-direction derivative of the unmanned surface vessel. ; This indicates the real-time speed of the unmanned vessel; , , and These represent the unmanned vessel's position, speed, and heading, respectively. This represents the acceleration of the unmanned vessel. and This indicates the speed constraint for the unmanned vessel.

3. The method for heterogeneous collaborative capture of unmanned aerial vehicles and unmanned surface vessels based on pigeon-like thinking optimization according to claim 2, characterized in that, The detection area of ​​the drone is obtained by collecting the coverage area of ​​the detector, and the calculation formula is as follows: In the formula: Indicates the coverage area of ​​the data acquisition detector; This refers to the drone's position relative to the ground. This indicates the maximum value of the drone's built-in line-of-sight parameters.

4. The method for heterogeneous collaborative encirclement and capture of unmanned aerial vehicles and unmanned vessels based on pigeon-like thinking optimization as described in claim 3, characterized in that, The relative distance between the drone and the target to be captured at sea is calculated using the following formula: in, Indicates the relative distance between the drone and the target to be captured at sea; Indicates the horizontal distance between the drone and the tracked target; The selected drone's ID number; Number all drone models.

5. The method for heterogeneous collaborative capture of unmanned aerial vehicles and unmanned surface vessels based on pigeon-like thinking optimization according to claim 1, characterized in that, Step S4, which involves optimizing the leader-follower model based on pigeon-like thinking to obtain the optimal leader-follower model, specifically involves... Step S5.1: In the original pigeon-like thinking optimization map and compass operator, the unmanned boat or drone is defined as a pigeon individual in the pigeon population. The random speed and position of each pigeon individual are set, and the fitness value of each pigeon is compared to obtain the current global best solution. The initial speed and position of each pigeon individual are updated, and the fitness value of each pigeon individual is compared to obtain the latest global best solution. Define pigeon The initial position and velocity are and ; And the pigeons mentioned The initial position and velocity state update formulas are as follows: in =1,2,3… ; It refers to population size; and These are the size of the search space, map, and compass factors; This represents the current global best solution; Represents pigeon Location; Represents pigeon Distance to pigeon 1 Represents pigeon With pigeons The distance between them; Represents pigeon The speed difference with pigeon 1; Represents pigeon With pigeons Speed ​​difference; Represents pigeon exist The speed of time; Represents the pigeon at time t Location; A limiting function representing speed; This represents the pigeon at time t+1. Location; Represents the pigeon at time t speed; Step S5.2: Using the energy estimation function The formula for selecting leaders in a group is expressed as follows: in, The energy of a leader that influences their attributes; Energy values ​​representing the performance of group search; Represents the energy estimation function; The energy values ​​of each pigeon in the pigeon group are sorted in descending order, and pigeons with energy values ​​greater than the preset leader energy threshold are selected as the candidate leader set. The calculation formula is in, =0, 1, 2, 3, ..., T0; T0 is the number of cycles. Represents fitness value; This represents the estimated position of the leader at time t; This represents the estimated position of the leader at time t+1; This represents an estimate of the globally optimal solution representing the leader's central position at time t. This represents the estimated position of the center at time t; The energy value of the group search performance The calculation formula is in, For the individual scale in pigeon-like thinking; Represents the pigeon at time t The estimated location; Represents the time t+1. An estimate of the location of the pigeon; Step S5.3: All individuals in the pigeon flock are ranked according to their fitness values. Perform energy adaptation and randomly divide into The set of candidate leaders is defined as a set of sub-intervals. in, Indicates the number of candidates for drones or unmanned vessels; set One individual is randomly selected from among them to serve as the new leader; The center of the pigeon colony is updated according to the new leader. Change to sub-interval center The center of the sub-interval The position update formula is: in, Indicates the position of pigeon j; Pigeons representing different objectives; , Represents a function; , These are the individual's neighborhood set and the number of neighbors, respectively. Step S5.4: Based on the behavior control of the invalid space, through the Jacobian matrix... To avoid collisions between individual drones or unmanned vessels, the Jacobian matrix... for in, ; This represents the detector's position relative to the ground in the x-direction. This represents the detector's position relative to the ground in the y-direction. The unmanned vessel's position relative to the ground in the x-direction; The position of the unmanned vessel relative to the ground in the y-direction; This represents the vector form of the unmanned vessel's position; It is a vector form representing the detector range; Represents the velocity feedback gain matrix; Step S5.5 Based on the Jacobian matrix The total control over the task of gaining encirclement is: in, Indicates the overall control over the encirclement; This indicates the repulsive force between individuals; Indicates zero-space behavior control; This indicates the acceleration of the unmanned vessel; It is the gain factor; The potential function of the detector is determined by the selected detector; This is the minimum distance for the detector; It is the gain factor; This is the maximum distance of the detector; This represents the vector form of the unmanned vessel's position; It is a vector form representing the detector range; This represents the detector's position relative to the ground in the x-direction. The unmanned vessel's position relative to the ground in the x-direction; The position of the unmanned vessel relative to the ground in the y-direction; Let y be the position of the detector relative to the ground.

Images