A collaborative target encirclement control method for multiple unmanned vehicles based on control obstacle function

Abstract

The present invention discloses a method for collaborative target encirclement control of multiple unmanned boats based on a control obstacle function. The method comprises: establishing a collaborative target encirclement control model of a multi-unmanned boat-target system model; when a multi-unmanned boat network encircles a moving target in real time, the position of the moving target acquired and estimated by the multiple unmanned boats is input into the model, the model outputs the forward and steering power of the unmanned boat to control the unmanned boat in real time, and the unmanned boat outputs the lateral velocity and angular velocity to the model to realize closed-loop control, thereby realizing collaborative target encirclement control of multiple unmanned boats. The method of the present invention can estimate the global position and velocity information of the target by the unmanned boat, realize collaborative target encirclement control when the position and velocity of the target are unknown, and when the unmanned boat is in the presence of model uncertainty and external interference, the unmanned boat can also accurately track the expected speed in real time, while considering safety design to ensure that the unmanned boats do not collide with each other and obstacles, and ultimately improve the overall control performance.

Description

Technical Field

[0001] The present invention relates to a method for cooperative target encirclement control of multiple unmanned boats, and in particular to a method for cooperative target encirclement control of multiple unmanned boats based on a control obstacle function. Background Art

[0002] In recent years, to adapt to increasingly complex maritime mission conditions, multi-UAV systems have gained increasing attention in various scenarios due to their integration of swarm intelligence distributed collaboration technology, strong environmental adaptability, and the autonomy and high portability of single UAVs. The field of collaborative control has also seen increasing development. In many marine operational scenarios, such as maritime rescue, maritime counter-terrorism, and resource exploration, multiple UAV systems are often required to perform tasks such as interception and multi-directional acquisition of target information. This requires the ability to surround targets. Therefore, multi-UAV target surround collaborative control technology has been widely studied in recent years as an important supporting technology in the field of UAV collaboration.

[0003] However, existing research in the field of cooperative control of unmanned aerial vehicles (UAVs) primarily focuses on collaborative path tracking and collaborative target tracking. These efforts do not address the specific requirements of target encirclement tasks and are therefore not directly applicable to such tasks. While some research has been conducted in the field of mobile robotics for target encirclement tasks, these studies often require the target to be stationary or have known velocity and position information. However, in practical applications, the target's velocity is often unknown. Furthermore, in the UAV operating environment, considering the nonlinear hydrodynamic conditions and the influence of disturbances, the safety of the formation becomes a significant challenge, including avoiding collisions between adjacent UAVs and between UAVs and external obstacles. Despite the importance of this issue, little research has been conducted on this issue. Summary of the Invention

[0004] In order to solve the problems existing in the background technology, the present invention provides a multi-unmanned vehicle cooperative target encirclement control method based on the control obstacle function. The method of the present invention solves the problem of cooperative safe encirclement control of a formation of unmanned vehicles when the target position information is locally unknown and the speed information is globally unknown.

[0005] The technical solution adopted in the present invention is:

[0006] The multi-unmanned vehicle cooperative target encirclement control method based on the control obstacle function of the present invention includes:

[0007] 1) The main unmanned boat and several auxiliary unmanned boats are constructed into a multi-unmanned boat network, and a multi-unmanned boat-target system model is established in the scenario where the multi-unmanned boat network collaboratively surrounds a moving target. Based on the multi-unmanned boat-target system model, a collaborative target encirclement control model is constructed, including a target collaborative controller, a kinematic control law, a forward thrust controller based on a radial basis function neural network, a target optimization model based on a control obstacle function, a torque controller based on a radial basis function neural network, and a radial basis function neural network adaptive law.

[0008] 2) When a multi-unmanned boat network surrounds a moving target in real time, the actual position of the moving target obtained by the main unmanned boat of the multi-unmanned boat network and the estimated position of the moving target estimated by each slave unmanned boat are input into the target collaborative controller of the collaborative target encirclement control model. The target collaborative controller outputs the estimated position and estimated speed of the moving target of each unmanned boat to the kinematic control law. The kinematic control law outputs the expected speed of each unmanned boat to the forward thrust controller based on the radial basis function neural network. The forward thrust controller based on the radial basis function neural network outputs the estimated forward power of each unmanned boat to the target optimization model based on the control obstacle function. The target optimization model based on the control obstacle function outputs the optimized value of the forward power of each unmanned boat and obtains the expected angular velocity and forward power of each unmanned boat after coordinate system transformation. Force control input, the power control input of the forward power of each unmanned boat is input into the multi-unmanned boat network to control each unmanned boat in real time, the expected angular velocity of each unmanned boat is input into the torque controller based on the radial basis neural network, and the torque controller based on the radial basis neural network outputs the power control input of the steering power of each unmanned boat to the multi-unmanned boat network to control each unmanned boat in real time. Under external interference, each unmanned boat outputs its own lateral speed and heading angular velocity to the radial basis neural network adaptive law, and the radial basis neural network adaptive law outputs the updated forward and lateral weight estimation matrices and steering weight estimation values ​​to the forward thrust controller and torque controller based on the radial basis neural network to realize closed loop, until the unmanned boat network completely surrounds the moving target, realizing multi-unmanned boat collaborative target encirclement control.

[0009] In step 1), the multi-UAV-target system model in the scenario where the multi-UAV network collaboratively surrounds the moving target includes a target kinematic model and a multi-UAV dynamic model, specifically as follows:

[0010] a) Target kinematic model:

[0011]

[0012] p t =[x t ,y t ] T

[0013] v t =[v t,x ,v t,y ] T

[0014] Among them, p t and Represent the position of the moving target and its derivative, p t ∈R 2 , x t and y t They represent the horizontal and vertical coordinates of the moving target in the Earth-centered Earth-fixed coordinate system respectively; v t Indicates the speed of the moving target, v t ∈R 2 , v t,x and v t,y They represent the velocity components of the moving object in the X and Y directions in the Earth-centered Earth-fixed coordinate system.

[0015] b) Dynamic model of multiple unmanned vehicles:

[0016]

[0017] p i =[x i ,y i ] T

[0018] v i =[u i ,v i ] T

[0019]

[0020]

[0021]

[0022] R(Ф i )=[cos(Ф i ),sin(Ф i )); -sin(Ф i ),cos(Ф i )]

[0023] f i (v i ,r i )=[f u,i (v i ,r i )f v,i (v i ,r i)] T

[0024]

[0025] Γ(v i )=[1 / m u,i ,v i ;0,-u i ]

[0026]

[0027]

[0028] Among them, p i and Represent the position and derivative of the i-th unmanned boat, p i ∈R 2 , i=1,2,…,N, N represents the number of unmanned boat formations in the unmanned boat network, x i and y i are the horizontal and vertical coordinates of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system; v i and Represent the speed and derivative of the i-th unmanned boat, v i ∈R 2 ,u i and v i Respectively represent the forward speed and lateral speed of the i-th unmanned boat; Ф i and denote the heading angle and its derivative of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system, r i and They represent the angular velocity and its derivative of the i-th unmanned boat respectively; Indicates that the i-th unmanned boat is moving at speed v i and the angular velocity r i The nonlinear hydrodynamic effect on the represents the forward power of the i-th unmanned boat; represents the unknown but bounded external disturbance to the i-th unmanned boat; R(Ф i ) represents the heading angle Φ of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system i The rotation matrix under i (v i ,r i ) indicates that the i-th unmanned boat is moving at a speed of v i and the angular velocity r i The unknown nonlinear hydrodynamic force under u,i (v i ,r i ),fv,i (v i ,r i ) and f r,i (v i ,r i ) represent the speed of the i-th unmanned boat v i and the angular velocity r i The unknown nonlinear hydrodynamic force in the forward velocity direction, the lateral velocity direction and the angular velocity direction; Γ(v i ) indicates that the i-th unmanned boat is moving at a speed of v i The velocity parameter matrix under τ u,i and τ r,i They represent the power control inputs of the forward power and steering power of the i-th unmanned boat respectively; m u,i represents the additional mass of the i-th unmanned boat; d e,u d e,v and d e,r They represent the unknown but bounded external disturbances that the unmanned boat experiences in the forward velocity direction, lateral velocity direction, and angular velocity direction respectively; m r,i represents the additional moment of inertia of the i-th unmanned boat.

[0029] In step 1), the target collaborative controller is as follows:

[0030]

[0031]

[0032] in, and They represent the estimated value of the position of the mobile target by the i-th unmanned boat and its derivative respectively; γ1 and γ2 represent the preset first constant and the preset second constant respectively greater than 0; a ij represents the communication status between the i-th unmanned boat and the j-th unmanned boat, a ij = 1, the i-th unmanned boat and the j-th unmanned boat communicate with each other, a ij = 0, the i-th unmanned boat and the j-th unmanned boat do not communicate with each other; κ i represents the communication status between the i-th unmanned boat and the mobile target, κ i = 1, the i-th unmanned boat and the moving target communicate with each other and obtain the position information of the moving target, κ i = 0, the i-th unmanned boat and the moving target do not communicate with each other and do not obtain the position information of the moving target; t Indicates the position of the moving target; and They represent the estimated value of the velocity of the moving target by the i-th unmanned boat and its derivative respectively.

[0033] In step 1), the kinematic control law is as follows:

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040] σ d,i =2πi / N

[0041] Among them, v d,i represents the expected speed of the i-th unmanned boat; and Respectively represent the estimated values ​​of the position and speed of the moving target by the i-th unmanned boat; ι1 and ι2 respectively represent the preset third constant and the preset fourth constant greater than 0; η d It represents the expected distance of the mobile target surrounded by the multi-UAV network, which is a constant greater than 0; ρ ti Represents the estimated distance between the i-th unmanned boat and the moving target is the desired enclosing angular velocity; Λ is the preset constant matrix; is the collaborative item of the i-th unmanned boat; p i represents the position of the i-th unmanned boat; and They represent the i-th unmanned boat and the j-th unmanned boat respectively, i is not equal to the central angle of the arc formed by j and its respective adjacent unmanned boats, β i represents the direction angle between the i-th unmanned boat and the target, σ d,i represents the phase of the desired heading angle of the i-th unmanned boat; x t and y t They represent the horizontal and vertical coordinates of the moving target in the Earth-centered Earth-fixed coordinate system respectively; i and y i They represent the horizontal and vertical coordinates of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system respectively; N represents the number of unmanned boat formations in the unmanned boat network.

[0042] In the step 1), the forward thrust controller based on the radial basis function neural network is specifically as follows:

[0043]

[0044]

[0045]

[0046]

[0047] e v,i =v d,i -v i

[0048]

[0049]

[0050] in, represents the forward power of the i-th unmanned boat; v d,i and They represent the expected speed v of the i-th unmanned boat respectively d,i and its derivatives; W v,i and represents the forward and lateral weight matrix of the i-th unmanned boat and its forward and lateral weight estimation matrix. The subscript v in the forward and lateral weight estimation matrix represents the forward and lateral directions. W x and W y Represents the forward and lateral weight matrices W of the i-th unmanned boat respectively v,i The first and second neural network weights, W x ∈R l×1 , W y ∈R l×1 , l is the number of neurons; h v,i (ξ i ) represents the neural network input matrix ξ of the i-th unmanned boat i The radial basis function matrix, h x,i (ξ x,i ) and h y,i (ξ y,i ) represent the neural network input matrix ξ in the radial basis function matrix of the i-th unmanned boat. i The neural network input component ξ in the forward velocity direction x,i The first radial basis function and the neural network input component ξ in the lateral velocity direction y,i The second radial basis function, h x,i (ξ x,i )∈R l×1 , h y,i (ξ y,i )∈R l×1 ,[ξ x,i ξ y,i ] T ∈R2 ×5 ;e v,i and They represent the expected speed and actual speed v of the i-th unmanned boat respectively. i The velocity tracking error between and its derivative; Indicates the speed v of the i-th unmanned boat estimated by radial basis neural network i and the angular velocity r i The nonlinear hydrodynamic effect estimated value of; represents the unknown but bounded external disturbance to the i-th unmanned boat; ε i represents the first estimation error of the radial basis neural network, ε i ∈R 2×1 ; l2 represents the identity matrix, l2 = [1,1] T ; k1 and k2 represent the fifth and sixth preset constants respectively, which are greater than 0. Real-time update based on the adaptive law of radial basis neural network.

[0051] In step 1), the target optimization model based on the control barrier function is specifically as follows:

[0052]

[0053]

[0054]

[0055]

[0056] in, and They represent the forward power of the i-th unmanned boat and its optimized value respectively; p ij Indicates the position p of the i-th unmanned boat i and the position p of the jth unmanned boat j The relative distance between ij =p i -p j ψ ij represents the nonlinear coupling term between the i-th and j-th unmanned boats, ψ io represents the nonlinear coupling term between the i-th unmanned boat and the obstacle; p io Indicates the position p of the i-th unmanned boat i and the obstacle position p o The relative distance between io =p i -p o ; τ maxrepresents the maximum forward power of the i-th unmanned boat;

[0057]

[0058]

[0059]

[0060]

[0061] Among them, a max Indicates the maximum acceleration of the unmanned boat; v ij represents the speed v of the i-th unmanned boat i and the speed v of the jth unmanned boat j The relative speed between ij =v i -v j ; ε o Indicates the minimum safe distance to be maintained between the unmanned boat and neighboring unmanned boats or external obstacles; B ij represents the control obstacle function that the i-th unmanned boat satisfies in avoiding obstacles with its neighbor j-th unmanned boat; v io represents the speed v of the i-th unmanned boat i and the speed v of the obstacle o The relative distance between io =v i -v o ; B io It represents the control obstacle function that the i-th unmanned boat satisfies in avoiding external obstacles.

[0062] The optimal value of the forward power of the i-th unmanned boat is Perform coordinate system transformation as follows:

[0063]

[0064] Among them, τ u,i represents the power control input of the forward power of the i-th unmanned boat, represents the expected angular velocity of the i-th unmanned boat; Γ(v i ) indicates that the i-th unmanned boat is moving at a speed of v i The velocity parameter matrix under R(Ф i ) represents the heading angle Φ of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system i The rotation matrix below.

[0065] In the step 1), the torque controller based on the radial basis function neural network is specifically as follows:

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] Among them, τ r,i represents the power control input of the steering power of the i-th unmanned boat; and denote the expected angular velocity and its derivative of the i-th unmanned boat, r i represents the actual angular velocity of the i-th unmanned boat, e r,i and are the expected angular velocity of the i-th unmanned boat and the actual angular velocity r i The angular velocity tracking error and its derivative between W r,i and They represent the steering weight of the i-th unmanned boat and its steering weight estimation value, W r,i ∈R l×1 is the neural network weight; h r,i (ξ r,i ) represents the neural network input ξ based on the angular velocity of the i-th unmanned boat. r,i The third radial basis function, h r,i (ξ i )∈R l×1 , k3 and k4 represent the seventh and eighth preset constants respectively, which are greater than 0; sign(e r,i ) represents the expected angular velocity based on the i-th unmanned boat and the actual angular velocity r i The angular velocity tracking error e r,i The symbolic function of f r,i ′(v i ,r i ) represents the speed v of the i-th unmanned boat estimated by radial basis neural network i and the angular velocity r i The unknown nonlinear hydrodynamic force f in the direction of the angular velocity is r,i (v i ,r i )’s estimated value; m r,i represents the additional inertia moment of the i-th unmanned boat; d e,r represents the external disturbance of the unmanned boat in the direction of angular velocity; ε r,i represents the second estimation error of the radial basis neural network, εr,i ∈R 1×1 ; Real-time update based on the adaptive law of radial basis neural network.

[0072] In the step 1), the adaptive law of the radial basis neural network is as follows:

[0073]

[0074]

[0075] in, Represents the updated forward and lateral weight matrix W of the i-th unmanned boat v,i The forward and side weight estimation matrices of γ r and γ v Respectively represent the ninth and tenth preset constants greater than 0; e v,i Indicates the expected speed and actual speed v of the i-th unmanned boat i The speed tracking error between v,i (ξ i )express

[0076] The neural network input matrix ξ of the i-th unmanned boat i The radial basis function matrix of Represents the updated steering weight W of the i-th unmanned boat r,i Steering weight estimate e r,i represents the expected angular velocity of the i-th unmanned boat and the actual angular velocity r i Angular velocity tracking error between r,i (ξ r,i ) represents the neural network input ξ based on the angular velocity of the i-th unmanned boat. r,i The third radial basis function.

[0077] The electronic device of the present invention comprises: a memory and a processor coupled to each other, wherein the memory stores program data, and the processor calls the program data to execute the above-mentioned method.

[0078] The computer-readable storage medium of the present invention stores program data thereon, and when the program data is executed by a processor, the method described above is implemented.

[0079] The present invention designs a target collaborative estimator to estimate the position and velocity of the target object in view of the local unknown position and global unknown velocity of the target object on the water surface. On this basis, the present invention designs a target encirclement collaborative kinematic control law based on the multi-unmanned boat-target system model to ensure that the unmanned boat formation can achieve the encirclement of the target object while keeping the adjacent unmanned boats at equal distances. Taking into account the unknown nonlinearity of the multi-unmanned boat-target system model and the external interference it is subjected to, the present invention designs a forward thrust controller based on the radial basis function neural network, which can achieve compensation for nonlinear hydrodynamics and external interference through online adaptive network weight parameters. In addition, in order to ensure the safety of the unmanned boat network during collaborative encirclement control, that is, to avoid collisions with neighboring unmanned boats and external obstacles, the present invention constructs a target optimization model based on the control obstacle function to optimize the control law and generate a safe unmanned boat control input. Through the above design, the safe multi-unmanned boat collaborative target encirclement control is finally achieved in the presence of uncertain dynamic models and external obstacles.

[0080] The beneficial effects of the present invention are:

[0081] 1. The present invention designs a target state estimator for mobile targets whose position information is locally unknown and whose speed information is globally unknown. Under the condition that some unmanned boats know the global position information of the target, the global position information and global speed information of each unmanned boat relative to the target are estimated without relying on high-cost ranging sensors.

[0082] 2. The present invention designs a multi-UAV target encirclement coordinated motion guidance law based on the estimated target position and velocity, thereby realizing coordinated target encirclement control when the target position and velocity are unknown.

[0083] 3. The present invention uses an unmanned vehicle dynamics controller based on a radial basis neural network and a minimum learning parameter adaptive law to enable the unmanned vehicle to track the desired speed generated by the collaborative motion guidance law in real time and accurately. When the unmanned vehicle is in the presence of model uncertainty and external interference, it can also achieve accurate speed tracking, thereby improving the overall target encirclement motion control performance.

[0084] 4. The present invention takes into account the high complexity of the radial basis neural network adaptive algorithm and designs a minimum learning parameter adaptive law to reduce the number of online adaptive weights to two, thereby greatly reducing the complexity of the neural network algorithm, thereby improving the algorithm operation efficiency and realizing lightweight deployment of the controller.

[0085] 5. The present invention takes into account the safety design during the encirclement control process of multiple unmanned boats, and uses the control obstacle function to optimize the obtained dynamic controller, thereby achieving encirclement control of the target while ensuring that the unmanned boats do not collide with each other and external obstacles. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 This is the target encirclement collaborative control block diagram based on the control obstacle function proposed by the present invention;

[0087] Figure 2 is a simulated trajectory diagram of an unmanned boat and a target object in a simulation experiment according to an embodiment of the present invention;

[0088] Figure 3 : is a schematic diagram of the relative distance and separation angle error of the unmanned boat target in the simulation experiment of the embodiment of the present invention, wherein: Figure 3 (a) is a schematic diagram of the relative distance of the target surrounded by the unmanned boat in the simulation experiment of the embodiment of the present invention, Figure 3 (b) is a schematic diagram of the target encirclement separation angle error of the unmanned vehicle in the simulation experiment of the embodiment of the present invention;

[0089] Figure 4 The collision avoidance of the unmanned boat in the time T = 220s to 300s under the traditional method and the method of the present invention, as well as the corresponding distance and control obstacle function value, where: Figure 4 (a) is a diagram showing the collision avoidance of the unmanned boat in the time T = 220s to 300s under the traditional method and the corresponding distance and control obstacle function value. Figure 4 (b) is a schematic diagram of the situation where the unmanned boat avoids collision within T=220s~300s under the method of the present invention and the corresponding distance and control obstacle function value. DETAILED DESCRIPTION

[0090] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0091] like Figure 1 As shown, the multi-unmanned vehicle cooperative target encirclement control method based on the control obstacle function of the present invention is specifically as follows:

[0092] 1) The main unmanned boat and several auxiliary unmanned boats are constructed into a multi-unmanned boat network, and a multi-unmanned boat-target system model is established in the scenario where the multi-unmanned boat network collaboratively surrounds a moving target. Based on the multi-unmanned boat-target system model, a collaborative target encirclement control model is constructed, including a target collaborative controller, a kinematic control law, a forward thrust controller based on a radial basis function neural network, a target optimization model based on a control obstacle function, a torque controller based on a radial basis function neural network, and a radial basis function neural network adaptive law.

[0093] The multi-UAV-target system model in the scenario where multiple UAV networks collaborate to surround a moving target includes the target kinematic model and the multi-UAV dynamics model, as follows:

[0094] a) Target kinematic model:

[0095]

[0096] p t =[x t ,y t ] T

[0097] v t =[v t,x ,v t,y ] T

[0098] Among them, p t and Represent the position of the moving target and its derivative, p t ∈R 2 , x t and y t They represent the horizontal and vertical coordinates of the moving target in the Earth-centered Earth-fixed coordinate system respectively; v t Indicates the speed of the moving target, v t ∈R 2 , v t,x and v t,y They represent the velocity components of the moving object in the X and Y directions in the Earth-centered Earth-fixed coordinate system.

[0099] b) Dynamic model of multiple unmanned vehicles:

[0100]

[0101] p i =[x i ,y i ] T

[0102] v i =[u i ,v i ] T

[0103]

[0104]

[0105]

[0106] R(Ф i )=[cos(Фi ),sin(Ф i )); -sin(Ф i ),cos(Ф i )]

[0107] f i (v i ,r i )=[f u,i (v i ,r i )f v,i (v i ,r i )] T

[0108]

[0109] Γ(v i )=[1 / m u,i ,v i ;0,-u i ]

[0110]

[0111]

[0112] Among them, p i and Represent the position and derivative of the i-th unmanned boat, p i ∈R 2 , i=1,2,…,N, N represents the number of unmanned boat formations in the unmanned boat network, x i and y i are the horizontal and vertical coordinates of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system; v i and Represent the speed and derivative of the i-th unmanned boat, v i ∈R 2 ,u i and v i Respectively represent the forward speed and lateral speed of the i-th unmanned boat; Ф i and denote the heading angle and its derivative of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system, r i and They represent the angular velocity and its derivative of the i-th unmanned boat respectively; Indicates that the i-th unmanned boat is moving at speed v i and the angular velocity r i The nonlinear hydrodynamic effect on the represents the forward power of the i-th unmanned boat; represents the unknown but bounded external disturbance to the i-th unmanned boat; R(Ф i ) represents the heading angle Φ of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system i The rotation matrix under i (v i ,r i ) indicates that the i-th unmanned boat is moving at a speed of v i and the angular velocity r i The unknown nonlinear hydrodynamic force under u,i (v i ,r i ),f v,i (v i ,r i ) and f r,i (v i ,r i ) represent the speed of the i-th unmanned boat v i and the angular velocity r i The unknown nonlinear hydrodynamic force in the forward velocity direction, the lateral velocity direction and the angular velocity direction; Γ(v i ) indicates that the i-th unmanned boat is moving at a speed of v i The velocity parameter matrix under τ u,i and τ r,i They represent the power control inputs of the forward power and steering power of the i-th unmanned boat respectively; m u,i represents the additional mass of the i-th unmanned boat; d e,u d e,v and d e,r They represent the unknown but bounded external disturbances that the unmanned boat experiences in the forward velocity direction, lateral velocity direction, and angular velocity direction respectively; m r,i represents the additional moment of inertia of the i-th unmanned boat.

[0113] The target collaborative controller is as follows:

[0114]

[0115]

[0116] in, and They represent the estimated value of the position of the mobile target by the i-th unmanned boat and its derivative respectively; γ1 and γ2 represent the preset first constant and the preset second constant respectively greater than 0; a ij represents the communication status between the i-th unmanned boat and the j-th unmanned boat, a ij = 1, the i-th unmanned boat and the j-th unmanned boat communicate with each other, a ij= 0, the i-th unmanned boat and the j-th unmanned boat do not communicate with each other; κ i represents the communication status between the i-th unmanned boat and the mobile target, κ i = 1, the i-th unmanned boat and the moving target communicate with each other and obtain the position information of the moving target, κ i = 0, the i-th unmanned boat and the moving target do not communicate with each other and do not obtain the position information of the moving target; t Indicates the position of the moving target; and They represent the estimated value of the velocity of the moving target by the i-th unmanned boat and its derivative respectively.

[0117] The kinematic control law is as follows:

[0118]

[0119]

[0120]

[0121]

[0122]

[0123]

[0124] σ d,i =2πi / N

[0125] Among them, v d,i represents the expected speed of the i-th unmanned boat; and Respectively represent the estimated values ​​of the position and speed of the moving target by the i-th unmanned boat; ι1 and ι1 respectively represent the preset third constant and the preset fourth constant greater than 0; η d It represents the expected distance of the mobile target surrounded by the multi-UAV network, which is a constant greater than 0; ρ ti Represents the estimated distance between the i-th unmanned boat and the moving target is the desired enclosing angular velocity; Λ is the preset constant matrix; is the collaborative item of the i-th unmanned boat; p i represents the position of the i-th unmanned boat; and They represent the i-th unmanned boat and the j-th unmanned boat respectively, i is not equal to the central angle of the arc formed by j and its respective adjacent unmanned boats, β i represents the direction angle between the i-th unmanned boat and the target, σ d,i represents the phase of the desired heading angle of the i-th unmanned boat; xt and y t They represent the horizontal and vertical coordinates of the moving target in the Earth-centered Earth-fixed coordinate system respectively; i and y i They represent the horizontal and vertical coordinates of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system respectively; N represents the number of unmanned boat formations in the unmanned boat network.

[0126] The forward thrust controller based on radial basis neural network is as follows:

[0127]

[0128]

[0129]

[0130]

[0131] e v,i =v f,i -v i

[0132]

[0133]

[0134] in, represents the forward power of the i-th unmanned boat; v d,i and They represent the expected speed v of the i-th unmanned boat respectively d,i and its derivatives; W v,i and represents the forward and lateral weight matrix of the i-th unmanned boat and its forward and lateral weight estimation matrix. The subscript v in the forward and lateral weight estimation matrix represents the forward and lateral directions. W x and W y Represents the forward and lateral weight matrices W of the i-th unmanned boat respectively v,i The first and second neural network weights, W x ∈R l×1 , W y ∈R l×1 , l is the number of neurons; h v,i (ξ i ) represents the neural network input matrix ξ of the i-th unmanned boat i The radial basis function matrix, h x,i (ξ x,i ) and h y,i (ξ y,i ) represent the neural network input matrix ξ in the radial basis function matrix of the i-th unmanned boat.i The neural network input component ξ in the forward velocity direction x,i The first radial basis function and the neural network input component ξ in the lateral velocity direction y,i The second radial basis function, h x,i (ξ x,i )∈R l×1 , h y,i (ξ y,i )∈R l×1 ,[ξ x,i ξ y,i ] T ∈R 2 ×5 ;e v,i and They represent the expected speed and actual speed v of the i-th unmanned boat respectively. i The velocity tracking error between and its derivative; Indicates the speed v of the i-th unmanned boat estimated by radial basis neural network i and the angular velocity r i The nonlinear hydrodynamic effect estimated value of; represents the unknown but bounded external disturbance to the i-th unmanned boat; ε i represents the first estimation error of the radial basis neural network, ε i ∈R 2×1 ; I2 represents the identity matrix, l2 = [1,1] T ; k1 and k2 represent the fifth and sixth preset constants respectively, which are greater than 0. Real-time update based on the adaptive law of radial basis neural network.

[0135] The target optimization model based on the control obstacle function is as follows:

[0136]

[0137]

[0138]

[0139]

[0140] in, and They represent the forward power of the i-th unmanned boat and its optimized value respectively; p ij Indicates the position p of the i-th unmanned boat i and the position p of the jth unmanned boat j The relative distance between ij =p i -pj ψ ij represents the nonlinear coupling term between the i-th and j-th unmanned boats, ψ io represents the nonlinear coupling term between the i-th unmanned boat and the obstacle; p io Indicates the position p of the i-th unmanned boat i and the obstacle position p o The relative distance between io =p i -p o ; τ max represents the maximum forward power of the i-th unmanned boat;

[0141]

[0142]

[0143]

[0144]

[0145] Among them, a max Indicates the maximum acceleration of the unmanned boat; v ij represents the speed v of the i-th unmanned boat i and the speed v of the jth unmanned boat j The relative speed between ij =v i -v j ; ε o Indicates the minimum safe distance to be maintained between the unmanned boat and neighboring unmanned boats or external obstacles; B ij represents the control obstacle function that the i-th unmanned boat satisfies in avoiding obstacles with its neighbor j-th unmanned boat; v io represents the speed v of the i-th unmanned boat i and the speed v of the obstacle o The relative distance between io =v i -v o ; B io It represents the control obstacle function that the i-th unmanned boat satisfies in avoiding external obstacles.

[0146] The optimal value of the forward power of the i-th unmanned boat is Perform coordinate system transformation as follows:

[0147]

[0148] Among them, τ u,i represents the power control input of the forward power of the i-th unmanned boat, represents the expected angular velocity of the i-th unmanned boat; Γ(vi ) indicates that the i-th unmanned boat is moving at a speed of v i The velocity parameter matrix under R(Ф i ) represents the heading angle Φ of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system i The rotation matrix below.

[0149] The torque controller based on radial basis neural network is as follows:

[0150]

[0151]

[0152]

[0153]

[0154]

[0155] Among them, τ r,i represents the power control input of the steering power of the i-th unmanned boat; and denote the expected angular velocity and its derivative of the i-th unmanned boat, r i represents the actual angular velocity of the i-th unmanned boat, e r,i and are the expected angular velocity of the i-th unmanned boat and the actual angular velocity r i The angular velocity tracking error and its derivative between W r,i and They represent the steering weight of the i-th unmanned boat and its steering weight estimation value, W r,i ∈R l×1 is the neural network weight; h r,i (ξ r,i ) represents the neural network input ξ based on the angular velocity of the i-th unmanned boat. r,i The third radial basis function, h r,i (ξ i )∈R l×1 , k3 and k4 represent the seventh and eighth preset constants respectively, which are greater than 0; sign(e r,i ) represents the expected angular velocity based on the i-th unmanned boat and the actual angular velocity r i The angular velocity tracking error e r,i The symbolic function of f r,i ′(v i ,r i) represents the speed v of the i-th unmanned boat estimated by radial basis neural network i and the angular velocity r i The unknown nonlinear hydrodynamic force f in the direction of the angular velocity is r,i (v i ,r i )’s estimated value; m r,i represents the additional inertia moment of the i-th unmanned boat; d e,r represents the external disturbance of the unmanned boat in the direction of angular velocity; ε r,i represents the second estimation error of the radial basis neural network, ε r,i ∈R 1×1 ; Real-time update based on the adaptive law of radial basis neural network.

[0156] The adaptive law of radial basis neural network is as follows:

[0157]

[0158]

[0159] in, Represents the updated forward and lateral weight matrix W of the i-th unmanned boat v,i The forward and side weight estimation matrices of γ r and γ v Respectively represent the ninth and tenth preset constants greater than 0; e v,i Indicates the expected speed and actual speed v of the i-th unmanned boat i The speed tracking error between v,i (ξ i )express

[0160] The neural network input matrix ξ of the i-th unmanned boat i The radial basis function matrix of Represents the updated steering weight W of the i-th unmanned boat r,i Steering weight estimate e r,i represents the expected angular velocity of the i-th unmanned boat and the actual angular velocity r i Angular velocity tracking error between r,i (ξ r,i ) represents the neural network input ξ based on the angular velocity of the i-th unmanned boat. r,i The third radial basis function.

[0161] 2) When a multi-unmanned boat network surrounds a moving target in real time, the actual position of the moving target obtained by the main unmanned boat of the multi-unmanned boat network and the estimated position of the moving target estimated by each slave unmanned boat are input into the target collaborative controller of the collaborative target encirclement control model. The target collaborative controller outputs the estimated position and estimated speed of the moving target of each unmanned boat to the kinematic control law. The kinematic control law outputs the expected speed of each unmanned boat to the forward thrust controller based on the radial basis function neural network. The forward thrust controller based on the radial basis function neural network outputs the estimated forward power of each unmanned boat to the target optimization model based on the control obstacle function. The target optimization model based on the control obstacle function outputs the optimized value of the forward power of each unmanned boat and obtains the expected angular velocity and forward power of each unmanned boat after coordinate system transformation. Force control input, the power control input of the forward power of each unmanned boat is input into the multi-unmanned boat network to control each unmanned boat in real time, the expected angular velocity of each unmanned boat is input into the torque controller based on the radial basis neural network, and the torque controller based on the radial basis neural network outputs the power control input of the steering power of each unmanned boat to the multi-unmanned boat network to control each unmanned boat in real time. Under external interference, each unmanned boat outputs its own lateral speed and heading angular velocity to the radial basis neural network adaptive law, and the radial basis neural network adaptive law outputs the updated forward and lateral weight estimation matrices and steering weight estimation values ​​to the forward thrust controller and torque controller based on the radial basis neural network to realize closed loop, until the unmanned boat network completely surrounds the moving target, realizing multi-unmanned boat collaborative target encirclement control.

[0162] Finally, Matlab / Simulink simulation is performed on the control method of the present invention to verify the effectiveness of the multi-unmanned vehicle dynamic target encirclement collaborative control method based on the control obstacle function proposed in the present invention.

[0163] During verification, N=3, that is, there are 3 unmanned boats in total, and the target speed is set as follows:

[0164] v t =[u t cos(Ф t ),v t sin(Ф t )] T

[0165] u t =0.2m / s

[0166]

[0167] Ф t (0) = 0 rad / s

[0168] Among them, Ф t and They represent the heading angle and angular velocity of the moving object in the Earth-centered Earth-fixed coordinate system respectively.

[0169] The unknown nonlinear coupling force caused by the hydrodynamic force is set as follows:

[0170]

[0171]

[0172]

[0173] The additional mass of the unmanned boat is set to m u,i = 25.8 kg, the additional moment of inertia is set to m r,i =2.76kg·m 2 , the external interference is set to d e,u d e,v and d e,r They represent the external disturbances of the unmanned boat in the forward velocity, lateral velocity and angular velocity directions, Indicates that the t The rotation matrix, w i represents the external disturbance in the Earth-centered Earth-fixed coordinate system, Ω=diag(0.5, 0.5, 0.5), K=diag(1,1,0.5), γ i ∈R 3 represents Gaussian white noise with a mean of 0 and a variance of 4. The initial positions and initial heading angles of the three unmanned boats are set to: [-4m, -4m, 0rad] T ,[-6m,2m,0rad] T ,[-2m,-6m,0rad] T , the initial speed is 0m / s; the initial position of the target is set to [0m,0m] T The unmanned boat can only obtain the relative orientation angle β of the target i , the expected encirclement distance is set as η d =10m, the desired enveloping angular velocity is set to ω d =0.3rad / s, expected separation angle vector σ d Set to:

[0174]

[0175] In the specific implementation process, the simulation time is set to 400s, and an obstacle is set at the position [42m, 60m]. The simulation results are as follows Figure 2 shown.

[0176] Figure 2The figure shows the trajectory of the three unmanned boats and the target unmanned boat under the action of this algorithm in the above embodiment. The dotted line represents the trajectory of the target object, the other three solid lines represent the trajectory of the three unmanned boats, and the dotted circle represents the boundary of the external obstacle. Figure 3 (a) and Figure 3 As shown in (b), the relative distance and relative separation angle errors of each unmanned boat relative to the target unmanned boat in this process are shown. It can be observed that the unmanned boats consistently surround the target unmanned boat and continuously maintain equidistant angular intervals. The corresponding relative distance and relative separation angle errors gradually converge to the expected distance interval of 10m and the interval of 0 from the initial stage. Figure 4 As shown in the figure, the time period T = 220s to 300s is displayed, and the unmanned boat encounters external obstacles, such as Figure 4 As shown in (a), under the traditional obstacle avoidance method that does not consider the use of control obstacle function, the unmanned boat collided with an external obstacle at T=241s and T=4.29s, and collided with internal unmanned boats. The corresponding relative distances between the obstacle and the neighboring unmanned boat were less than the safe distance at these moments. At the same time, the control obstacle function was also less than 0, proving that the collision avoidance failed. This is because the control obstacle function constraint of the safety control filter in the traditional algorithm is only based on the unmanned boat dynamics design fitted by the radial basis neural network, but does not fully consider the network estimation error and external disturbances. When the combined uncertainty caused by the network estimation error and external disturbance is large, the control command obtained based on this control obstacle function constraint will not be able to ensure that the actual unmanned boat's control obstacle function also meets the constraint, thereby increasing the risk of collision and failing to ensure the safety of the entire process. The safety control filter in the method of the present invention takes into account the combined uncertainty caused by the network estimation error and external disturbance. Therefore, it can be obtained from Figure 4 It can be observed in (b) that the unmanned boat successfully avoids collisions during these periods of high collision risk. The corresponding relative distances to obstacles and neighboring unmanned boats are always greater than the safe distance at these moments, and the control obstacle function is also greater than 0, which proves that the safety control filter in the method of the present invention can ensure the safety of the unmanned boat throughout the entire process and achieve the control goal.

[0177] The above content is only the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function, characterized in that: include: Step 1) A main unmanned boat and several auxiliary unmanned boats are constructed into a multi-unmanned boat network, and a multi-unmanned boat-target system model is established in a scenario where the multi-unmanned boat network collaboratively surrounds a moving target object. Based on the multi-unmanned boat-target system model, a collaborative target encirclement control model is constructed, including a target collaborative controller, a kinematic control law, a forward thrust controller based on a radial basis function neural network, a target optimization model based on a control obstacle function, a torque controller based on a radial basis function neural network, and a radial basis function neural network adaptive law; Step 2) When the multi-unmanned boat network surrounds the moving target in real time, the actual position of the moving target obtained by the main unmanned boat of the multi-unmanned boat network and the estimated position of the moving target estimated by each slave unmanned boat are input into the target collaborative controller of the collaborative target encirclement control model. The target collaborative controller outputs the estimated position and estimated speed of each unmanned boat for the moving target to the kinematic control law. The kinematic control law outputs the expected speed of each unmanned boat to the forward thrust controller based on the radial basis function neural network. The forward thrust controller based on the radial basis function neural network outputs the estimated forward power of each unmanned boat to the target optimization model based on the control obstacle function. The target optimization model based on the control obstacle function outputs the optimized value of the forward power of each unmanned boat and obtains the expected angular velocity and forward power of each unmanned boat after coordinate system transformation. Power control input, the power control input of the forward power of each unmanned boat is input into the multi-unmanned boat network to control each unmanned boat in real time, the expected angular velocity of each unmanned boat is input into the torque controller based on the radial basis neural network, the torque controller based on the radial basis neural network outputs the power control input of the steering power of each unmanned boat to the multi-unmanned boat network to control each unmanned boat in real time, each unmanned boat outputs its own lateral speed and heading angular velocity to the radial basis neural network adaptive law under external interference, the radial basis neural network adaptive law outputs the updated forward and lateral weight estimation matrix and steering weight estimation value to the forward thrust controller and torque controller based on the radial basis neural network to realize closed loop, until the unmanned boat network completely surrounds the moving target, realizing multi-unmanned boat collaborative target encirclement control.

2. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In step 1), the multi-UAV-target system model in the scenario where the multi-UAV network collaboratively surrounds the moving target includes a target kinematic model and a multi-UAV dynamic model, specifically as follows: a) Target kinematic model: p t =[x t ,y t ] T v t =[v t,x ,v t,y ] T Among them, p t and Represent the position of the moving target and its derivative, x t and y t They represent the horizontal and vertical coordinates of the moving target in the Earth-centered Earth-fixed coordinate system respectively; v t Indicates the speed of the moving target, v t,x and v t,y They represent the velocity components of the moving target in the X and Y directions in the Earth-centered Earth-fixed coordinate system respectively; b) Dynamic model of multiple unmanned vehicles: p i =[x i ,y i ] T v i =[u i ,v i ] T R(F i )=[cos(F i ),sin(F i );-sin(F i ),cos(F i )] f i (v i ,r i )=[f u,i (v i ,r i )f v,i (v i ,r i )] T Γ(v i )=[1 / m u,i ,v i ;0,-u i ] Among them, p i and They represent the position of the i-th unmanned boat and its derivative, i = 1, 2, ..., N, N represents the number of unmanned boat formations in the unmanned boat network, x i and y i are the horizontal and vertical coordinates of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system; V i and They represent the speed and derivative of the i-th unmanned boat, u i and v i Respectively represent the forward speed and lateral speed of the i-th unmanned boat; Ф i and denote the heading angle and its derivative of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system, r i and They represent the angular velocity and its derivative of the i-th unmanned boat respectively; Indicates that the i-th unmanned boat is moving at speed v i and the angular velocity r i The nonlinear hydrodynamic effect on the represents the forward power of the i-th unmanned boat; represents the unknown but bounded external disturbance to the i-th unmanned boat; R(Ф i ) represents the heading angle Φ of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system i The rotation matrix under i (v i ,r i ) indicates that the i-th unmanned boat is moving at a speed of V i and the angular velocity r i The unknown nonlinear hydrodynamic force under u,i (v i ,r i ),f v,i (v i ,r i ) and f r,i (v i ,r i ) represent the speed of the i-th unmanned boat v i and the angular velocity r i The unknown nonlinear hydrodynamic force in the forward velocity direction, the lateral velocity direction and the angular velocity direction; Γ(v i ) indicates that the i-th unmanned boat is moving at a speed of v i The velocity parameter matrix under τ u,i and τ r,i They represent the power control inputs of the forward power and steering power of the i-th unmanned boat respectively; m u,i represents the additional mass of the i-th unmanned boat; d e,u d e,v and d e,r They represent the unknown but bounded external disturbances that the unmanned boat experiences in the forward velocity direction, lateral velocity direction, and angular velocity direction respectively; m r,i represents the additional moment of inertia of the i-th unmanned boat.

3. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In step 1), the target collaborative controller is as follows: in, and They represent the estimated value of the position of the mobile target by the i-th unmanned boat and its derivative respectively; γ1 and γ2 represent the preset first constant and the preset second constant respectively greater than 0; a ij represents the communication status between the i-th unmanned boat and the j-th unmanned boat, a ij = 1, the i-th unmanned boat and the j-th unmanned boat communicate with each other, a ij = 0, the i-th unmanned boat and the i-th unmanned boat do not communicate with each other; κ i represents the communication status between the i-th unmanned boat and the mobile target, κ i = 1, the i-th unmanned boat and the moving target communicate with each other and obtain the position information of the moving target, κ i = 0, the i-th unmanned boat and the moving target do not communicate with each other and do not obtain the position information of the moving target; t Indicates the position of the moving target; and They represent the estimated value of the velocity of the moving target by the i-th unmanned boat and its derivative respectively.

4. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In step 1), the kinematic control law is as follows: v d,i =2πi / N Among them, v d,i represents the expected speed of the i-th unmanned boat; and Respectively represent the estimated values ​​of the position and speed of the moving target by the i-th unmanned boat; ι1 and ι2 respectively represent the preset third constant and the preset fourth constant greater than 0; η d represents the expected distance of the mobile target surrounded by the multi-UAV network; ρ ti Represents the estimated distance between the i-th unmanned boat and the moving target is the desired enclosing angular velocity; Λ is the preset constant matrix; is the collaborative item of the i-th unmanned boat; p i represents the position of the i-th unmanned boat; and They represent the central angles of the arcs formed by the i-th unmanned boat and the j-th unmanned boat and their respective adjacent unmanned boats, β i represents the direction angle between the i-th unmanned boat and the target, σ d,i represents the phase of the desired heading angle of the i-th unmanned boat; x t and y t They represent the horizontal and vertical coordinates of the moving target in the Earth-centered Earth-fixed coordinate system respectively; i and y i They represent the horizontal and vertical coordinates of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system respectively; N represents the number of unmanned boat formations in the unmanned boat network.

5. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In the step 1), the forward thrust controller based on the radial basis function neural network is specifically as follows: yes v,i =v d,i -v i in, represents the forward power of the i-th unmanned boat; v d,i and They represent the expected speed v of the i-th unmanned boat respectively d,ii and its derivatives; W v,i and represents the forward and lateral weight matrix of the i-th unmanned boat and its forward and lateral weight estimation matrix, W x and W y Represents the forward and lateral weight matrices W of the i-th unmanned boat respectively v,i The first and second neural network weights; h v,i (ξ i ) represents the neural network input matrix ξ of the i-th unmanned boat i The radial basis function matrix, h x,i (ξ x,i ) and h y,i (ξ y,i ) represent the neural network input matrix ξ in the radial basis function matrix of the i-th unmanned boat. i The neural network input component ξ in the forward velocity direction x,i The first radial basis function and the neural network input component ξ in the lateral velocity direction y,i The second radial basis function of e v,i and They represent the expected speed and actual speed v of the i-th unmanned boat respectively. i The velocity tracking error between and its derivative; Indicates the speed v of the i-th unmanned boat estimated by radial basis neural network i and the angular velocity r i The nonlinear hydrodynamic effect estimated value of; represents the unknown but bounded external disturbance to the i-th unmanned boat; ε i Represents the first estimation error of the radial basis neural network.

6. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In step 1), the target optimization model based on the control barrier function is specifically as follows: in, and They represent the forward power of the i-th unmanned boat and its optimized value respectively; p ij Indicates the position p of the i-th unmanned boat i and the position p of the jth unmanned boat j The relative distance between ij =p i -p j ψ ij represents the nonlinear coupling term between the i-th and j-th unmanned boats, ψ io represents the nonlinear coupling term between the i-th unmanned boat and the obstacle; p io Indicates the position p of the i-th unmanned boat i and the obstacle position p o The relative distance between io =p i -p o ; τ max represents the maximum forward power of the i-th unmanned boat; The optimal value of the forward power of the i-th unmanned boat is Perform coordinate system transformation as follows: Among them, τ u,i represents the power control input of the forward power of the i-th unmanned boat, represents the expected angular velocity of the i-th unmanned boat; Γ(v i ) indicates that the i-th unmanned boat is moving at a speed of v i The velocity parameter matrix under R(Ф i ) represents the heading angle Φ of the i-th unmanned boat in the Earth-centered Earth-fixed coordinate system i The rotation matrix below.

7. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In the step 1), the torque controller based on the radial basis function neural network is specifically as follows: Among them, τ r,i represents the power control input of the steering power of the i-th unmanned boat; and denote the expected angular velocity and its derivative of the i-th unmanned boat, r i represents the actual angular velocity of the i-th unmanned boat, e r,i and are the expected angular velocity of the i-th unmanned boat and the actual angular velocity r i The angular velocity tracking error and its derivative between W r,i and They represent the steering weight of the i-th unmanned boat and its steering weight estimation value respectively; h r,i (ξ r,i ) represents the neural network input ξ based on the angular velocity of the i-th unmanned boat. r,i The third radial basis function; k3 and k4 represent the seventh preset constant and the eighth preset constant respectively greater than 0; sign(e r,i ) represents the expected angular velocity based on the i-th unmanned boat and the actual angular velocity r i The angular velocity tracking error e r,i The symbolic function of f r,i ′(v i ,r i ) represents the speed v of the i-th unmanned boat estimated by radial basis neural network i and the angular velocity r i The unknown nonlinear hydrodynamic force f in the direction of the angular velocity is r,i (v i ,r i )’s estimated value; m r,i represents the additional inertia moment of the i-th unmanned boat; d e,r represents the external disturbance of the unmanned boat in the direction of angular velocity; ε r,i Represents the second estimation error of the radial basis neural network.

8. The method for cooperative target encirclement control of multiple unmanned vehicles based on a control obstacle function according to claim 1, characterized in that: In the step 1), the adaptive law of the radial basis neural network is as follows: in, Represents the updated forward and lateral weight matrix W of the i-th unmanned boat v,i The forward and side weight estimation matrices of γ r and γ v Respectively represent the ninth and tenth preset constants greater than 0; e v,i Indicates the expected speed and actual speed v of the i-th unmanned boat i The speed tracking error between v,i (ξ i ) represents the neural network input matrix ξ of the i-th unmanned boat i The radial basis function matrix of Represents the updated steering weight W of the i-th unmanned boat r,i Steering weight estimate e r,i represents the expected angular velocity of the i-th unmanned boat and the actual angular velocity r i Angular velocity tracking error between r,i (ξ r,i ) represents the neural network input ξ based on the angular velocity of the i-th unmanned boat. r,i The third radial basis function.

9. An electronic device, characterized in that: include: A memory and a processor coupled to each other, wherein the memory stores program data, and the processor calls the program data to execute the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having program data stored thereon, characterized in that: When the program data is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images