A multi-to-one cooperative pursuit controller design method for unmanned surface vehicle in obstacle environment

Abstract

The application discloses a design method of a multi-to-one cooperative pursuit controller for unmanned ships in an obstacle environment, and comprises the following steps: considering unknown system dynamics and complex environmental disturbances, constructing kinematic equations and dynamic equations of a tracking unmanned ship and kinematic equations of an escaping unmanned ship; based on the position of the escaping unmanned ship, realizing estimation of the position and speed of the escaping unmanned ship by multiple virtual pursuers; based on a surrounding algorithm, an approaching algorithm and a trade-off algorithm, designing a distributed cooperative pursuit strategy of the multiple virtual pursuers for the escaping unmanned ship; based on the distributed cooperative pursuit strategy, obtaining pursuit speed of the virtual pursuers; constructing a control obstacle function to generate an optimal trajectory generator to avoid collision in the cooperative pursuit process; based on a line-of-sight guidance law, designing an extended state observer according to a speed signal given by an unmanned ship system; based on the line-of-sight guidance law and the extended state observer, designing a model-free control law to realize pursuit control of multiple unmanned ships for one escaping unmanned ship.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ship and ocean engineering, and can be widely applied in military and civilian fields, mainly studies the problem of multi-unmanned ship cooperative pursuit under unknown model and obstacle environment, and proposes a design method of multi-to-one cooperative pursuit controller of unmanned ship under obstacle environment. BACKGROUND

[0002] In recent years, with the wide application of unmanned ships in military and civilian fields, researchers have put forward new demands for the improvement of control ability and the expansion of application of unmanned ships, such as cooperative tracking, interception and expulsion, target encirclement, etc. The application of these demands aims to effectively pursue target objects and provide more reliable support for military operations, criminal investigation and other fields. Among them, pursuit control is one of the key technologies in the pursuit and escape application of unmanned ships, which involves accurate tracking of escapees and multi-ship cooperative control tasks.

[0003] In the past few years, some researchers have conducted in-depth research on the pursuit problem, however, most of the current pursuit research is based on first-order system design, only considering the kinematics of the controlled target and ignoring the dynamics. With the continuous expansion of the application of unmanned ships in the marine field, multi-unmanned ship pursuit technology has become an important research direction. In the process of multi-unmanned ship cooperative pursuit, the cooperative pursuit control strategy with collision avoidance capability not only needs to consider the collision between individuals, but also needs to consider the obstacles in the environment, therefore, the collision avoidance problem in multi-ship cooperative pursuit is a challenging problem; at the same time, in the actual navigation environment, due to the limitation of communication distance and bandwidth, the acquisition of escapee information is also a problem that puzzles researchers, because the escapee may adopt strategies such as concealment, speed change, direction change, etc., which makes the trajectory prediction and tracking more difficult; in addition, due to the uncertainty of the system model and the complexity and randomness of the marine environment, the unmanned ship also needs to face the influence of complex disturbance factors such as system dynamics unknown and wind and current in the pursuit process, which further increases the difficulty of pursuing the escape unmanned ship. Therefore, it is still a problem worth studying to realize safe and fast pursuit control in complex marine environment.

[0004] In summary, the existing technology has the following shortcomings:

[0005] First, most of the existing cooperative pursuit strategies are based on first-order system design, only considering the kinematics of the controlled target and ignoring the dynamics, which may lead to the inability to guarantee the stability of the pursuit control in some cases. In addition, most of the existing cooperative pursuit strategies assume that the information of the escape unmanned ship is globally known, however, in a vast sea area, it is often difficult to achieve global knowledge of the escape unmanned ship information.

[0006] Second, existing cooperative pursuit strategies usually only focus on the feasibility of pursuing the escaping unmanned surface vehicle (USV), ignoring the static and dynamic obstacles faced by the tracking USV during the pursuit process and the collision risk of the neighbor tracking USV, which may lead to the safety of the multi-USV system being unable to be guaranteed during the pursuit process.

[0007] Third, existing model-based controller design needs to rely on the knowledge of model dynamics parameters, which may bring certain limitations to the design and adjustment of the controller, making it difficult for the controller to adapt to different scenarios and task requirements. SUMMARY

[0008] To solve the above problems, the technical scheme adopted by the present application is: a multi-to-one cooperative pursuit controller design method for unmanned surface vehicles in obstacle environments, comprising the following steps:

[0009] Considering unknown system dynamics and complex environmental disturbances, the kinematic equation of the tracking USV is constructed;

[0010] Based on the position of the escaping USV, the position of the virtual pursuer is estimated.

[0011] Based on the surrounding algorithm, the proximity algorithm and the trade-off algorithm, a distributed cooperative pursuit strategy for multiple virtual pursuers is designed.

[0012] Based on the distributed cooperative pursuit strategy, the pursuit speed of the virtual pursuer is obtained.

[0013] A control obstacle function is constructed to generate an optimal trajectory generator to avoid collision during cooperative pursuit.

[0014] Define the longitudinal trajectory tracking error and the lateral trajectory tracking error, and design a line-of-sight guidance law.

[0015] Based on the line-of-sight guidance law, the speed signal given by the USV system is designed to design an extended state observer.

[0016] Based on the line-of-sight guidance law and the extended state observer, a model-free control law is designed to realize the pursuit control of multiple USVs on one escaping USV.

[0017] Further, the position of the virtual pursuer is estimated based on the position of the escaping USV, and a distributed target observer is used to estimate the position and speed of the escaping USV.

[0018]

[0019] wherein, represents the estimation of the position of the i-th virtual pursuer on the escaping USV p e . represents the position of the i-th virtual pursuer to the escaping UAV e derivative of the estimate; represents the position of the j-th virtual pursuer to the escaping UAV e estimate; represents the velocity of the i-th virtual pursuer to the escaping UAV e estimate; represents the velocity of the i-th virtual pursuer to the escaping UAV e derivative of the estimate; K1 is a positive definite coefficient matrix; K2 is a positive definite coefficient matrix; c is a normal number; a ij is a communication topology parameter of the multiple virtual pursuers cooperation; b i is a detection structure parameter of the virtual pursuer and the escaping UAV.

[0020] Further, the process of designing the distributed cooperative pursuit strategy of the multiple virtual pursuers to the escaping UAV based on the surrounding algorithm, the proximity algorithm and the trade-off algorithm is as follows:

[0021] The virtual pursuit signal of the i-th pursuit UAV is defined as:

[0022] and

[0023] wherein, represents the position of the i-th virtual pursuer; represents the velocity of the i-th virtual pursuer; represents the maximum velocity value of the i-th virtual pursuer;

[0024] Based on the state information of the escaping UAV estimated by the distributed target observer, the relative distance p i and the relative angle b i between the i-th virtual pursuer and the escaping UAV are defined as:

[0025]

[0026] The Apollonian circle method is introduced in the pursuit strategy, and the occupation angle q i is defined as the angle between the tangent line l1 and the tangent line l2, which is expressed by the sine rule as follows:

[0027] q i = 2arcsin(m i ) (7)

[0028] The virtual pursuer group is dispersed counterclockwise, and the coverage angle a i of the i-th virtual pursuer and the adjacent (i+1)-th virtual pursuer is defined as:

[0029]

[0030] where β (M+) = β1; θ (M+) = θ1; ζ i = 2π when i = M; ζ i = 0 when i ≠ M; α i < 0 means that both the ith virtual pursuer and the (i + 1)th virtual pursuer track the escaping UAV; otherwise, the escaping UAV can escape;

[0031] The occupation angle of all virtual pursuers is denoted as:

[0032]

[0033] where θ G is the total occupation angle of the encircling task.

[0034] The proposed distributed cooperative pursuit strategy includes an encircling task and an approaching task. The encircling task is to increase the total occupation angle θ G , and the encircling strategy is designed as:

[0035] λ is = k i ρ i (α i - α (i-) ) (10)

[0036] where α0= α M when i = 0; k i > 0 is the encircling parameter; (α i - α (i-1) ) > 0 means that the ith virtual pursuer encircles the escaping UAV counterclockwise; the approaching task is to shorten the distance to the escaping UAV, and the approaching strategy is designed as

[0037] λ ih = -h i ρ i (11)

[0038] where h i > 0 is the approaching parameter.

[0039] The weights of the encircling strategy and the approaching strategy are determined by the trade-off method, and the trade-off coefficient is designed. It is assumed that all virtual pursuers move at the maximum speed ;

[0040] From equation (9), we have and where η i ∈ [0, π / 2] is a trade-off parameter, and its specific form is as follows:

[0041]

[0042] where κ i ∈ [0, 1]; κ i = 2 |α i - α (i-) | / 4π - θ (i+) + θ (i-) is the enclosing weight; ι i ∈ [0, 1], is the approaching weight, and ρ i + ρ (i-) + ρ (i+) ≠ 0.

[0043] Further, the process of constructing the control barrier function to generate the optimal trajectory generator is as follows:

[0044] For static obstacles, consider the safety set Design the control barrier function as Construct the safety constraint condition of the control barrier function as where and is a positive parameter;

[0045] For dynamic obstacles, consider the safety set Design the control barrier function as Construct the safety constraint condition of the control barrier function as where and is a positive parameter;

[0046] For neighbor virtual pursuers, consider the safety set Design the control barrier function as Construct the safety constraint condition of the control barrier function as where and is a positive parameter;

[0047] The optimal trajectory generator based on the control barrier function is designed as follows:

[0048]

[0049] s.t.

[0050]

[0051]

[0052] where

[0053] Further, the definition of longitudinal trajectory tracking error and lateral trajectory tracking error, the process of designing the line-of-sight guidance law is as follows:

[0054] The longitudinal trajectory tracking error x is defined as iz and the lateral trajectory tracking error y is defined as iz As follows:

[0055]

[0056] Wherein, is the tangent angle, i.e. the velocity direction of the planned trajectory.

[0057] The line-of-sight guidance law is designed as follows:

[0058]

[0059] Wherein, is the velocity of the trajectory generator; is the look-ahead distance; and is a positive parameter.

[0060] Further, the extended state observer is designed as follows:

[0061]

[0062] Wherein, are the extended states of the observer; is the estimated value of u i ; is the estimated value of r i ; is the estimated value of σ iu ; is the estimated value of σ ir ; is the derivative of the estimated value of u i ; is the derivative of the estimated value of r i ; is the derivative of the estimated value of σ iu ; is the derivative of the estimated value of σ ir ; are the observer gain coefficients designed.

[0063] Further, the model-free control law is designed as follows:

[0064]

[0065] Wherein, γ iu and γ ir are control parameters; k iu and kir is a regulation parameter.

[0066] A cooperative pursuit controller for unmanned surface vehicle in obstacle environment, comprising a distributed target observer, a cooperative pursuit trajectory planning, a nominal trajectory generator, an optimal trajectory generator based on control barrier function, a line-of-sight guidance law module, an extended state observer and a model-free control law;

[0067] The input end of the distributed target observer is connected with the position signal of the given escape unmanned surface vehicle in the communication network, and the output end of the distributed target observer is connected with the cooperative pursuit trajectory planning;

[0068] The input end of the cooperative pursuit trajectory planning is connected with the i-th virtual pursuer given in the communication network, the coverage angle signal of the adjacent i+1-th virtual pursuer and the distributed target observer, and the output end of the cooperative pursuit trajectory planning is connected with the nominal trajectory generator;

[0069] The input end of the nominal trajectory generator is connected with the cooperative pursuit trajectory planning, and the output end of the nominal trajectory generator is connected with the trajectory generator based on control barrier function;

[0070] The input end of the optimal trajectory generator based on control barrier function is connected with the nominal trajectory generator, and the output end of the optimal trajectory generator based on control barrier function is connected with the line-of-sight guidance law module;

[0071] The input end of the line-of-sight guidance law module is connected with the optimal trajectory generator based on control barrier function and the state signal given by the unmanned surface vehicle system, and the output end of the line-of-sight guidance law module is connected with the model-free control law;

[0072] The input end of the extended state observer is connected with the speed signal given by the unmanned surface vehicle system, and the output end of the extended state observer is connected with the model-free control law;

[0073] The input end of the model-free control law is connected with the extended state observer and the line-of-sight guidance law, and the output end of the model-free control law is connected with the unmanned surface vehicle system.

[0074] The application provides a multi-to-one cooperative pursuit controller design method for unmanned surface vehicles in an obstacle environment, and the multi-unmanned surface vehicle cooperative pursuit problem containing unknown system dynamics, complex environmental disturbances and escape unmanned surface vehicle information uncertainty is researched, a distributed cooperative pursuit control method based on control barrier function is provided, and the problem that unmanned surface vehicles effectively avoid collision and cooperatively pursue escape unmanned surface vehicles in a complex marine environment is solved.

[0075] Compared with the prior art, the application has the following beneficial effects:

[0076] Firstly, compared with the existing cooperative pursuit strategy for first-order integrator, the proposed method not only considers the dynamics and kinematics characteristics of the unmanned ship, but also considers the local information known to the escaping unmanned ship. By using the known local information, the direction and path of the pursuit can be determined more effectively, thereby improving the pursuit speed and effect of the cooperative pursuit task.

[0077] Secondly, compared with the existing cooperative pursuit strategy that does not consider collision avoidance, the proposed method can avoid collision with static obstacles, dynamic obstacles, and adjacent pursuers in the environment by designing different control barrier functions, thereby improving the safety of the cooperative pursuit task.

[0078] Thirdly, compared with the existing model-based controller, the proposed method designs a model-free control law without any prior model parameters, which can better adapt to changes in different environments and ship characteristics, thereby improving the robustness of the cooperative pursuit task.

[0079] Through the above steps, a complete multi-pursuit unmanned ship-single-escape unmanned ship pursuit system is established, and effective pursuit tasks can be performed in complex environments. BRIEF DESCRIPTION OF DRAWINGS

[0080] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0081] Figure 1 Figure 1 is a diagram for the design of unmanned ship pursuit control method in obstacle environment;

[0082] Figure 2 Figure 2 is a pursuit trajectory curve diagram;

[0083] Figure 3 Figure 3 is a relative distance curve diagram between the pursuit unmanned ship and the escaping unmanned ship;

[0084] Figure 4 Figure 4 is a relative distance curve diagram between the pursuit unmanned ship and the obstacle;

[0085] Figure 5 Figure 5 is a position estimation error curve diagram of the distributed target observer;

[0086] Figure 6 Figure 6 is a speed estimation error curve diagram of the distributed target observer;

[0087] Figure 7 Figure 7 is a longitudinal trajectory tracking error curve diagram;

[0088] Figure 8 is a lateral trajectory tracking error plot;

[0089] Figure 9 is a total surge disturbance estimate plot;

[0090] Figure 10 is a total yaw disturbance estimate plot;

[0091] Figure 11 is a surge system control input plot;

[0092] Figure 12 is a yaw system control input plot. DETAILED DESCRIPTION

[0093] It should be noted that the embodiments and features of the embodiments in the present application can be combined with each other without conflict, and the present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0094] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. The description of the at least one exemplary embodiment is actually only illustrative, and is by no means intended to limit any of the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0095] It should be noted that the terms used herein are only intended to describe specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and furthermore, it should be understood that when the terms "comprise" and / or "include" are used in the specification, there is a feature, step, operation, device, component and / or combination thereof.

[0096] The foregoing is considered as illustrative only of the principles of the application. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the application to the exact construction and practice described. Accordingly, all suitable modifications and equivalents can be resorted to falling within the scope of the application. Unless otherwise indicated herein, the contents of all patents, patent applications, publications, and test methods cited herein are hereby incorporated by reference in their entirety for all purposes.

[0097] In the description of the present application, it is to be understood that the orientation or positional relationships indicated by terms such as "front", "back", "up", "down", "left", "right", "lateral", "vertical", "horizontal", "top", "bottom", and the like are generally based on the orientation or positional relationships shown in the drawings, and are merely intended to facilitate the description and simplify the description, and do not indicate or imply that the device or element referred to must have a particular orientation or be constructed and operated in a particular orientation, and therefore cannot be construed as limiting the scope of protection of the present application. The orientation terms "inner", "outer" refer to the inner and outer relative to the contour of the components themselves.

[0098] For the convenience of description, spatial relative terms such as "over", "above", "upper surface", "upper", and the like can be used herein to describe the spatial positional relationship of one device or feature with respect to other devices or features as shown in the drawings. It should be understood that the spatial relative terms are intended to include different orientations in use or operation in addition to the orientation of the device as described in the drawings. For example, if the device in the drawings is inverted, the device described as "above" or "over" other devices or structures will be positioned "below" or "under" the other devices or structures. Thus, the exemplary term "above" can include both "above" and "below" orientations. The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein are interpreted accordingly.

[0099] In addition, it should be noted that the use of the terms "first", "second", and the like do not have a special meaning, and are merely used to distinguish the corresponding components, and therefore cannot be construed as limiting the scope of protection of the present application.

[0100] The invention relates to a kind of obstacle environment under the design of unmanned ship cooperative pursuit method as shown in Figure 1 According to the observed escape unmanned ship information, a cooperative pursuit strategy is proposed, and a control barrier function is introduced to ensure the safety of the tracking unmanned ship. Based on the principle of line-of-sight guidance, a trajectory tracking guidance law is proposed, and an extended state observer is used to estimate the unknown system dynamics and external disturbances. Based on the estimated dynamic information, a model-free control law is designed to track the guidance signals in the forward and yaw directions.

[0101] Consider a multi-unmanned ship pursuit game system consisting of M tracking unmanned ships and an escaping unmanned ship, and construct an unmanned ship model with unknown system dynamics and complex environmental disturbances. Based on the communication topology of the virtual pursuer cooperation, a distributed target observer is designed to estimate the position and velocity information of the escaping unmanned ship. The virtual pursuit guidance signal of the ith tracking unmanned ship is defined, and the relative distance ρ i and the relative angle β i between the ith virtual pursuer and the escaping unmanned ship are obtained according to the estimated information of the target observer. A cooperative pursuit strategy is proposed based on the relative distance and angle, and the pursuit task is divided into encirclement task and approach task. At the same time, in order to more efficiently pursue the freely evading escaping unmanned ship, a trade-off coefficient is designed to achieve faster capture speed. In order to ensure the safety of the cooperative pursuit task, a collision avoidance control barrier function is designed for static obstacles, dynamic obstacles and adjacent virtual pursuers to achieve safe collision avoidance during the pursuit process. A kinematic guidance law based on the principle of line-of-sight guidance is designed to guide the ith tracking unmanned ship to track the planned trajectory. An extended state observer is used to estimate the unknown system dynamics and external disturbances, and a model-free control law is designed to track the guidance signals in the forward and yaw directions based on the estimated dynamic information.

[0102] A method for designing a multi-to-one cooperative pursuit controller for unmanned ships in an obstacle environment, comprising the following steps:

[0103] Consider the unknown system dynamics and complex environmental disturbances, and construct the kinematic equation of the tracking unmanned ship;

[0104] Based on the position of the escaping unmanned ship, estimate the position and velocity of the escaping unmanned ship by multiple virtual pursuers;

[0105] Based on the surrounding algorithm, the approach algorithm and the trade-off algorithm, design a distributed cooperative pursuit strategy for the escaping unmanned ship by multiple virtual pursuers;

[0106] Based on the distributed cooperative pursuit strategy, obtain the pursuit speed of the virtual pursuer;

[0107] Construct a control barrier function to generate an optimal trajectory generator to avoid collision during the cooperative pursuit process;

[0108] Define longitudinal trajectory tracking error and lateral trajectory tracking error, design the line-of-sight guidance law;

[0109] Based on the line-of-sight guidance law, the given velocity signal of the unmanned surface vehicle system, design the extended state observer;

[0110] Based on the line-of-sight guidance law and the extended state observer, design the model-free control law to realize the multi-to-one cooperative pursuit control of multiple unmanned surface vehicles to an escaping unmanned surface vehicle.

[0111] A method for designing a multi-to-one cooperative pursuit controller of unmanned surface vehicles in an obstacle environment includes the following steps:

[0112] Considering the unknown system dynamics and complex environmental disturbances, the kinematic equation of the ith tracking unmanned surface vehicle is constructed as

[0113]

[0114] Where, i = 1,...,M; (x i ,y i ) represents the position of the ith tracking unmanned surface vehicle in the earth coordinate system; represents the derivative of the position of the ith tracking unmanned surface vehicle in the earth coordinate system; ψ i represents the heading angle of the ith tracking unmanned surface vehicle in the earth coordinate system; represents the derivative of the heading angle of the ith tracking unmanned surface vehicle in the earth coordinate system; u i represents the surge velocity of the ith tracking unmanned surface vehicle in the body coordinate system; v i represents the lateral velocity of the ith tracking unmanned surface vehicle in the body coordinate system, r i represents the yaw angular velocity of the ith tracking unmanned surface vehicle in the body coordinate system.

[0115] The dynamic equation is

[0116]

[0117] Where, represents the derivative of the surge velocity of the ith tracking unmanned surface vehicle in the body coordinate system; represents the derivative of the lateral velocity of the ith tracking unmanned surface vehicle in the body coordinate system; represents the derivative of the yaw angular velocity of the ith tracking unmanned surface vehicle in the body coordinate system; τ iu represents the control input in the surge direction; τ ir represents the control input in the yaw direction; m iu is the inertia coefficient in the surge direction; m iv is the inertia coefficient in the lateral direction; m ir is the inertia coefficient in the yaw direction; f iu(·) represents an unknown nonlinear function in surge direction; f iv (·) represents an unknown nonlinear function in sway direction; f ir (·) represents an unknown nonlinear function in yaw direction; τ iwu represents environmental disturbance in surge direction; τ iwv represents environmental disturbance in sway direction; τ iwr represents environmental disturbance in yaw direction.

[0118] The kinematic equation of the escape USV is constructed as

[0119]

[0120] where (x e ,y e ) represents the real position of the escape USV in the earth coordinate system; represents the derivative of the real position of the escape USV in the earth coordinate system; ψ e represents the heading angle of the escape USV in the earth coordinate system; represents the derivative of the heading angle of the escape USV in the earth coordinate system; u e represents the surge velocity of the escape USV in the body coordinate system; v e represents the sway velocity of the escape USV in the body coordinate system; r e represents the yaw angular velocity of the escape USV in the body coordinate system.

[0121] Further, the distributed target observer is designed as follows:

[0122]

[0123] where, represents the estimation of the position p e of the escape USV by the i-th virtual pursuer; represents the derivative of the estimation of the position p e of the escape USV by the i-th virtual pursuer; represents the estimation of the position p e of the escape USV by the j-th virtual pursuer; represents the estimation of the velocity u e of the escape USV by the i-th virtual pursuer; represents the derivative of the estimation of the velocity u e of the escape USV by the i-th virtual pursuer; K1 is a positive definite coefficient matrix; K2 is a positive definite coefficient matrix; c is a normal number; a ij is a communication topology structure parameter of the multiple virtual pursuers; b i is a detection structure parameter of the virtual pursuer and the escape USV.

[0124] Further, the cooperative pursuit trajectory planning includes a surrounding algorithm, an approaching algorithm, and a trade-off algorithm.

[0125] To realize the cooperative pursuit strategy of multiple USVs, the virtual pursuit signal of the ith USV is defined as

[0126] and where, represents the position of the ith virtual pursuer; represents the speed of the ith virtual pursuer; represents the maximum speed value of the ith virtual pursuer.

[0127] Based on the state information of the escaping USV estimated by the distributed target observer in step a, the relative distance p between the ith virtual pursuer and the escaping USV is defined as i and the relative angle b is i :

[0128]

[0129] To pursue the escaping USV with a faster speed, the Apollonian circle method is introduced into the pursuit strategy. The occupation angle q i defined as the angle between the tangent line l1 and the tangent line l2, can be expressed by the sine rule as follows:

[0130] q i = 2 arcsin (m i ) (7)

[0131] The virtual pursuer group is dispersed counterclockwise, and the coverage angle a i between the ith virtual pursuer and the adjacent (i+1)th virtual pursuer is defined as:

[0132]

[0133] where, b (M+) = b1; q (M+) = q1; when i = M, z i = 2p; when i ≠ M, z i = 0. a i < 0 indicates that both the ith virtual pursuer and the (i+1)th virtual pursuer can pursue the escaping USV; otherwise, the escaping USV can escape.

[0134] The occupation angle of all virtual pursuers can be expressed as

[0135]

[0136] where, q Gto encircle the target.

[0137] The proposed distributed cooperative pursuit strategy includes encircling task and approaching task. The encircling task is to increase the total occupancy angle θ G , the encircling strategy is designed as

[0138] λ is = k i ρ i (α i - α (i-) ) (10)

[0139] where α 0 = α M when i = 0; k i > 0 is the encircling parameter; (α i - α (i-1) ) > 0 means that the i th virtual pursuer encircles the escaping UAV counterclockwise.

[0140] The approaching task is to shorten the distance to the escaping UAV, and the approaching strategy is designed as:

[0141] λ ih = - h i ρ i (11)

[0142] where h i > 0 is the approaching parameter.

[0143] The role of the approaching algorithm is to shorten the distance to the escaping UAV;

[0144] In addition, in order to more efficiently pursue the escaping UAV that freely evades, the present application adopts a trade-off method to determine the weight of the encircling strategy and the approaching strategy, and designs a trade-off coefficient to realize a faster capture speed. It is assumed that all virtual pursuers move at the maximum speed . From equation (9), we can know that and where η i ∈ [0, π / 2] is a trade-off parameter, and its specific form is as follows:

[0145]

[0146] where κ i ∈ [0, 1]; κ i = 2 | α i - α (i-) | / 4 π - θ (i+) + θ (i-) is the encircling weight; ι i ∈ [0, 1], is the approaching weight, and ρ i + ρ (i-) + ρ(i+) ≠ 0.

[0147] Based on the above distributed cooperative pursuit strategy design, the pursuit speed of the ith virtual pursuer can be obtained as

[0148] u ic = u is + u ih (13)

[0149] where u is = λ is [-sin β i , cos β i ] T is the encircling speed of the ith virtual pursuer; u ih = λ ih [cos β i , sin β i ] T is the approaching speed of the ith virtual pursuer.

[0150] Combining the above pursuit speed, the pursuit trajectory of the ith virtual pursuer can be obtained according to

[0151] Further, the design process of the optimal trajectory generator based on the control barrier function is as follows:

[0152] In order to ensure the safety in the process of distributed cooperative pursuit, the following control barrier function is constructed to realize collision avoidance in the process of cooperative pursuit.

[0153] For static obstacles, the safety set is considered as The control barrier function is designed as The safety constraint condition of the control barrier function is constructed as where and is a positive parameter.

[0154] For dynamic obstacles, the safety set is considered as The control barrier function is designed as The safety constraint condition of the control barrier function is constructed as where and is a positive parameter.

[0155] For neighbor virtual pursuers, the safety set is considered as The control barrier function is designed as The safety constraint condition of the control barrier function is constructed as where and is a positive parameter.

[0156] ​In summary, the optimal trajectory generator based on control barrier function is designed as follows:

[0157]

[0158] where,

[0159] Define the longitudinal trajectory tracking error x iz and the lateral trajectory tracking error y iz as follows:

[0160]

[0161] where, is the tangent angle, i.e., the velocity direction of the planned trajectory.

[0162] The line-of-sight guidance law is designed as follows:

[0163]

[0164] where, is the velocity of the trajectory generator; is the look-ahead distance; and is a positive parameter.

[0165] The extended state observer is designed as follows:

[0166]

[0167] where, are the extended states of the observer; is the estimated value of u i ; is the estimated value of r i ; is the estimated value of σ iu ; is the estimated value of σ ir ; is the derivative of the estimated value of u i ; is the derivative of the estimated value of r i ; is the derivative of the estimated value of σ iu ; is the derivative of the estimated value of σ ir ; are the observer gain coefficients designed.

[0168] The model-free control law is designed as follows:

[0169]

[0170] where, γiu and γ ir is a control parameter; k iu and k ir is a tuning parameter.

[0171] The simulation results are shown in Figures 2-12 Fig. 6. Figure 2 Fig. 6 is a curve diagram of the pursuit trajectory, Figure 3 Fig. 6 is a curve diagram of the relative distance between the pursuit USV and the escape USV, it can be seen that at 45 seconds, the No. 3 pursuit USV successfully captures the escape USV, and the relative distance between them is less than the capture radius; Figure 4 Fig. 6 is a curve diagram of the relative distance between the pursuit USV and the obstacle, it can be seen that the distance between the pursuit USV and the static obstacle, the dynamic obstacle, and the neighbor pursuit USV is always greater than the limit obstacle avoidance distance, and no collision occurs; Figure 5 Fig. 6 is a curve diagram of the position estimation error of the distributed target observer, it can be seen that the position estimation error of each virtual pursuer can converge to the domain of the origin, that is, each virtual pursuer can accurately estimate the position information of the escape USV; Figure 6 Fig. 6 is a curve diagram of the velocity estimation error of the distributed target observer, it can be seen that the velocity estimation error of each virtual pursuer can converge to the domain of the origin, that is, each virtual pursuer can accurately estimate the velocity information of the escape USV; Figure 7 Fig. 6 is a curve diagram of the longitudinal trajectory tracking error, it can be seen that the longitudinal trajectory tracking error of each pursuit USV can converge to the domain of the origin; Figure 8 Fig. 6 is a curve diagram of the lateral trajectory tracking error, it can be seen that the lateral trajectory tracking error of each pursuit USV can converge to the domain of the origin, and through Figure 7 and Figure 8 it can be seen that each pursuit USV can accurately track the trajectory of the trajectory generator; Figure 9 Fig. 6 is a curve diagram of the total disturbance estimation in the surge direction of the system, it can be seen that the extended state observer can accurately estimate the total disturbance in the surge direction of the system; Figure 10 Fig. 6 is a curve diagram of the total disturbance estimation in the yaw direction of the system, it can be seen that the extended state observer can accurately estimate the total disturbance in the yaw direction of the system; Figure 11 Fig. 6 is a curve diagram of the control input in the surge direction of the system, it can be seen that the torque signal of the pursuit USV in the surge direction is bounded; Figure 12 Fig. 6 is a curve diagram of the control input in the yaw direction of the system, it can be seen that the torque signal of the pursuit USV in the yaw direction is bounded.

[0172] It can be seen from the above simulation result diagram that the distributed cooperative pursuit control method based on the control barrier function proposed in the present application is a complete multi-pursuit USV-single-escape USV pursuit system, and can effectively perform the pursuit task in a complex environment.

[0173] The present application is not limited to the embodiments, and any equivalent concept or change within the technical scope disclosed in the present application is included in the protection scope of the present application.

[0174] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A design method for a multi-to-one cooperative pursuit controller for unmanned surface vessels in obstacle environments, characterized in that: The method comprises the following steps: considering unknown system dynamics and complex environmental disturbances, constructing kinematic equations and dynamic equations of the tracking unmanned surface vehicle and the escaping unmanned surface vehicle; based on the position of the escaping unmanned surface vehicle, realizing estimation of the position and speed of the escaping unmanned surface vehicle by multiple virtual pursuers; based on the surrounding algorithm, the approaching algorithm and the weighing algorithm, designing a distributed cooperative hunting strategy of the multiple virtual pursuers for the escaping unmanned surface vehicle; based on the distributed cooperative hunting strategy, obtaining the hunting speed of the virtual pursuer; constructing a control barrier function to generate an optimal trajectory generator to avoid collision in the cooperative hunting process; defining longitudinal trajectory tracking error and lateral trajectory tracking error, and designing a line-of-sight guidance law; based on the line-of-sight guidance law and the speed signal given by the unmanned surface vehicle system, designing an extended state observer; based on the line-of-sight guidance law and the extended state observer, designing a model-free control law to realize the hunting control of multiple unmanned surface vehicles for one escaping unmanned surface vehicle.

2. The method of claim 1, wherein: The distributed target observer is used to realize the estimation of the position and speed of the escaping unmanned surface vehicle by multiple virtual pursuers based on the position of the escaping unmanned surface vehicle. wherein, represents the estimation of the position p e of the escaping UAV by the i-th virtual pursuer; represents the derivative of the estimation of the position p e of the escaping UAV by the i-th virtual pursuer; represents the estimation of the position p e of the escaping UAV by the j-th virtual pursuer; represents the estimation of the velocity u e of the escaping UAV by the i-th virtual pursuer; represents the derivative of the estimation of the velocity u e of the escaping UAV by the i-th virtual pursuer; K1 is a positive definite coefficient matrix; K2 is a positive definite coefficient matrix; c is a normal number; a ij is a communication topology parameter of the coordination of the plurality of virtual pursuers; b i is a detection structure parameter of the virtual pursuer and the escaping UAV.

3. The method of claim 1, wherein: The process of designing the distributed cooperative hunting strategy of the multiple virtual pursuers for the escaping unmanned surface vehicle based on the surrounding algorithm, the approaching algorithm and the weighing algorithm is as follows: the virtual pursuit signal of the ith tracking unmanned surface vehicle is defined as: and wherein, represents the position of the i-th virtual pursuer; represents the speed of the i-th virtual pursuer; represents the maximum speed value of the i-th virtual pursuer; Based on the state information of the escaping USV estimated by the distributed target observer, the relative distance ρ between the i-th virtual pursuer and the escaping USV is defined i and the relative angle β i : The Apollonian circle method is introduced in the pursuit strategy, the angle θ i defined as the angle between the tangent line l1 and the tangent line l2, expressed by the sine rule as follows: θ i = 2arcsin(μ i ) (7) The virtual pursuers group disperses counterclockwise, the covering angle a of the i-th virtual pursuer and the adjacent (i+1)-th virtual pursuer i is defined as: where β (M+) = β1; θ (M+) = θ1; when i = M, ζ i = 2π; when i ≠ M, ζ i = 0; α i < 0 means that the ith virtual pursuer and the (i+1)th virtual pursuer both track the escaping UAV; otherwise, the escaping UAV can escape. the occupation angle of all virtual pursuers is represented as: where θ G is the total occupancy angle of the encompassing task; The proposed distributed cooperative pursuit strategy includes a surround task and an approach task, the surround task is to increase the total occupancy angle θ G The surround strategy is designed as: λ is = k i ρ i (α i -α (i-) ) (10) where, when i = 0, a0= a M ; k i > 0 is the enclosing parameter; (a i - a (i-1) ) > 0 means the ith virtual pursuer encloses the escaping UAV counterclockwise; the approaching task is to shorten the distance to the escaping UAV, and the approaching strategy is designed as: λ ih = -h i ρ i (11) where h i > 0 is a proximity parameter; The weights of the surrounding strategy and the approaching strategy are determined by using a trade-off method, a trade-off coefficient is designed, and all virtual pursuers are assumed to move at the maximum speed movement; From equation (9) and where η i is a trade-off parameter, which is specified as follows: where κ i ∈ [0, 1]; κ i = 2 | α i - α (i-) | / 4 π - θ (i+) + θ (i-) is an enclosing weight; ι i ∈ [0, 1], is a proximity weight, and p i + p (i-) + p (i+) ≠ 0.

4. The method of claim 1, wherein: The process of constructing the control barrier function to generate the optimal trajectory generator is as follows: For static obstacles, consider the safety set The control barrier function is designed as The safety constraint condition of the control barrier function is constructed as where and is a positive parameter; For dynamic obstacles, consider the safety set The control barrier function is designed as The safety constraint condition of the control barrier function is constructed as where, and is a positive parameter; For the neighbor virtual pursuer, consider the safety set The control barrier function is designed as The safety constraint condition of the control barrier function is constructed as wherein, and is a positive parameter; The optimal trajectory generator based on the control barrier function is designed as follows: wherein, 5. The method of claim 1, wherein: The process of defining the longitudinal trajectory tracking error and the lateral trajectory tracking error and designing the line-of-sight guidance law is as follows: Define longitudinal trajectory tracking error x iz and lateral trajectory tracking error y iz as follows: wherein, is the tangent angle, i.e. the velocity direction of the planned trajectory; The line-of-sight guidance law is designed as follows: wherein, is a speed of the trajectory generator; is a look-ahead distance; and is a positive parameter.

6. The method of claim 1, wherein: The extended state observer is designed as follows: wherein are the extended states of the observer; is an estimate of u i ; is an estimate of r i ; is an estimate of σ iu ; is an estimate of σ ir ; is a derivative of the estimate of u i ; is a derivative of the estimate of r i ; is a derivative of the estimate of σ iu ; is a derivative of the estimate of σ ir ; are the designed observer gain coefficients.

7. The method of claim 1, wherein: The model-free control law is designed as follows: where γ iu and γ ir are control parameters; k iu and k ir are tuning parameters.

8. An obstacle environment unmanned surface vehicle cooperative pursuit controller, characterized in that: The distributed target observer, the cooperative hunting trajectory planning, the nominal trajectory generator, the optimal trajectory generator based on the control barrier function, the line-of-sight guidance law module, the extended state observer and the model-free control law are included. The input end of the distributed target observer is connected with the position signal of the escaping unmanned surface vehicle given by the communication network, and the output end of the distributed target observer is connected with the cooperative hunting trajectory planning. The input end of the cooperative hunting trajectory planning is connected with the covering angle signal of the ith virtual pursuer, the adjacent ith+1 virtual pursuer and the distributed target observer, and the output end of the cooperative hunting trajectory planning is connected with the nominal trajectory generator. The input end of the nominal trajectory generator is connected with the cooperative hunting trajectory planning, and the output end of the nominal trajectory generator is connected with the trajectory generator based on the control barrier function. The input end of the optimal trajectory generator based on the control barrier function is connected with the nominal trajectory generator, and the output end of the optimal trajectory generator based on the control barrier function is connected with the line-of-sight guidance law module. The input end of the line-of-sight guidance law module is connected with the optimal trajectory generator based on the control barrier function and the state signal given by the unmanned surface vehicle system, and the output end of the line-of-sight guidance law module is connected with the model-free control law. The input end of the extended state observer is connected with a given speed signal of the unmanned ship system, and the output end of the extended state observer is connected with a model-free control law. The input end of the model-free control law is connected with the extended state observer and a line-of-sight guidance law, and the output end of the model-free control law is connected with the unmanned ship system.

Images