Unmanned ship cluster safety formation reconstruction control system based on local detection information

Abstract

The application discloses a kind of unmanned ship cluster safety formation reconstruction control systems based on local detection information, applied to unmanned ship cluster, any one described unmanned ship corresponding communication connects a collision vector module and a safety formation reconstruction controller;The collision vector module is used to output the nearest collision point and collision vector corresponding to static obstacle, dynamic obstacle and other unmanned ship of current unmanned ship;The safety formation reconstruction controller includes safety space artificial potential field module, repulsion space artificial potential field module, mixed space artificial potential field module, based on artificial potential field nominal guidance law module, reciprocal control obstacle function constraint module and quadratic optimization module.The present application can enable unmanned ship to avoid unknown dynamic, static obstacles based on the local information of autonomous detection in complex dynamic environment, without needing to obtain the accurate position and shape information of obstacle in advance, to improve the flexibility and environmental adaptability of system.

Description

Technical Field

[0001] This invention relates to the field of multi-unmanned vessel control technology, and mainly studies the problem of unmanned vessel swarm safety priority formation reconfiguration control, and proposes an unmanned vessel swarm safety formation reconfiguration control system based on local detection information. Background Technology

[0002] In recent years, due to the limitations of single unmanned surface vessels (USVs), such as limited mission capabilities, low mission efficiency, high safety risks, limited data acquisition, and poor adaptability, multi-USV swarm control methods have attracted widespread attention from researchers both domestically and internationally. As an unmanned marine transportation platform, underactuated USV swarms have wide applications in both civilian and military fields. In the military field, USV swarms have significant application value, capable of performing tasks such as formation escort, swarm tracking, and swarm warfare. In the civilian field, USV swarms can greatly expand the scope of marine operations, with applications including marine environmental monitoring, collaborative resource exploration, and collaborative disaster search and rescue. Therefore, how to achieve reasonable and effective formation control of underactuated USV swarms has become a research hotspot in recent years.

[0003] It is important to note that current research primarily focuses on formation and maintenance of formations. However, when multiple unmanned surface vessels (USVs) operate collaboratively in formation, the complexity of the aquatic environment and the dynamic changes in mission requirements inevitably lead to formation reconfiguration. In actual operations, USV swarms may need to switch from one mission to another or from their current channel to another, often requiring changes or reconstruction of the USV formation. During this process, critical safety issues must be addressed, such as avoiding collisions with obstacles and nearby vessels, posing a significant challenge to the system. To address this problem, researchers both domestically and internationally have proposed various collision avoidance algorithms, including artificial potential fields, deep reinforcement learning, model predictive control, and obstacle control functions. However, control schemes based on these methods require prior environmental information and the accurate location and shape of obstacles. In real-world marine environments, this information is often difficult to obtain accurately. Summary of the Invention

[0004] To address the technical problem of existing unmanned surface vessel (USV) swarm formation control systems failing to accurately acquire prior environmental information and the precise location and shape of obstacles, this invention provides a USV swarm safety formation reconfiguration control system based on local detection information. This invention proposes a collision avoidance control system based on local detection information, enabling USVs to avoid unknown dynamic and static obstacles in complex dynamic environments based on autonomously detected local information, without needing to obtain the precise location and shape information of obstacles beforehand. This significantly improves the system's flexibility and environmental adaptability.

[0005] The technical means employed in this invention are as follows:

[0006] A safe formation reconstruction control system for unmanned surface vessels (USVs) based on local detection information is applied to USVs, the USVs cluster comprising N USVs, the control system comprising N collision vector modules and N safe formation reconstruction controllers, and each USV having a corresponding communication connection to a collision vector module and a safe formation reconstruction controller.

[0007] The collision vector module is used to output the nearest collision point and collision vector of the current unmanned surface vessel (USV) to static obstacles, dynamic obstacles, and other USVs.

[0008] The safe formation reconstruction controller includes a safe space-like artificial potential field module, a repulsive space-like artificial potential field module, a hybrid space-like artificial potential field module, a nominal guidance law module based on the artificial potential field, a reciprocal control obstacle function constraint module, and a quadratic optimization module, wherein:

[0009] The current unmanned surface vessel (USV) position and its formation reconstruction target position are connected to the input of the safe space artificial potential field module; the current USV position, static obstacle collision vector, and formation reconstruction target position are connected to the input of the repulsion space artificial potential field module; the safe space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module; the repulsion space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module; the hybrid space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module; the hybrid space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module. The gradient of the potential field function is connected to the input of the nominal guidance law module based on the artificial potential field; the current UAV's position, bow angle, sway velocity, and bow angular velocity are connected to the input of the nominal guidance law module based on the artificial potential field; the nominal guidance law output by the nominal guidance law module based on the artificial potential field is connected to the input of the secondary optimization module; the current UAV's position, bow angle, nearest collision point of dynamic obstacles, and nearest collision point of the UAV are connected to the input of the reciprocal control obstacle function constraint module; the dynamic obstacle reciprocal control obstacle function and neighboring ship reciprocal control obstacle function output by the reciprocal control obstacle function constraint module are connected to the input of the secondary optimization module; the safe formation reconstruction guidance law output by the secondary optimization module is connected to the control end of the UAV.

[0010] Furthermore, the kinematic model of the unmanned vessel is as follows:

[0011]

[0012] Where, x i ,y i This is the location of the unmanned ship in Earth coordinate system. It is the actual speed of the unmanned vessel, μ i It is the sway velocity of the unmanned ship in the ship's coordinate system, υ i It is the sway velocity of the unmanned vessel in the ship's coordinate system; r i It is the bow roll angular velocity, ψ iW =ψ i +β i ψ is the actual direction angle of movement of the unmanned vessel. i It is the bow roll angle; β i =atan2(v i / u i ) is the sideslip angle.

[0013] Furthermore, the collision vector module is used to output:

[0014]

[0015] Where, d iS Let p be the nearest collision point of the current unmanned vessel corresponding to the S-th static obstacle. iS =[x iS ,y iS ] T The collision vector, p iS =[x iS ,y iS ] T Let d be a line drawn from the position of the i-th unmanned vessel, perpendicular to the point on the boundary of the S-th static obstacle; iD Let p be the nearest collision point of the current unmanned vessel corresponding to the Dth dynamic obstacle. iD =[x iD ,y iD ] T The collision vector, p iD =[x iD ,y iD ] T Let d be a line drawn from the position of the i-th unmanned vessel, perpendicular to the point on the boundary of the D-th dynamic obstacle; ij Let p be the nearest collision point p between the current unmanned vessel and the j-th unmanned vessel. ij =[x ij ,y ij ] T The collision vector, p ij =[x ij ,y ij ] T Let p be the perpendicular point of a line drawn from the position of the i-th unmanned vessel to the boundary of the j-th unmanned vessel; i =[x i ,y i ] TLet be the position of the i-th unmanned vessel.

[0016] Furthermore, the safe space-type artificial potential field module is used to output:

[0017]

[0018] Among them, F iSS For the artificial potential field function of the safe space class, ▽F iSS For the gradient of the artificial potential function in the safe space class, p ie =p i -p id p i =[x i ,y i ] T p is the position of the i-th unmanned ship. id =[x id ,y id ] T It is the target position for reconstructing the formation of the i-th unmanned vessel.

[0019] Furthermore, the repulsive space-like artificial potential field module is used to output:

[0020]

[0021] in, To exclude spatial artificial potential field functions, To exclude the gradient of the spatial artificial potential function, p ie =p i -p id p i =[x i ,y i ] T p is the position of the i-th unmanned ship. id =[x id ,y id ] T δ is the target position for reconstructing the formation of the i-th unmanned vessel. d It is the sensing range radius of the unmanned vessel, k1 is a strictly positive real number, and g i ′ S It is g iS The derivative, It is along the collision vector d iS A unit vector in direction.

[0022] Furthermore, the hybrid spatial artificial potential field module is used to output:

[0023]

[0024] Among them, F ihLet F be the artificial potential field function in the mixed space. ih The gradient of the artificial potential function in the mixed space. Describe the set ε i The cardinality, λ i These are weighting coefficients based on an exponential function. k λ >0 is an adjustment parameter, r min It is the smallest switching range.

[0025] Furthermore, the nominal guidance law module based on the artificial potential field is used to output:

[0026]

[0027] in, For the nominal guidance law, k u and k r It is the controller gain, ρ i =||p i ||,p i =[x i ,y i ] T It is the position of the i-th unmanned ship, γ Fi =arctan(▽F ih ).

[0028] Furthermore, the reciprocity control barrier function constraint module is used to output:

[0029]

[0030] Among them, h iD h is the obstacle function for reciprocal control of dynamic obstacles. ij For the reciprocal control barrier function of neighboring ships, It is the relative angle between the unmanned vessel and the nearest point of collision, p i =[x i ,y i ] T , parameter d D k represents the minimum safe distance between an unmanned surface vessel and a dynamic obstacle. H1 >0 is a control parameter. It is the relative angle between the unmanned vessel and the nearest point of collision, p i =[x i ,y i ] T , parameter d N k represents the minimum safe distance between an unmanned vessel and its neighboring vessel. H2 >0 is a control parameter.

[0031] Furthermore, the secondary optimization module is used to output:

[0032]

[0033] Where, α iu ,α ir For the guidance law of safe formation reconstruction, k2 is a κ-type function, and k3 is a κ-type function. L f h iD +L g h iD r i The dynamic obstacle reciprocity control obstacle function h represents the obstacle function. iD Li derivative, L f h ij +L g h ij r represents the reciprocal control barrier function for neighboring ships. ij Li Daoshu.

[0034] Compared with the prior art, the present invention has the following advantages:

[0035] First, compared with existing formation control methods that mainly focus on the formation and maintenance of formation, the control method proposed in this invention realizes the formation reconstruction task of unmanned vessel clusters.

[0036] Second, compared with existing formation control methods that cannot guarantee the avoidance of collisions with obstacles outside the formation of unmanned vessels, the control system proposed in this invention realizes collision avoidance between unmanned vessel clusters and static and dynamic obstacles and between vessels.

[0037] Third, compared with existing collision avoidance methods, the control system proposed in this invention does not require accurate prior environmental information or accurate obstacle shape and position information. It can achieve collision avoidance between arbitrarily shaped convex static obstacles and ships based only on local detection information (nearest collision point and collision vector), which improves the flexibility and environmental adaptability of unmanned ship swarm formation. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a block diagram of the unmanned vessel swarm safety formation reconstruction control system based on local detection information in an embodiment of the present invention.

[0040] Figure 2a This is a schematic diagram of the safe formation reconstruction of the unmanned vessel cluster in an embodiment of the present invention.

[0041] Figure 2b This is a collision avoidance effect diagram of the unmanned vessel cluster safety formation reconstruction in an embodiment of the present invention.

[0042] Figure 3 This is a diagram of the artificial potential field function values ​​for the safe formation reconstruction of the unmanned vessel cluster in an embodiment of the present invention.

[0043] Figure 4 This is a numerical diagram of the modulus of the minimum collision vector between the unmanned vessel and the static obstacle in the unmanned vessel cluster safety formation reconstruction in this embodiment of the invention.

[0044] Figure 5 This is a numerical diagram of the magnitude of the collision vector between the unmanned vessel and the dynamic obstacle in the reconstructed safe formation of the unmanned vessel swarm in this embodiment of the invention.

[0045] Figure 6 This is a numerical diagram of the modulus of the minimum collision vector between unmanned vessels in the unmanned vessel cluster safety formation reconstruction in this embodiment of the invention. Detailed Implementation

[0046] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0047] like Figure 1 As shown, this invention provides a safe formation reconstruction control system for an unmanned surface vessel (USV) swarm based on local detection information. The USV swarm comprises N USVs, and the control system includes N collision vector modules and N safe formation reconstruction controllers. Each USV is connected to one collision vector module and one safe formation reconstruction controller via communication. The collision vector module outputs the nearest collision point and collision vector of the current USV to static obstacles, dynamic obstacles, and other USVs. The safe formation reconstruction controller includes a safe space-like artificial potential field module, a repulsive space-like artificial potential field module, a hybrid space-like artificial potential field module, a nominal guidance law module based on the artificial potential field, a reciprocal control obstacle function constraint module, and a quadratic optimization module.

[0048] To address the problem of safe formation reconfiguration control for unmanned surface vessel (USV) swarms, this invention proposes a safe formation reconfiguration control system based on local detection information. This system considers that each USV can measure the nearest point (nearest collision point) and the relative position vector of each nearest obstacle (nearest collision vector) within its sensing range in a global coordinate system using its onboard radar and shipborne sensors. First, an artificial potential field function is designed based on the collision vectors, enabling each USV to reach its target while avoiding collisions with convex static obstacles. Then, a reciprocal obstacle control function is defined based on the nearest collision points to guide each USV to avoid collisions with convex dynamic obstacles and neighboring USVs. Finally, a constrained quadratic programming problem is constructed to compute the optimal guidance signal. Therefore, under this control method, the USV swarm can avoid obstacles and complete formation reconfiguration without using any prior information about the position and shape of obstacles, improving the environmental adaptability and flexibility of the USV formation in complex sea environments.

[0049] In this application, the kinematic model of the unmanned surface vessel is as follows:

[0050]

[0051] Where x i ,y i This refers to the position of the unmanned ship in the Earth coordinate system; ψ i It is the bow roll angle; μ i It is the sway velocity of the unmanned vessel in the hull coordinate system; υ i It is the sway velocity of the unmanned vessel in the ship's coordinate system; r i It is the bow roll rate.

[0052] Assuming the actual speed and direction of the unmanned vessel can be measured, the kinematic model of the unmanned vessel can be rewritten as follows:

[0053]

[0054] in This is the actual speed of the unmanned vessel; ψ iW =ψ i +β i β is the actual direction angle of movement of the unmanned vessel. i =atan2(v i / u i ) is the sideslip angle;

[0055] The output of the collision vector module corresponds to the nearest collision point p of the i-th unmanned vessel with the S-th static obstacle. iS =[x iS ,y iS ] T and collision vector d iSConnected to the input of the unmanned surface vessel (USV). The output of the collision vector module shows the nearest collision point p of the i-th USV corresponding to the D-th dynamic obstacle. iD =[x iD ,y iD ] T and collision vector d iD Connected to the input of the unmanned surface vessel (USV). The output of the collision vector module shows the nearest collision point p between the i-th USV and the j-th USV. ij =[x ij ,y ij ] T and collision vector d ij It is connected to the input terminal of the unmanned vessel.

[0056] Any of the aforementioned safe formation reconstruction controllers comprises the following modules: a safe space-like artificial potential field module, a repulsive space-like artificial potential field module, a hybrid space-like artificial potential field module, a nominal guidance law module based on the artificial potential field, a reciprocal control obstacle function constraint module, and a secondary optimization module.

[0057] The position of the i-th unmanned vessel is output x. i ,y i The target position x of the formation reconstruction of the i-th unmanned vessel id ,y id Connected to the input of the artificial potential field module for safe space. Output of the i-th unmanned surface vessel: the position x of the unmanned surface vessel. i ,y i Collision vector d iS The target position x of the formation reconstruction of the i-th unmanned vessel id ,y id Connected to the input of the repulsive space type artificial potential field module. The output of the safe space type artificial potential field module is the safe space type artificial potential field function F. iSS and the gradient ▽F of the artificial potential field function in the safe space iSS Connected to the input of the hybrid space-type artificial potential field module. The output of the repulsive space-type artificial potential field module is the repulsive space-type artificial potential field function. gradient of artificial potential field function in repulsive space Connected to the input of the hybrid space-type artificial potential field module. The output of the hybrid space-type artificial potential field module is the hybrid space artificial potential field function F. ih and the gradient ▽F of the artificial potential field function in the mixed space ih Connected to the input of the nominal guidance law module based on a quasi-artificial potential field. Output of the unmanned surface vessel: Unmanned surface vessel position x. i ,y i Bow roll angle ψ i sway velocity υ i and bow roll rate r iIt is connected to the input of the nominal guidance law module based on a quasi-artificial potential field. The output of the nominal guidance law module based on a quasi-artificial potential field is the nominal guidance law. Connected to the input of the secondary optimization module. Output of the unmanned surface vessel: the position x of the unmanned surface vessel. i ,y i Bow roll angle ψ i The closest collision point p iD =[x iD ,y iD ] T and the nearest collision point p ij =[x ij ,y ij ] T Connected to the input of the reciprocal control obstacle function constraint module. The output of the reciprocal control obstacle function constraint module is the dynamic obstacle reciprocal control obstacle function h. iD Reciprocal control barrier function h with neighboring ships ij It is connected to the input of the secondary optimization module. The output of the secondary optimization module is the safety formation reconstruction guidance law α. iu ,α ir It is connected to the input terminal of the unmanned vessel.

[0058] As a preferred embodiment of the present invention, the collision vector module is designed as follows:

[0059] Draw a line perpendicular to a point on the boundary of obstacle o from the current position of the i-th unmanned vessel. This point is the nearest collision point p. io =[x io ,y io ] T Collision vector d io as follows

[0060] d io =p i -p io (3)

[0061] Where p i =[x i ,y i ] T .

[0062] The output of the collision vector module is the nearest collision point p corresponding to the S-th static obstacle. iS =[x iS ,y iS ] T and collision vector d iS The nearest collision point p corresponding to the Dth dynamic obstacle. iD =[x iD ,y iD ] T and collision vector diD The nearest collision point p corresponding to the j-th unmanned vessel ij =[x ij ,y ij ] T and collision vector d ij .

[0063]

[0064] As a preferred embodiment of the present invention, the design of the safe space-type artificial potential field module is as follows:

[0065] The output of the unmanned surface vessel is the position x of the i-th unmanned surface vessel. i ,y i The target position x of the formation reconstruction of the i-th unmanned ship. id ,y id As the input signal to the safe space-type artificial potential field module, the output of the safe space-type artificial potential field module is the safe space-type artificial potential field function F. iSS The gradient ▽F of the artificial potential function in the safe space class iSS .

[0066]

[0067] Where p ie =p i -p id p i =[x i ,y i ] T p is the position of the i-th unmanned ship. id =[x id ,y id ] T It is the target position for reconstructing the formation of the i-th unmanned vessel.

[0068] As a preferred embodiment of the present invention, the design of the repulsive space-like artificial potential field module is as follows:

[0069] The output of the unmanned surface vessel is the position x of the i-th unmanned surface vessel. i ,y i and the collision vector d corresponding to the S-th static obstacle iS and the target position x for the formation reconstruction of the i-th unmanned vessel. id ,y id As the input signal to the repulsive space-like artificial potential field module, the output of the repulsive space-like artificial potential field module is the repulsive space-like artificial potential field function. Gradient of the artificial potential function in repulsive space

[0070]

[0071] Where p ie =p i -p id p i =[x i ,y i ] T p is the position of the i-th unmanned ship. id =[x id ,y id ] T δ is the target position for reconstructing the formation of the i-th unmanned vessel. d K1 is the radius of the unmanned vessel's sensing range, and k1 is a strictly positive real number.

[0072]

[0073] Where g i ′ S It is g iS The derivative, It is along the collision vector d iS A unit vector in direction.

[0074] As a preferred embodiment of the present invention, the design of the hybrid space-like artificial potential field module is as follows:

[0075] The output of the safe space artificial potential field module is the safe space artificial potential field function F. iSS The gradient ▽F of the artificial potential function in the safe space class iSS And the output function of the repulsive space artificial potential field module. Gradient of the artificial potential function in repulsive space This serves as the input signal for the hybrid space-type artificial potential field module. The output of the hybrid space-type artificial potential field module is the hybrid space artificial potential field function F. ih The gradient ▽F of the artificial potential function in the mixed space ih .

[0076]

[0077] in Describe the set ε i The cardinality, λ i The weighting coefficients are based on the exponential function as follows:

[0078]

[0079] in k λ >0 is an adjustment parameter, r min It is the smallest switching range.

[0080] As a preferred embodiment of the present invention, the design of the nominal guidance law module based on a quasi-artificial potential field is as follows:

[0081] Output of unmanned surface vessel (USV) position x i ,y i Bow roll angle ψ i sway velocity υ i and bow roll rate r i And the output of the hybrid space artificial potential field module, the hybrid space artificial potential field function F. ih The gradient ▽F of the artificial potential function in the mixed space ih This serves as the input signal for the nominal guidance law module based on a quasi-artificial potential field. The output of this module is the nominal guidance law.

[0082]

[0083] Where k u and k r It is the controller gain, ρ i =||p i ||,p i =[x i ,y i ] T It is the position of the i-th unmanned ship, γ Fi =arctan(▽F ih ).

[0084] As a preferred embodiment of the present invention, the reciprocal control barrier function constraint module is designed as follows:

[0085] Output of unmanned surface vessel (USV) position x i ,y i Bow roll angle ψ i and the nearest collision point p corresponding to the dynamic obstacle. iD =[x iD ,y iD ] T and the nearest collision point p corresponding to the adjacent ship ij =[x ij ,y ij ] T This serves as the input signal to the reciprocal control obstacle function constraint module. The output of the reciprocal control obstacle function constraint module is the dynamic obstacle reciprocal control obstacle function h. iD Reciprocal control barrier function h with neighboring ships ij .

[0086] Dynamic obstacle reciprocity control obstacle function h iD as follows

[0087]

[0088] Where the distance function D iD and direction function H iD as follows

[0089]

[0090] in It is the relative angle between the unmanned vessel and the nearest point of collision, p i =[x i ,y i ] T , parameter d D k represents the minimum safe distance between an unmanned surface vessel and a dynamic obstacle. H1 >0 is a control parameter.

[0091] Neighboring ship reciprocal control barrier function h ij as follows

[0092]

[0093] Where the distance function D ij and direction function H ij as follows

[0094]

[0095] in It is the relative angle between the unmanned vessel and the nearest point of collision, p i =[x i ,y i ] T , parameter d N k represents the minimum safe distance between an unmanned vessel and its neighboring vessel. H2 >0 is a control parameter.

[0096] As a preferred embodiment of the present invention, the secondary optimization module is designed as follows:

[0097] The output nominal guidance law based on the artificial potential field nominal guidance law module And the output dynamic obstacle reciprocity control obstacle function h of the reciprocity control obstacle function constraint module. iD Reciprocal control barrier function h with neighboring ships ij This serves as the input signal to the secondary optimization module. The output of the secondary optimization module is the safety formation reconstruction guidance law α. iu ,α ir .

[0098] The safe guidance angular velocity of the unmanned surface vessel is obtained by solving a quadratic optimization problem.

[0099]

[0100] Where k2 is a κ-class function, and k3 is a κ-class function. L f h iD +L g h iD r i The dynamic obstacle reciprocity control obstacle function h represents the obstacle function. iD Li derivative, L f h ij +L g h ij r represents the reciprocal control barrier function for neighboring ships. ij Li Daoshu.

[0101] Furthermore, the guidance law α for safe formation reconstruction iu ,α ir as follows

[0102]

[0103] To better explain the present invention, the present invention will be further illustrated below using a specific example of six unmanned ships. Each unmanned ship in the system satisfies the motion model shown in (1). The control objective of the six ships is to transform from a rectangular formation to a wedge formation. During the formation reconstruction process, the unmanned ships achieve collision avoidance with arbitrarily shaped convex static obstacles and neighboring ships based on local detection information.

[0104] The control method for each unmanned vessel is as follows: Figure 1 As shown in Figure 2-6, the simulation results are as follows. Figure 2 shows that the six unmanned vessels transformed from a rectangular formation to a wedge formation. Furthermore, no collisions occurred between any unmanned vessel and static or dynamic obstacles or adjacent vessels. Figure 3 The change of the artificial potential field function value over time shows that the convergence of the unmanned vessel to the target position is consistent with the monotonic convergence of the artificial potential field function value to 0. Figure 4 Let represent the magnitude of the minimum collision vector between the unmanned vessel and the four static obstacles. It can be seen that no collision occurred between the unmanned vessel and the static obstacles. Figure 5 This represents the magnitude of the collision vector between the unmanned surface vessel (USV) and the dynamic obstacle. It can be seen that no collision occurred between the USV and the dynamic obstacle. Figure 6 This represents the magnitude of the minimum collision vector between the unmanned vessels. It can be seen that no collisions occurred between the unmanned vessels.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A safe formation reconstructing control system for unmanned surface vessel swarms based on local detection information, characterized in that, The system is applied to an unmanned surface vessel (USV) swarm, which includes N USVs. The control system includes N collision vector modules and N safe formation reconstruction controllers. Each USV has a corresponding communication connection to a collision vector module and a safe formation reconstruction controller. The collision vector module is used to output the nearest collision point and collision vector of the current unmanned surface vessel (USV) to static obstacles, dynamic obstacles, and other USVs. The safe formation reconstruction controller includes a safe space-like artificial potential field module, a repulsive space-like artificial potential field module, a hybrid space-like artificial potential field module, a nominal guidance law module based on the artificial potential field, a reciprocal control obstacle function constraint module, and a quadratic optimization module, wherein: The current unmanned surface vessel (USV) position and its formation reconstruction target position are connected to the input of the safe space artificial potential field module; the current USV position, static obstacle collision vector, and formation reconstruction target position are connected to the input of the repulsion space artificial potential field module; the safe space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module; the repulsion space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module; the hybrid space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module; the hybrid space artificial potential field function and its gradient are connected to the input of the hybrid space artificial potential field module. The gradient of the potential field function is connected to the input of the nominal guidance law module based on the artificial potential field; the current UAV's position, bow angle, sway velocity, and bow angular velocity are connected to the input of the nominal guidance law module based on the artificial potential field; the nominal guidance law output by the nominal guidance law module based on the artificial potential field is connected to the input of the secondary optimization module; the current UAV's position, bow angle, nearest collision point of dynamic obstacles, and nearest collision point of the UAV are connected to the input of the reciprocal control obstacle function constraint module; the dynamic obstacle reciprocal control obstacle function and neighboring ship reciprocal control obstacle function output by the reciprocal control obstacle function constraint module are connected to the input of the secondary optimization module; the safe formation reconstruction guidance law output by the secondary optimization module is connected to the control end of the UAV.

2. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 1, characterized in that, The kinematic model of the unmanned vessel is as follows: in, This is the location of the unmanned ship in the Earth coordinate system. This is the actual speed of the unmanned boat. It is the pitch velocity of the unmanned vessel in the ship's coordinate system. It is the sway velocity of the unmanned vessel in the ship's coordinate system; It is the bow roll rate. It is the actual direction angle of movement of the unmanned vessel. It is the bow yaw angle; It is the sideslip angle. .

3. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 1, characterized in that, The collision vector module is used for output: in, The current unmanned vessel corresponds to the first The nearest collision point of a static obstacle The collision vector, For from the first Draw a line perpendicular to the position of the unmanned vessel. A vertical point on the boundary of a static obstacle; The current unmanned vessel corresponds to the first The nearest collision point of a dynamic obstacle The collision vector, For from the first Draw a line perpendicular to the position of the unmanned vessel. A vertical point on the boundary of a dynamic obstacle; The current unmanned vessel corresponds to the first The nearest collision point of the unmanned vessel The collision vector, For from the first Draw a line perpendicular to the position of the unmanned vessel. The perpendicular point on the boundary of the unmanned vessel; For the first Location of the unmanned vessel.

4. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 1, characterized in that, The safe space-type artificial potential field module is used for output: in, For the artificial potential field function of the safe space class, The gradient of the artificial potential function in the safe space class. , It is the first The location of the unmanned boat It is the first The formation of unmanned ships reconstructs the target location.

5. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 4, characterized in that, The repulsive space-type artificial potential field module is used for output: in, To exclude spatial artificial potential field functions, To exclude the gradient of the spatial artificial potential function, , It is the first The location of the unmanned boat It is the first The formation of unmanned ships reconstructs the target location. It is the radius of the unmanned ship's sensing range. It is a strictly positive real number. yes The derivative of It is along The unit vector of direction. The current unmanned vessel corresponds to the first The nearest collision point of a static obstacle The collision vector.

6. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 5, characterized in that, The hybrid spatial artificial potential field module is used for output: in, Let be the artificial potential field function in the mixed space. The gradient of the artificial potential function in the mixed space. Represents a set The base number, These are weighting coefficients based on an exponential function. , It's about adjusting parameters. It is the smallest switching range.

7. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 6, characterized in that, The nominal guidance law module based on a quasi-artificial potential field is used for output: in, For the nominal guidance law, and It is the controller gain. , It is the first The location of the unmanned boat , It is a bow yaw. , It is the sideslip angle. It is the pitch velocity of the unmanned vessel in the ship's coordinate system. It is the sway velocity of the unmanned vessel in the ship's coordinate system.

8. The unmanned surface vessel swarm safety formation reconstructing control system based on local detection information according to claim 7, characterized in that, The reciprocal control barrier function constraint module is used for output: in, For dynamic obstacles, the reciprocal control obstacle function is... For the reciprocal control barrier function of neighboring ships, It is the relative angle between the unmanned vessel and the nearest point of collision with respect to a dynamic obstacle. This indicates the minimum safe distance between the unmanned vessel and dynamic obstacles. These are control parameters. It is the relative angle between the unmanned vessel and the nearest point of collision with a neighboring vessel. This indicates the minimum safe distance between unmanned vessels and adjacent vessels. These are control parameters. For from the first Draw a line perpendicular to the position of the unmanned vessel. A vertical point on the boundary of a dynamic obstacle. For the first The location of the unmanned vessel For from the first Draw a line perpendicular to the position of the unmanned vessel. The vertical point on the boundary of the unmanned vessel.

9. A safe formation reconstructing control system for unmanned surface vessels based on local detection information according to claim 8, characterized in that, The secondary optimization module is used to output: in, To reconstruct guidance laws for safe formations, It is Class function, It is Class function, Represents the dynamic obstacle reciprocity control obstacle function Li Daoshu, Represents the reciprocal control barrier function of neighboring ships. Li Daoshu, For the nominal guidance law, It is the bow roll rate.

Images