A data-driven collaborative tracking control method for multiple unmanned ship swarms

Abstract

The present invention discloses a data-driven collaborative tracking control method for a cluster of multiple unmanned ships, which belongs to the field of intelligent control of unmanned ship clusters. The present invention first constructs a communication topology for distributed collaborative control of a cluster of multiple unmanned ships; constructs a dynamic linearized data model of a single unmanned ship's speed and heading system with interference terms based on unknown external interference and nonlinear uncertainty; constructs a smooth anti-interference trajectory tracking algorithm for a single unmanned ship to track a dynamic target point; constructs a pseudo-Jacobian matrix estimation and distributed disturbance compensation controller for distributed model-free adaptive control of multiple unmanned ships; constructs a reset algorithm and a controller of an actual unmanned ship; and finally, realizes collaborative tracking control of a cluster of multiple unmanned ships. The present invention breaks through the limitations of traditional control methods, does not rely on complex system model information, and only uses the input and output data of the system to realize a completely data-driven intelligent control strategy.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent control of unmanned ship clusters, and in particular relates to a data-driven collaborative tracking control method for multiple unmanned ship clusters. Background Art

[0002] An unmanned ship is an unmanned ship that can perform predetermined tasks and plays an important role in enemy ship tracking, ocean monitoring and maritime rescue. As a highly complex nonlinear time-varying system, an unmanned ship has the characteristics of strong coupling and under-actuation, and faces challenges such as model uncertainty, unmodeled dynamics and strong external interference. At present, most control methods are model-based, and these methods rely heavily on accurate dynamic models when designing controllers, resulting in overly complex structures. However, it is very difficult to obtain accurate model data in real time for unmanned ships, and most surge / heading control methods do not take into account the model information of under-actuated unmanned ships, such as the maneuverability and maneuverability of the ship, and inherent limitations such as mechanical saturation of the actuator. In addition, these methods rarely consider the performance of unmanned ships under actual complex operating conditions, such as uncertain factors such as model perturbations and interference forces of the marine environment. This poses a challenge to the collaborative control of multiple unmanned ship clusters.

[0003] At present, unmanned ship trajectory tracking control algorithms are mainly divided into two types: model-based and model-free. When the model information is accurate, the model-based control method is effective, but the complex nonlinear characteristics of unmanned ships make it very difficult to establish an accurate model, which affects the control effect. Some existing model-free algorithms, such as PID control, lack adaptive adjustment capabilities. Although the adaptive fuzzy control algorithm can provide a certain degree of adaptability, it is necessary to design an adaptive law for all fuzzy system weights, which increases the computational burden and is not suitable for actual engineering needs. Therefore, reducing the computational burden and improving the applicability of the algorithm has become an important research direction.

[0004] Model-free adaptive control is a control strategy that does not rely on the precise mathematical model of the controlled object. It uses the input / output data of the system for online learning and is particularly suitable for complex systems that are difficult to model. Model-free adaptive control can estimate the system in real time online, so it can still provide excellent control effects in the face of inaccurate model parameters, complex or time-varying systems. Therefore, applying model-free adaptive control to high-performance control of unmanned ships has become a hot area of ​​current research.

[0005] In real application scenarios, single unmanned ships face the dual constraints of operating range and flexibility, which will undoubtedly prolong the time to complete the task and affect the overall work efficiency. Therefore, it is urgent to further build an unmanned ship cluster collaborative operation system based on the basic platform of single unmanned ships. The collaboration of multiple unmanned ships can not only cover a larger operating area and obtain more comprehensive and richer ocean data, but also further enhance flexibility through mutual communication and coordination. As the result of the development of marine robot technology to a certain stage, the collaboration of multiple unmanned ships will play a mainstream role in ocean exploration. The high degree of parallelism and redundancy of its system is unmatched by the single unmanned ship system.

[0006] The unmanned ship trajectory tracking algorithm needs to link the guidance system with the motion control system, and achieve trajectory tracking by adjusting the speed and heading. It also needs to consider adjusting the speed of the unmanned ship and the setting of the safety distance in real time according to the distance from the target point. At the same time, environmental interference such as wind, current, and waves may affect the speed and heading of the unmanned ship, resulting in the inability to accurately track the target. In order to achieve high-performance collaborative tracking control of unmanned ship clusters in complex marine environments, interference and uncertainty estimation is a key issue. The multi-unmanned ship cluster system is large in scale and has complex associated topology. It is necessary to design a distributed unmanned ship cluster control scheme, the input and output data required for communication between adjacent unmanned ships, and realize fully data-driven and distributed control of multi-unmanned ship cluster collaborative tracking. These challenges need to be solved through advanced control algorithms and technologies.

[0007] Currently, there is little research on data-driven collaborative tracking control of multiple unmanned ship swarms, especially considering the nonlinear uncertainty of unmanned ships and wave interference in actual situations, so it remains an important issue in this field. Summary of the invention

[0008] The purpose of this invention is to provide a data-driven multi-unmanned ship cluster cooperative tracking control method. The proposed controller does not need to use the unmanned ship model information, but only uses the system input / output data to achieve the unmanned ship cluster cooperative tracking of the reference target. It takes nonlinear uncertainty, wave interference and cluster cooperative tracking into consideration, and has wide applicability.

[0009] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0010] In a first aspect, the present invention discloses a data-driven multi-unmanned ship cluster collaborative tracking control method, comprising:

[0011] 1) According to the initial position of each unmanned ship in the multi-unmanned ship cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked, the communication topology of the distributed collaborative control of the multi-unmanned ship cluster is constructed;

[0012] 2) Based on unknown external disturbances and nonlinear uncertainties, a dynamic linearized data model of the speed and heading system of a single unmanned ship with disturbance terms is constructed;

[0013] 3) Construct a smooth anti-interference trajectory tracking algorithm for a single unmanned ship to track a dynamic target point, and obtain the expected tracking speed, expected tracking heading and expected yaw speed of the single unmanned ship;

[0014] 4) Construct pseudo-Jacobian matrix estimation, disturbance estimation and distributed disturbance compensation controller for distributed model-free adaptive control of multiple unmanned ships;

[0015] 5) Based on the pseudo-Jacobian matrix estimation obtained in step 4), a reset algorithm is constructed to update the distributed disturbance compensation controller to obtain the actual controller of the unmanned ship that meets the physical requirements;

[0016] 6) Based on the expected tracking speed, expected tracking heading and expected yaw speed obtained in step 3), the pseudo-Jacobian matrix estimation and interference estimation obtained in step 4), and the reset algorithm and actual controller obtained in step 5), distributed collaborative tracking control is performed on the multi-unmanned ship cluster to achieve collaborative tracking of the target point to be tracked by the multi-unmanned ship cluster, and the heading and speed of each unmanned ship are consistent with the target point to be tracked.

[0017] In a second aspect, the present invention discloses a multi-unmanned ship cluster collaborative tracking control system for implementing the method, comprising:

[0018] A communication topology building module is used to build a communication topology for distributed collaborative control of a multi-unmanned ship cluster based on the initial position of each unmanned ship in the multi-unmanned ship cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked;

[0019] A model acquisition module is used to construct a dynamic linear data model of the unmanned ship's speed and heading system with interference terms based on unknown external interference and nonlinear uncertainty;

[0020] A pseudo-Jacobian matrix and controller acquisition module, which is used to obtain the pseudo-Jacobian matrix and distributed disturbance compensation controller of the model based on the model-free adaptive control principle;

[0021] The tracking module is used to obtain the state information of the target point to be tracked and the state information of the unmanned ship to design the expected output of the unmanned ship trajectory tracking control;

[0022] An estimation module, which is used to estimate the pseudo-Jacobian matrix through an estimation algorithm, and to estimate unknown external interference and nonlinear uncertainty;

[0023] An updating module, which updates the distributed disturbance compensation controller based on the results obtained by the tracking module and the estimation module;

[0024] The control module converts the distributed disturbance compensation controller obtained by the update module into a controller of an unmanned ship that actually meets the physical requirements, and performs distributed collaborative tracking control on a cluster of multiple unmanned ships.

[0025] In a third aspect, the present invention discloses an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the data-driven multi-unmanned ship cluster collaborative tracking control method.

[0026] In a fourth aspect, the present invention discloses a machine-readable storage medium, which stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, they are used to implement the data-driven multi-unmanned ship cluster collaborative tracking control method.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1) Based on model-free adaptive control, the present invention considers the equivalent dynamic linearization data model of the unmanned ship speed and heading system with unknown external interference and nonlinear uncertainty, and designs a data-driven unmanned ship speed and heading controller, so that the unmanned ship cluster can track the target by adjusting the speed and heading. This controller has the characteristics of model-free, adaptive and distributed, and exhibits extremely high robustness and reliability. Model-free means that the controller does not require model information of the unmanned ship, and only uses system input / output data for control. Adaptability means that the controller can adjust the controller parameters to maintain system stability according to changes in the external environment, and can adapt to changes and intensity of external interference. Distributed means that the controller can distribute control of the unmanned ship cluster. The unmanned ship only needs to know a small amount of information around it to respond, which reduces the delay of information transmission and improves the response speed of the system, thereby improving the data-driven unmanned ship cluster collaborative tracking control capability in complex environments.

[0029] 2) The present invention designs a smooth anti-interference trajectory tracking algorithm for unmanned ships under the influence of nonlinear uncertainty and wave interference. An estimation algorithm is designed to estimate the pseudo-Jacobian matrix of model-free adaptive control and unknown external interference and nonlinear uncertainty, and a distributed disturbance compensation controller is designed to reduce the influence of uncertainty and interference, improve the stability of the system, and have strong robustness; at the same time, a trajectory tracking algorithm is introduced to link the guidance system with the motion control system, and trajectory tracking is achieved by adjusting the speed and heading, and considering adjusting the speed of the unmanned ship in real time according to the distance from the target point and the setting of the safety distance, which provides key technical support for data-driven multi-unmanned ship cluster collaborative tracking control. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a multi-unmanned ship cluster communication topology diagram according to an exemplary embodiment;

[0031] Figure 2 is a flow chart of a data-driven multi-unmanned ship cluster collaborative tracking control method according to an exemplary embodiment;

[0032] Figure 3 is a schematic diagram of collaborative tracking of multiple unmanned ship clusters according to an embodiment of the present invention;

[0033] Figure 4 is a schematic diagram of the heading consistency error of an unmanned ship according to an embodiment of the present invention;

[0034] Figure 5 is a schematic diagram of the speed consistency error of an unmanned ship according to an embodiment of the present invention;

[0035] Figure 6 Schematic diagram of the speed change of the unmanned ship according to an embodiment of the present invention;

[0036] Figure 7 is a schematic diagram of thrust change of an unmanned ship according to an embodiment of the present invention;

[0037] Figure 8 Schematic diagram of the steering torque change of the unmanned ship according to an embodiment of the present invention;

[0038] Fig. 9 is a schematic diagram of unmanned ship interference and interference estimation according to an embodiment of the present invention;

[0039] Fig.10 It is a schematic diagram of interference estimation error of an unmanned ship according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The present invention is further described and illustrated below in conjunction with specific embodiments. The embodiments are merely exemplary of the present disclosure and do not define the scope of limitation. The technical features of each embodiment of the present invention may be combined accordingly without conflicting with each other.

[0041] Figure 2 is a flow chart of a data-driven multi-unmanned ship cluster collaborative tracking control method according to an exemplary embodiment. Figure 2 As shown, the method may include the following steps:

[0042] S1: Construct the communication topology of distributed collaborative control of multiple unmanned ship clusters based on the initial positions of each unmanned ship in the cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked;

[0043] S2: Based on unknown external disturbances and nonlinear uncertainties, a dynamic linearized data model of the speed and heading system of a single unmanned ship with disturbance terms is constructed;

[0044] S3: Construct a smooth anti-interference trajectory tracking algorithm for a single unmanned ship to track a dynamic target point, and obtain the expected tracking speed, expected tracking heading and expected yaw speed of the single unmanned ship;

[0045] S4: Construct pseudo-Jacobian matrix estimation, disturbance estimation and distributed disturbance compensation controller for distributed model-free adaptive control of multiple unmanned ships;

[0046] S5: Based on the pseudo-Jacobian matrix estimation obtained in S4, a reset algorithm is constructed to update the distributed disturbance compensation controller to obtain the actual controller of the unmanned ship that meets the physical requirements;

[0047] S6: Based on the expected tracking speed, expected tracking heading and expected yaw speed obtained in S3, the pseudo-Jacobian matrix estimation and interference estimation obtained in S4, and the reset algorithm and actual controller obtained in S5, distributed collaborative tracking control is performed on the multi-unmanned ship cluster to achieve collaborative tracking of the target point to be tracked by the multi-unmanned ship cluster, and the heading and speed of each unmanned ship are consistent with the target point to be tracked.

[0048] It can be seen from the above embodiments that the present invention aims at the problem of collaborative tracking control of a cluster of multiple unmanned ships based on data-driven control, while considering the influence of nonlinear uncertainty and wave interference, and constructs a trajectory tracking algorithm, an estimation algorithm and a distributed model-free adaptive control. The trajectory tracking algorithm, the estimation algorithm and the distributed model-free adaptive control are used to perform collaborative tracking control on a cluster of multiple unmanned ships. The distributed controller does not need to rely on complex system model information, but only uses the input and output data of the system to realize a completely data-driven intelligent control strategy, and its distributed characteristics make it unnecessary to use all network resources, and only the communication of adjacent unmanned ships can be used to coordinate the consistency of all unmanned ships, saving communication resources and energy loss. It has significant advantages such as high computing efficiency, fast response speed, wide application range and good real-time performance, which significantly reduces the computing load, effectively saves energy consumption, and reduces the adverse effects of wave interference on ship movement.

[0049] In the specific implementation of S1: according to the initial position of each unmanned ship in the multi-unmanned ship cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked, the communication topology of the distributed collaborative control of the multi-unmanned ship cluster is constructed; specifically:

[0050] The present invention will Defined as the communication topology diagram between unmanned ships. and Represents nodes and edge sets respectively. Representation Node You can get the node If the unmanned ship Unmanned boats available The information transmitted , and also called unmanned ships For unmanned ships Neighbors, defining unmanned ships Neighborhood set . is the adjacency matrix, for The weight of ,but ,otherwise At the same time, we define . is defined as the Laplacian matrix, where It is a picture The in-degree matrix of The element is defined as all The sum of If there is at least one path between any two nodes, then the graph is strongly connected. In addition, node 0 is defined as the reference target. Unmanned ship Can the information of the target point to be tracked be directly obtained? The information of the reference target can be directly obtained. ,otherwise, For ease of understanding, some necessary symbols need to be explained. Definition and is the n-dimensional identity matrix and diagonal matrix. is the transpose of the matrix. and They are the absolute value operator of a scalar and the norm operator of a vector respectively.

[0051] In the specific implementation of S2: Based on unknown external interference and nonlinear uncertainty, a dynamic linear data model of the speed and heading system of a single unmanned ship with interference terms is constructed; specifically:

[0052] Considering the ship fixed coordinate system and the earth coordinate system, the origin of the ship fixed coordinate system is set as the center of the unmanned ship. For the horizontal motion, the kinematic and dynamic models can be described as:

[0053] (1)

[0054] in, is the unmanned ship in the geodetic coordinate system The first derivative of the X-axis coordinate; is the unmanned ship in the geodetic coordinate system The first derivative of the Y-axis coordinate; For unmanned ships Heading angle The first derivative of , and Unmanned Ship Surge speed , Swing speed and yaw speed The first derivative of . is a continuous smooth function. , and Unmanned Ship The speed thrust, rolling moment and heading torque. , and Unmanned Ship The disturbances suffered by surge, sway and yaw.

[0055] To simplify the unmanned ship The study of horizontal motion control, the present invention adopts the Euler discretization method to express the kinematic model of its horizontal motion as follows:

[0056] (2)

[0057] in, is the discrete sampling time, k is the moment; is the unmanned ship at time k Surge velocity; is the unmanned ship at time k Swing speed; is the unmanned ship at time k Yaw speed; is the unmanned ship in the geodetic coordinate system at time k The X-axis coordinate of is the unmanned ship in the geodetic coordinate system at time k The Y-axis coordinate of is the unmanned ship at time k The heading angle. , and are the unknown inertial parameters of surge, sway and yaw of the unmanned ship respectively. , and The unmanned boats at time k are The speed thrust, rolling moment and heading torque. , and The unmanned boats at time k are Unknown disturbances in surge, sway and yaw direction.

[0058] Under-actuated unmanned ship The swing moment is In order to simplify the analysis without losing its generality, we only consider the following unmanned ships with external interference Heading and speed system:

[0059] (3)

[0060] For a multi-input and multi-output nonlinear system,

[0061] (4)

[0062] in, , and Unmanned Ship The heading and speed of the system at time k are the output, control input and disturbance, and It is an unknown integer and can be set by yourself; is a nonlinear function. The perturbation of the system is bounded, that is, , is the limit value of the disturbance, And is a constant. Obviously, the unmanned ship of formula (3) The heading and speed system is the general form of equation (4).

[0063] Since the control input signal obtained by the control algorithm may exceed the physical limit of the unmanned ship actuator, that is, input saturation, this will bring disaster to the unmanned ship control system. The constraints of the actuator are as follows:

[0064] (5)

[0065] in, It is an unmanned ship without considering actuator constraints. The control input, It's an unmanned ship The upper limit of the actuator.

[0066] The present invention uses a compact dynamic linearization-model-free adaptive control method to estimate the unmanned ship The heading and speed system of the dynamic model of formula (3) is The heading system of the unmanned ship cannot directly meet the "quasi-linear" assumption requirement of model-free adaptive control. The linear sum of the heading and heading velocity (yaw velocity) is taken as a component of the system output, that is, ,in, is the unmanned ship heading yaw speed feedback gain, select the appropriate The yaw moment When it increases, the heading system output If the unmanned boat The ship speed (surge speed) is another component of the system output In this way, a compact dynamic linearized data model can be constructed according to the model-free adaptive control principle. Equation (4) can be transformed into the following compact dynamic linearized data model:

[0067] (6)

[0068] in, is the output increment, To control the input increment, is the unknown disturbance increment, where the unknown disturbance is the sum of external disturbance and nonlinear uncertainty; It's an unmanned ship The pseudo-Jacobian matrix of the dynamic linearization data model of the heading and speed system, and The signs of all elements in remain unchanged for any time instant k. It's an unmanned ship The pseudo-Jacobian matrix of the interference correlation is The signs of all elements in remain unchanged for any time instant k.

[0069] In the specific implementation of S3: a smooth anti-interference trajectory tracking algorithm for a single unmanned ship to track a dynamic target point is constructed to obtain the expected tracking speed, expected tracking heading and expected yaw speed of the single unmanned ship; specifically:

[0070] The trajectory tracking algorithm needs to track the dynamic target point under time constraints, while controlling the heading and speed to meet the requirements at a specific time point. The speed adjustment and safety distance setting are considered, and the trajectory accompanied by the speed under the influence of water flow is considered. Therefore, the following tracking law is adopted:

[0071] (7)

[0072] (8)

[0073] in, , and for Unmanned ship at all times The expected tracking speed, expected tracking heading and expected yaw speed; is the forward speed of the reference target at time k, For unmanned ships The maximum speed of For unmanned ships The distance to the target point to be tracked, For a safe distance, Parameters for adjusting the speed convergence rate; is the coordinate of the reference target at time k, It is the differential parameter in the proportional integral differential control, and the desired yaw speed is adjusted by controlling the differential term of the error of the desired tracking heading. As the tracking distance decreases to a safe distance, the speed of the ship gradually transitions to the speed of the target point, and at the same time, The item increases the convergence speed of the speed. When the heading of the target point changes, the unmanned ship needs to change its heading. Considering the existence of water flow interference, the speed will change due to the interference of water flow. The term is used to compensate and mitigate the impact of changes in water flow disturbance. It is used to control the smooth change of speed. If it is too small, the speed will change too suddenly. If it is too large, the speed will be significantly affected by the water flow. Let the expected output of the unmanned ship be .

[0074] In the specific implementation of S4: construct pseudo-Jacobian matrix estimation, disturbance estimation and distributed disturbance compensation controller for distributed model-free adaptive control of multiple unmanned ships; specifically:

[0075] First, define unmanned ships The consistency output error is

[0076] (9)

[0077] The design goal of the controller is to make the consistent output error of each unmanned ship converge with the increase of time. , the control performance is evaluated by minimizing the following function:

[0078] (10)

[0079] in, is the performance index function of the controller; is the unmanned ship at time k The consistency error is caused by nonlinear uncertainties such as unknown external disturbances and model parameter perturbations. are the parameters of the communication topology; is the weight coefficient. Take the partial derivative The extreme value of , the distributed disturbance compensation controller can be obtained as:

[0080] (11)

[0081] in yes The step size factor of , which makes the control algorithm design more flexible. It can be seen that the proposed distributed disturbance compensation controller contains parameters related to the communication topology , its purpose is to optimize the performance of the distributed disturbance compensation controller so that the compensation part of the controller can better compensate for unknown disturbances.

[0082] In formula (11), , and are unknown and cannot be used directly. Therefore, it is necessary to estimate them. The performance function of the unknown pseudo-Jacobian matrix is ​​defined as:

[0083] (12)

[0084] in, is the performance index function of the pseudo-Jacobian matrix; the weight factor , which is used to penalize large changes in the pseudo-Jacobian matrix estimate. is the pseudo-Jacobian matrix at time k-1 The estimated value of . Minimizing the performance function (12) yields the pseudo-Jacobian matrix The estimated value is

[0085] (13)

[0086] The pseudo-Jacobian matrix can also be obtained Estimated value of

[0087] (14)

[0088] Estimated value of

[0089] (15)

[0090] in, is the step size factor, which provides greater flexibility in the design of the control algorithm; is the weight factor; is the output of the dynamic linearization data model An estimated value of is the pseudo-Jacobian matrix at time k An estimated value of is the pseudo-Jacobian matrix at time k An estimated value of is the interference term at time k The estimated value of .

[0091] In the specific implementation of S5: a reset algorithm is constructed based on the pseudo-Jacobian matrix estimation obtained in S4, and the distributed disturbance compensation controller is updated to obtain the actual controller of the unmanned ship that meets the physical requirements; specifically:

[0092] In order to ensure that the pseudo-Jacobian matrix estimates the parameters and Ability to track actual and , the following reset algorithm is given: The element in row m and column n of The absolute value of the element exceeds the set threshold, or its sign is inconsistent with the initial value of the element. The sign of is opposite, then Reset to ;

[0093] when The element in row m and column n of The absolute value of the element exceeds the threshold, or its sign is different from the initial value of the element. The sign of is opposite, then Reset to ;

[0094] in, It is initial , It is initial ; The algorithm reset mechanism is introduced to make the pseudo-Jacobian matrix estimation algorithm better able to track time-varying parameters.

[0095] Substituting the estimated parameters into equation (11), we get the control input:

[0096] (16)

[0097] in, is the control input of the model-free adaptive control based on the estimation algorithm mentioned in equation (5); Finally, the unmanned boat is The actual control input is obtained by the actuator constraints .

[0098] In the specific implementation of S6: based on the expected tracking speed, expected tracking heading and expected yaw speed obtained in S3 (i.e., equations (7) and (8)), the pseudo-Jacobian matrix estimation and interference estimation obtained in S4 (i.e., equations (13), (14) and (15)), and the reset algorithm and actual controller obtained in S5 (i.e., equation (16)), distributed collaborative trajectory tracking control is performed on the multi-unmanned ship cluster.

[0099] The present invention adopts the trajectory tracking algorithm, estimation algorithm and distributed model-free adaptive control to perform collaborative tracking control on a cluster of multiple unmanned ships.

[0100] In a specific embodiment of the present invention, the S6 is: based on the initial position of the unmanned ship , the initial position of the reference target and initial velocity Get the expected tracking speed and the desired tracking heading , and then the expected yaw speed is obtained from the expected tracking heading error , we can get the overall expected output of the system Then, based on the actual situation, , and Substituting into equation (9), we get the consistent output error , then set the initial value , and ,Will , and Substituting the value of into equation (13), equation (14) and equation (15), we get , and , based on , , and From formula (16), we can get ,Will Substituting the value of into formula (5) we get , input into the unmanned ship actuator for control.

[0101] Based on the actual situation at the current moment and the partial values ​​obtained, the above steps are used to obtain the navigation speed and heading angle of the unmanned ship at the new moment, until the navigation speed and heading angle of the unmanned ship reach the cluster coordination consistency, and the multi-unmanned ship cluster system coordinated tracking control is completed. If there is no subsequent task requirement, the unmanned ship will maintain the cluster coordinated formation to operate; if there is a new task requirement, the unmanned ship will perform cluster system coordinated tracking control according to the new reference target.

[0102] The effectiveness of the proposed method of the present invention is verified through simulation;

[0103] Unmanned ship model parameters used in the simulation experiment , , , , , .

[0104] The initial positions of the four unmanned ships are , , , The initial heading angle and speed are , , , The controller parameters are , , , , , , , , , . Set the reference target speed to , the initial position is , the trajectory is circular motion, safe distance Set as Considering the disturbance in the surge direction and the disturbance in the yaw direction is . Set the initial value , , The initial speed, initial thrust, and initial steering torque of all unmanned ships are set to 0. The communication topology is shown in the figure below. Figure 1 As shown, node 0 is the reference target, and nodes 1-4 represent unmanned ships 1-4. Figure 1 Knowable , , , , , , , , , , , . , , , .

[0105] The simulation results of the proposed data-driven multi-unmanned ship swarm collaborative tracking control method for tracking dynamic targets under complex conditions with current interference are shown in Figure 2. Figure 3 As shown. Figure 3 It can be seen that under the complex conditions of ocean current interference, the proposed data-driven multi-unmanned ship cluster collaborative tracking control method still has a good control effect. The unmanned ship cluster collaboratively tracks the target in formation and maintains a certain safety distance. Figure 4 It is the heading consistency error of multiple unmanned ship clusters. Figure 5 is the speed consistency error of multiple unmanned ship clusters. Figure 4 and Figure 5 It can be seen that as time goes by, the consistency error of the heading and speed of the multi-unmanned ship cluster converges to a bounded value close to 0, achieving consistency of heading and speed, with good coordination and meeting the design requirements. Figure 6 It is the change of the speed of multiple unmanned ship clusters. It can be seen that with the increase of time, the speed of the unmanned ships first gradually increases to a speed greater than the reference target (that is, the distance between the unmanned ships and the reference target is far, much greater than the safe distance), and then gradually approaches the speed of the reference target (that is, the distance between the unmanned ships and the reference target is close, gradually approaching the safe distance), maintaining formation navigation at a safe distance, which is in line with the design goal of our smooth anti-interference trajectory tracking algorithm. Figure 7 and Figure 8 They are the thrust and steering torque of multiple unmanned ship actuators. It can be seen that the changes of the unmanned ship actuators are always within a reasonable range, and the thrust and steering torque are adjusted in real time to cope with the interference of surge and yaw direction currents, thereby improving system stability. Fig. 9 is the interference estimation error in the surge direction of each unmanned ship. It can be seen that the interference estimation error in the surge direction of each unmanned ship gradually converges to near zero, which has a good estimation effect. Fig.10 is the disturbance estimation error in the yaw direction of each unmanned ship. It can be seen that the disturbance estimation error in the yaw direction of each unmanned ship gradually converges to near zero, indicating that the proposed method is reasonable. The above results show that the proposed data-driven multi-unmanned ship cluster collaborative tracking control method is feasible and effective.

[0106] The multi-unmanned ship cluster cooperative tracking controller developed by the present invention breaks through the limitations of traditional control methods. The distributed controller does not need to rely on complex system model information, but only uses the input and output data of the system to realize a completely data-driven intelligent control strategy. Its distributed characteristics make it unnecessary to use all network resources, and only use the communication of adjacent unmanned ships to coordinate the consistency of all unmanned ships, saving communication resources and energy loss. It has significant advantages such as high computing efficiency, fast response speed, wide application range and good real-time performance, which significantly reduces the computing load and effectively saves energy consumption. Its anti-interference compensation part can effectively reduce the adverse effects of wave interference on ship movement, providing key technical support for the stable operation of multi-unmanned ship cluster cooperative tracking.

[0107] In this embodiment, a data-driven multi-unmanned ship cluster collaborative tracking control system is also provided, including:

[0108] A communication topology building module is used to build a communication topology for distributed collaborative control of a multi-unmanned ship cluster based on the initial position of each unmanned ship in the multi-unmanned ship cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked;

[0109] A model acquisition module is used to construct a dynamic linear data model of the unmanned ship's speed and heading system with interference terms based on unknown external interference and nonlinear uncertainty;

[0110] A pseudo-Jacobian matrix and controller acquisition module, which is used to obtain the pseudo-Jacobian matrix and distributed disturbance compensation controller of the model based on the model-free adaptive control principle;

[0111] A tracking module is used to obtain reference target state information and unmanned ship state information to design the expected output of unmanned ship trajectory tracking control;

[0112] An estimation module, which is used to estimate the pseudo-Jacobian matrix through an estimation algorithm, and to estimate unknown external interference and nonlinear uncertainty;

[0113] An updating module, which updates the distributed disturbance compensation controller based on the results obtained by the tracking module and the estimation module;

[0114] The control module converts the distributed disturbance compensation controller obtained by the update module into a controller of an unmanned ship that actually meets the physical requirements, and performs distributed collaborative tracking control on a cluster of multiple unmanned ships.

[0115] For the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment, and the implementation methods of the remaining modules will not be repeated here. The system embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of the present invention. Ordinary technicians in this field can understand and implement it without paying creative work.

[0116] The embodiments of the system of the present invention can be applied to any device with data processing capabilities, and the device with data processing capabilities can be a device or apparatus such as a computer. The system embodiments can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, the corresponding computer program instructions in the non-volatile memory are read into the memory by the processor of any device with data processing capabilities and run.

[0117] An embodiment of the present invention further provides an electronic device, including a memory and a processor;

[0118] The memory is used to store computer programs;

[0119] The processor is used to implement the above-mentioned data-driven multi-unmanned ship cluster collaborative tracking control process when executing the computer program.

[0120] An embodiment of the present invention also provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the process of the above-mentioned data-driven multi-unmanned ship cluster collaborative tracking control method is implemented.

[0121] Obviously, the above-described embodiments and drawings are only some examples of the present application. For those of ordinary skill in the art, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that, although the work done in this development process may be complicated and lengthy, for those of ordinary skill in the art, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application. Without departing from the concept of the present application, several variations and improvements can also be made, which all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.

Claims

1. A data-driven multi-unmanned ship cluster collaborative tracking control method, characterized in that: include: 1) According to the initial position of each unmanned ship in the multi-unmanned ship cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked, the communication topology of the distributed collaborative control of the multi-unmanned ship cluster is constructed; 2) Based on unknown external disturbances and nonlinear uncertainties, a dynamic linearized data model of the speed and heading system of a single unmanned ship with disturbance terms is constructed; the dynamic linearized data model of the speed and heading system of a single unmanned ship with disturbance terms is expressed as: ； in, is the output increment, For unmanned ships The speed and heading system The output of the moment, For unmanned ships The heading system is The output of the moment, For unmanned ships The speed system is Output at the moment; It's an unmanned ship The pseudo-Jacobian matrix of the heading and speed system dynamic linearization data model, It's an unmanned ship The pseudo-Jacobian matrix of the interference correlation; To control the input increment, For unmanned ships The speed and heading system Control input at all times; is the unknown interference increment, For unmanned ships The speed and heading system The unknown disturbances are the sum of external disturbances and nonlinear uncertainties. 3) Construct a smooth anti-interference trajectory tracking algorithm for a single unmanned ship to track a dynamic target point, and obtain the expected tracking speed, expected tracking heading and expected yaw speed of the single unmanned ship; 4) Construct pseudo-Jacobian matrix estimation, disturbance estimation and distributed disturbance compensation controller for distributed model-free adaptive control of multiple unmanned ships; 5) Based on the pseudo-Jacobian matrix estimation obtained in step 4), a reset algorithm is constructed to update the distributed disturbance compensation controller to obtain the actual controller of the unmanned ship that meets the physical requirements; 6) Based on the expected tracking speed, expected tracking heading and expected yaw speed obtained in step 3), the pseudo-Jacobian matrix estimation and interference estimation obtained in step 4), and the reset algorithm and actual controller obtained in step 5), distributed collaborative tracking control is performed on the multi-unmanned ship cluster to achieve collaborative tracking of the target point to be tracked by the multi-unmanned ship cluster, and the heading and speed of each unmanned ship are consistent with the target point to be tracked.

2. The method according to claim 1, characterized in that In step 1), the communication topology of the distributed collaborative control of multiple unmanned ship clusters is expressed as ;in, Represents the communication topology The nodes in , i.e., each unmanned ship in the multi-unmanned ship cluster; represents the edge set, if the unmanned ship Capable of acquiring unmanned ships The information transmitted is expressed as , and also called unmanned ships For unmanned ships Neighbors, defining unmanned ships Neighborhood set ; is the adjacency matrix, for The weight of ,but ,otherwise At the same time, define ; In addition, node 0 is defined to represent the target point to be tracked. Unmanned ship Can the information of the target point to be tracked be obtained directly? Can directly obtain the information of the target point to be tracked, then ,otherwise, .

3. The method according to claim 2, characterized in that In step 3), the construction of a smooth anti-interference trajectory tracking algorithm for a single unmanned ship to track a dynamic target point includes: Consider the unmanned ship approaching the target point to be tracked The speed adjustment and safety distance setting are adopted to design the expected tracking speed, expected tracking heading and expected yaw speed of the single unmanned ship using the following tracking law; ； ； in, , and for Unmanned ship at all times The expected tracking speed, expected tracking heading and expected yaw speed; for Always refer to the target's forward speed. For unmanned ships The maximum speed of For unmanned ships The distance to the target point to be tracked, For a safe distance, To adjust the parameters of the speed convergence rate, for The unmanned ship in the geodetic coordinate system at the time of Axis coordinates; for The unmanned ship in the geodetic coordinate system at the time of Axis coordinates; for Always refer to the coordinates of the target, is the differential parameter in proportional integral differential control; let the expected output of the unmanned ship be ,in is the gain of yaw speed.

4. The method according to claim 3, characterized in that Step 4) includes: Defining unmanned ships The consistency output error is: ； For unmanned ships , the control performance is evaluated by minimizing the following function: ； in, is the performance index function of the controller; for Unmanned ship at the moment Consistency error for nonlinear uncertainties; are the parameters of the communication topology; is the weight coefficient; take the partial derivative The extreme value of , the distributed disturbance compensation controller is: ； in, for The actual controller at the moment; is the controller update coefficient of unmanned ship i, for The step factor, ; is the pseudo-Jacobian matrix The estimated value of is the pseudo-Jacobian matrix The estimated value of is the unknown interference increment The estimated values ​​are: ； ； ； Among them, the step factor and step factor , is the weight factor; is the output of the dynamic linearization data model The estimated value of .

5. The method according to claim 4, characterized in that The step 5) comprises: The pseudo-Jacobian matrix at time Estimated value of ; The pseudo-Jacobian matrix at time Estimated value of ; when The element in row m and column n of The absolute value of the element exceeds the set threshold, or its sign is inconsistent with the initial value of the element. The sign of is opposite, then Reset to ; when The element in row m and column n of The absolute value of the element exceeds the set threshold, or its sign is inconsistent with the initial value of the element. The sign of is opposite, then Reset to ; Finally, the distributed disturbance compensation controller described in step 4) Substitute unmanned ship The constraints of the actuators are used to obtain the actual controller of the unmanned ship that meets the physical requirements: ； in, It's an unmanned ship The upper limit of the actuator.

6. A multi-unmanned ship cluster cooperative tracking control system for implementing the method described in claim 1, characterized in that: include: A communication topology building module is used to build a communication topology for distributed collaborative control of a multi-unmanned ship cluster based on the initial position of each unmanned ship in the multi-unmanned ship cluster, the master-slave relationship, and whether each unmanned ship can obtain the status information of the target point to be tracked; The model acquisition module is used to construct a dynamic linearized data model of the unmanned ship speed and heading system with interference terms based on unknown external interference and nonlinear uncertainty; the dynamic linearized data model of the speed and heading system of a single unmanned ship with interference terms is expressed as: ； in, is the output increment, For unmanned ships The speed and heading system The output of the moment, For unmanned ships The heading system is The output of the moment, For unmanned ships The speed system is Output at the moment; It's an unmanned ship The pseudo-Jacobian matrix of the heading and speed system dynamic linearization data model, It's an unmanned ship The pseudo-Jacobian matrix of the interference correlation; To control the input increment, For unmanned ships The speed and heading system Control input at all times; is the unknown interference increment, For unmanned ships The speed and heading system The unknown disturbances are the sum of external disturbances and nonlinear uncertainties. A pseudo-Jacobian matrix and controller acquisition module, which is used to obtain the pseudo-Jacobian matrix and distributed disturbance compensation controller of the model based on the model-free adaptive control principle; The tracking module is used to obtain the state information of the target point to be tracked and the state information of the unmanned ship to design the expected output of the unmanned ship trajectory tracking control; An estimation module, which is used to estimate the pseudo-Jacobian matrix through an estimation algorithm, and to estimate unknown external interference and nonlinear uncertainty; An updating module, which updates the distributed disturbance compensation controller based on the results obtained by the tracking module and the estimation module; The control module converts the distributed disturbance compensation controller obtained by the update module into a controller of an unmanned ship that actually meets the physical requirements, and performs distributed collaborative tracking control on a cluster of multiple unmanned ships.

7. An electronic device, characterized in that: It includes a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the data-driven multi-unmanned ship cluster collaborative tracking control method as described in any one of claims 1 to 5.

8. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores machine-executable instructions, which, when called and executed by a processor, are used to implement the data-driven multi-unmanned ship cluster collaborative tracking control method as described in any one of claims 1 to 5.

Images