Multi-unmanned ship surrounding method based on dual-radar positioning and exploration excitation function

Abstract

The invention discloses a multi-unmanned-ship surrounding method based on double-radar positioning and exploration excitation functions, and the method comprises the following steps: S1, obtaining the motion parameters of surrounding ships at the current moment, and determining the position information of an escape target through double-radar positioning; s2, designing an exploration excitation function model, and obtaining the course, the navigational speed, the position and the expected position of the surrounding ship at the next moment through the motion parameters of the surrounding ship at the current moment and the position information of the escape target at the current moment; and S3, according to the course, the navigational speed, the position and the expected position of the surrounding ship at the next moment, judging whether the surrounding task is ended or not. According to the invention, through double-radar sensing and a specific exploration excitation function, an intelligent decision strategy is output to guide the unmanned ship to complete a surrounding task, so that the positioning speed and the damage resistance of the system are improved, and the surrounding efficiency and success rate are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of radar and marine unmanned systems, and in particular relates to a multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function. Background Art

[0002] With the continuous innovation of electronic information technology, the research on multi-unmanned ship coordination technology has achieved rapid development and has become a cutting-edge technology that countries around the world are competing to promote. Combining the advantages of group intelligence coordination and single-ship autonomous execution, multi-unmanned ship coordination has high flexibility, good maneuverability, high task execution rate, and a wide range of operations, meeting the application requirements of regional search, sea patrol, coordinated encirclement, cluster optimization scheduling, etc. The multi-unmanned ship system forms an encirclement formation through mutual cooperation between encircling ships, reducing the chance of escape of the fleeing target and reducing energy consumption.

[0003] The positioning accuracy of a single radar is low and its anti-interference ability is poor. At the same time, the existing collaborative encirclement methods of unmanned ships mostly rely on traditional control theory, including virtual structure method, pilot following method, etc. The above methods have poor scalability. As the number of unmanned ships increases, the calculation complexity increases nonlinearly, and the response strategy for flexible escape targets is relatively lagging, and the encirclement efficiency is low. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function. Through dual radar perception and specific exploration excitation function, an intelligent decision-making strategy is output to guide the unmanned ship to complete the encirclement task, improve the positioning speed and system anti-damage capability, and improve the encirclement efficiency and success rate.

[0005] In order to achieve the purpose of the present invention, the present invention provides a multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function, comprising the following steps:

[0006] S1. Obtain the motion parameters of the surrounding ship at the current moment, and determine the location information of the escape target through dual radar positioning;

[0007] S11, determine the distance between the first encircling ship A and the second encircling ship B at the current moment :

[0008] ;

[0009] in, is the position of the first encircling ship A at the current moment, is the current position of the second encircling ship B, for the current moment;

[0010] S12: Determine the angle of the escape target at the current moment relative to the first encircling ship A. :

[0011] ;

[0012] in, The distance at which the radar on the first encircling ship A detects the fleeing target, The distance at which the radar on the second encircling ship B detected the fleeing target;

[0013] S13, using the first encircling ship A as the dual radar positioning master node, calculating the current escape target position :

[0014] .

[0015] S2. Construct an exploration incentive function model to obtain the next-moment heading, speed, position, and expected position of the surrounding ship through the motion parameters of the surrounding ship at the current moment and the position information of the escape target at the current moment;

[0016] S21. Determine the network model and enclose the state matrix :

[0017] ;

[0018] in, for Moment The speed of the encircling ship, for Moment The encircling ship is heading, for Moment The horizontal coordinate of the Cartesian coordinate system of the surrounding ships, for Moment The Cartesian coordinates of the surrounding ships are: is the total number of encircling ships;

[0019] S22. Determine the expected position of each encircling ship :

[0020] ;

[0021] in, is the enclosing radius, , is the adjustment factor, is the radius of the surrounding ship; when The larger the value, the farther the target point of the encircling ship is from the escaping target, and the larger the encirclement, otherwise the smaller the encirclement;

[0022] S23, determine The relative distance between the encircling ships and the desired encirclement position , Set the single ship exploration incentive function :

[0023] ;

[0024] ;

[0025] in, is the single ship exploration incentive value, is the encirclement coefficient in the current encirclement scene;

[0026] S24. Set cooperative ship exploration incentive function :

[0027] ;

[0028] in, , , is the weight parameter that affects the encirclement efficiency;

[0029] S25, based on the single ship exploration incentive function and multi-ship exploration incentive function Set up the training network structure, and get the next position of each encircling ship after inputting the state matrix ;

[0030] S251, set the exploration incentive function and multi-ship exploration incentive function is the initial condition of the network model;

[0031] S252: Input a 1x6 state matrix to the model , through the fully connected layer function, the 1x128 first transformation matrix is ​​obtained;

[0032] S253, the 1x128 first conversion matrix is ​​activated by a RELU function to obtain a 1x64 second conversion matrix;

[0033] S254: The 1x64 second conversion matrix is ​​activated by a tanh function to output a 1x2 action matrix. , after execution, the next moment position is obtained .

[0034] S3, judging whether the encirclement task is completed according to the next course, speed, position and expected position of the encircling ship; when the position error is less than or equal to the radius of the encircling ship, the encirclement is successful; when the position error is greater than the radius of the encircling ship, the encirclement is unsuccessful, and the steps S2 and S3 are sequentially executed until the encirclement is successful and the encirclement is ended;

[0035] According to the next moment position With the desired position Make a judgment to get the position error :

[0036] ;

[0037] when When If , the encirclement is unsuccessful, and S2 and S3 are executed in sequence until the encirclement is successful.

[0038] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for surrounding multiple unmanned ships based on dual radar positioning and exploration excitation function is implemented.

[0039] A computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function is implemented.

[0040] Compared with the prior art, the significant progress of the present invention is: 1. The dual radar positioning of the present invention can provide a wide detection distance for unmanned ships, and compared with single radar positioning, it improves the positioning accuracy and positioning speed of the escaping target; 2. The exploration incentive function of the present invention can improve the convergence speed in task decision-making, solve the problem of large computational complexity of a large number of unmanned ship clusters, and the constraints on the decision to surround the ship are in line with reality, which flexibly improves the efficiency and success rate of the encirclement.

[0041] In order to more clearly illustrate the functional characteristics and structural parameters of the present invention, further description is given below in conjunction with the accompanying drawings and specific implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0043] Figure 1 is a flow chart of the steps of the present invention;

[0044] Figure 2 It is a multi-unmanned ship cooperative encirclement model of the present invention;

[0045] Figure 3 is a schematic diagram of dual radar collaborative positioning of the present invention;

[0046] Figure 4 It is the encirclement training model of the present invention. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0048] The present invention is applied in the maritime attack and defense confrontation scenario, where the encircled area is an open sea area. Figure 2 The positions of the encircling ships are , and their expected target points are , To escape the target location; To ensure the formation of an encirclement formation, the number of encircling ships ,Select any two unmanned ships to deploy navigation radars, denoted as the first encirclement ship A and the second encirclement ship B.

[0049] The invention discloses a multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function, combined with Figure 1 , including the following steps:

[0050] S1. Obtain the motion parameters of the surrounding ship at the current moment, and determine the location information of the escape target through dual radar positioning;

[0051] S11, determine the distance between the first encircling ship A and the second encircling ship B at the current moment :

[0052] ;

[0053] in, is the position of the first encircling ship A at the current moment, is the current position of the second encircling ship B, for the current moment;

[0054] S12: Determine the angle of the escape target at the current moment relative to the first encircling ship A. :

[0055] ;

[0056] in, The distance at which the radar on the first encircling ship A detects the fleeing target, The distance at which the radar on the second encircling ship B detected the fleeing target;

[0057] S13, Combination Figure 3 , taking the first encircling ship A as the dual radar positioning master node, calculate the current escape target position :

[0058] .

[0059] S2. Construct an exploration incentive function model to obtain the next-moment heading, speed, position, and expected position of the surrounding ship through the motion parameters of the surrounding ship at the current moment and the position information of the escape target at the current moment;

[0060] S21. Determine the network model and enclose the state matrix :

[0061] ;

[0062] in, for Moment The speed of the encircling ship, for Moment The encircling ship is heading, for Moment The horizontal coordinate of the Cartesian coordinate system of the surrounding ships, for Moment The Cartesian coordinates of the surrounding ships are: is the total number of encircling ships;

[0063] S22. Determine the expected position of each encircling ship :

[0064] ;

[0065] in, is the enclosing radius, , is the adjustment factor, is the radius of the surrounding ship; when The larger the value, the farther the target point of the encircling ship is from the escaping target, and the larger the encirclement, otherwise the smaller the encirclement;

[0066] S23, determine The relative distance between the encircling ships and the desired encirclement position , Set the single ship exploration incentive function :

[0067] ;

[0068] ;

[0069] in, is the single ship exploration incentive value, is the encirclement coefficient in the current encirclement scene;

[0070] S24. Set cooperative ship exploration incentive function :

[0071] ;

[0072] in, , , is the weight parameter that affects the encirclement efficiency;

[0073] S25, Combination Figure 4 , according to the single ship exploration incentive function and multi-ship exploration incentive function Set up the training network structure, and get the next position of each encircling ship after inputting the state matrix ;

[0074] S251, set the exploration incentive function and multi-ship exploration incentive function is the initial condition of the network model;

[0075] S252: Input a 1x6 state matrix to the model , through the fully connected layer function, the 1x128 first transformation matrix is ​​obtained;

[0076] S253, the 1x128 first conversion matrix is ​​activated by a RELU function to obtain a 1x64 second conversion matrix;

[0077] S254: The 1x64 second conversion matrix is ​​activated by a tanh function to output a 1x2 action matrix. , after execution, the next moment position is obtained .

[0078] S3, judging whether the encirclement task is completed according to the next course, speed, position and expected position of the encircling ship; when the position error is less than or equal to the radius of the encircling ship, the encirclement is successful; when the position error is greater than the radius of the encircling ship, the encirclement is unsuccessful, and the steps S2 and S3 are sequentially executed until the encirclement is successful and the encirclement is ended;

[0079] According to the next moment position With the desired position Make a judgment to get the position error :

[0080] ;

[0081] when When If , the encirclement is unsuccessful, and S2 and S3 are executed in sequence until the encirclement is successful.

[0082] This embodiment uses MALTAB to conduct computer simulation research, and the above parameters are as follows:

[0083] The encirclement area is 10nm*10nm, and the number of encirclement ships is 3; the initial position of the first encirclement ship A is (2000, 4000); the initial position of the second encirclement ship B is (4000, 2000); the initial position of the encirclement ship C is (15000, 4000); the initial position of the escape target is (6000, 6000) (unit: meters);

[0084] Surrounding radius of unmanned ship (Unit: meter);

[0085] Adjustment factor ; Enclosure radius (unit: meter);

[0086] In order to meet the requirements of small encirclement formation, reasonable distance between ships and convergence speed in the encirclement area of ​​10nm, the distance between each encircling ship and its expected target point is used as the activation function of the single ship exploration. As a reference for the design of the excitation function, set , ;

[0087] In order to meet the requirements of convergence speed, learning efficiency and encirclement success rate within the 10nm encirclement area, the weight parameter is set in the cooperative ship exploration incentive function. ;

[0088] Under the above known conditions, taking the first encircling ship A as an example, the above S1 to S3 is an iterative process, and the distance error data after the 30th, 60th, 90th, 120th, 150th, and 193rd iterations are recorded as follows. It can be seen that at the 193rd iteration, the distance error of the first encircling ship A is smaller than the radius of the unmanned ship. When the distance errors of the three encircling ships A, B, and C all meet the encirclement success conditions, the encirclement is successful:

[0089] Time / second 30 60 90 120 150 180 193 Distance error / m 4517.7 4272.8 2763.1 1367.9 320.8 70.6 48.3

[0090] The simulation results show that the multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function has achieved better encirclement efficiency and encirclement effect.

[0091] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0092] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function, characterized in that: The following steps are involved: S1. Obtain the motion parameters of the surrounding ship at the current moment, and determine the location information of the escape target through dual radar positioning; S2. Construct an exploration incentive function model to obtain the next-moment heading, speed, position, and expected position of the surrounding ship through the motion parameters of the surrounding ship at the current moment and the position information of the escape target at the current moment; S3. Determine whether the encirclement mission is completed based on the next heading, speed, position, and expected position of the encircling ship; when the position error is less than or equal to the radius of the encircling ship, the encirclement is successful; when the position error is greater than the radius of the encircling ship, the encirclement is unsuccessful, and continue to execute steps S2 and S3 in sequence until the encirclement is successful and the encirclement is ended.

2. The method for encircling multiple unmanned ships based on dual radar positioning and exploration excitation function according to claim 1 is characterized in that: The specific implementation process of step S1 is as follows: S11, determine the distance between the first encircling ship A and the second encircling ship B at the current moment : ； in, is the position of the first encircling ship A at the current moment, is the current position of the second encircling ship B, for the current moment; S12: Determine the angle of the escape target at the current moment relative to the first encircling ship A. : ； in, The distance at which the radar on the first encircling ship A detects the fleeing target, The distance at which the radar on the second encircling ship B detected the fleeing target; S13, using the first encircling ship A as the dual radar positioning master node, calculating the current escape target position : 。 3. The method for encircling multiple unmanned ships based on dual radar positioning and exploration excitation function according to claim 2 is characterized in that: The specific implementation process of step S2 is as follows: S21. Determine the network model and enclose the state matrix : ； in, for Moment The speed of the encircling ship, for Moment The encircling ship is heading, for Moment The horizontal coordinate of the Cartesian coordinate system of the surrounding ships, for Moment The Cartesian coordinates of the surrounding ships are: is the total number of encircling ships; S22. Determine the expected position of each encircling ship : ； in, is the enclosing radius, , is the adjustment factor, is the radius of the surrounding ship; when The larger the value, the farther the target point of the encircling ship is from the escaping target, and the larger the encirclement, otherwise the smaller the encirclement; S23, determine The relative distance between the encircling ships and the desired encirclement position , Set the single ship exploration incentive function : ； ； in, is the single ship exploration incentive value, is the encirclement coefficient in the current encirclement scene; S24. Set cooperative ship exploration incentive function : ； in, , , is the weight parameter that affects the encirclement efficiency; S25, based on the single ship exploration incentive function and multi-ship exploration incentive function Set up the training network structure, and get the next position of each encircling ship after inputting the state matrix .

4. The method for surrounding multiple unmanned ships based on dual radar positioning and exploration excitation function according to claim 3 is characterized in that: The step S25 comprises the following steps: S251, set the exploration incentive function and multi-ship exploration incentive function is the initial condition of the network model; S252: Input a 1x6 state matrix to the model , through the fully connected layer function, the 1x128 first transformation matrix is ​​obtained; S253, the 1x128 first conversion matrix is ​​activated by a RELU function to obtain a 1x64 second conversion matrix; S254: The 1x64 second conversion matrix is ​​activated by a tanh function to output a 1x2 action matrix. , after execution, the next moment position is obtained .

5. The method for encircling multiple unmanned ships based on dual radar positioning and exploration excitation function according to claim 4 is characterized in that: The specific implementation process of step S3 is as follows: According to the next moment position With the desired position Make a judgment to get the position error : ； when When If , the encirclement is unsuccessful, and S2 and S3 are executed in sequence until the encirclement is successful.

6. The method for encircling multiple unmanned ships based on dual radar positioning and exploration excitation function according to claim 3 is characterized in that: The adjustment factor is 5.

7. The multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function according to claim 3 is characterized in that: The single-vessel exploration incentive value is 10, the encirclement coefficient of the current encirclement scene is 100.

8. The method for encircling multiple unmanned ships based on dual radar positioning and exploration excitation function according to claim 3 is characterized in that: The weight parameter .

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, a multi-unmanned ship encirclement method based on dual radar positioning and exploration excitation function as described in any one of claims 1-8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a multi-unmanned ship surrounding method based on dual radar positioning and exploration excitation function as described in any one of claims 1-8 is implemented.

Images