Unmanned boat collaborative hunting method and device based on improved whale algorithm

Abstract

The invention provides an unmanned ship collaborative hunting method and device based on an improved whale algorithm. According to coordinate values of an unmanned ship, a target and an obstacle in a hunting space, hunting income models of various strategy types of the unmanned ship and anti-hunting income models of various strategy types of the target are established; a total income model of the unmanned ship is established through all the encircling income models, and a total escape risk model of the target is established through all the anti-encircling income models; establishing a whale algorithm model, solving the strategy action of the unmanned ship through the whale algorithm model, and carrying out iterative optimization on the solved strategy action of the unmanned ship according to the total income model and the total escape risk model to obtain an optimal strategy action; the whale algorithm is applied to the hunting action of the unmanned ship, so that the hunting efficiency and success rate of the unmanned ship are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned boat cooperative capture, and in particular to an unmanned boat cooperative capture method and device based on an improved whale algorithm. Background Art

[0002] Unmanned boats are surface vessels that can operate autonomously and intelligently, and are the inevitable product of the trend of unmanned ship development. With its flexible autonomy, cost savings, and ability to perform special tasks, unmanned boats have become a research hotspot in the field of ships at home and abroad, and the research results have been widely used in many fields. However, single unmanned boats have defects such as low work efficiency and limited mission types, which makes it impossible for a single boat to meet the needs of more complex time-limited missions.

[0003] The multi-unmanned boat system can improve the task execution capability and system fault tolerance through multi-boat task allocation, communication interconnection and collaborative control, which is an important direction for the development of unmanned boats. Multi-unmanned boats usually require collaboration between multiple boats. According to the current state of the unmanned boats, the targets and distances of the unmanned boats, the collaboration between multiple unmanned boats can ultimately achieve the encirclement of the target.

[0004] Among them, the unmanned boat capture problem faces many difficulties. For example, how to design an effective collaborative capture strategy, how to improve the efficiency and success rate of the unmanned boat collaborative capture. Therefore, a more efficient and intelligent unmanned boat collaborative capture method is needed to solve these problems.

[0005] In view of this, overcoming the defects of the prior art is an urgent problem to be solved in the field of this technology. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an effective collaborative capture strategy to improve the efficiency and success rate of collaborative capture by unmanned boats.

[0007] The present invention adopts the following technical solution:

[0008] In a first aspect, a method for cooperative capture of unmanned boats based on an improved whale algorithm is provided, comprising:

[0009] Projecting the capture space into a coordinate system to represent coordinate values ​​of different positions in the capture space;

[0010] According to the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space, a capture benefit model between different strategic actions and capture benefits in various strategic types of the unmanned boat is established; according to the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space, an anti-encirclement benefit model between different strategic actions and anti-encirclement benefits in various strategic types of the target is established;

[0011] The total benefit model of the unmanned boat is established through all the capture benefit models, and the total escape risk model of the target is established through all the anti-capture benefit models;

[0012] A whale algorithm model is established, and the strategic actions of the unmanned boat are solved by the whale algorithm model. The solved strategic actions of the unmanned boat are iteratively optimized according to the total benefit model and the total escape risk model to obtain the optimal strategic actions.

[0013] Preferably, the projecting of the capture space into a coordinate system to represent the coordinate values ​​of different positions in the capture space specifically includes:

[0014] The capture space is flattened and discretized into a grid network;

[0015] Projecting the grid network into a coordinate system, and corresponding the vertices of each grid area in the grid network to coordinate values ​​in the coordinate system;

[0016] According to the positions of the unmanned boat, the target and the obstacle in the grid network, the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space are obtained respectively.

[0017] Preferably, the method of establishing a capture benefit model between different strategic actions and capture benefits in various strategy types of the unmanned boat based on the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space specifically includes:

[0018] According to the coordinate values ​​of the unmanned boats, the target and the obstacle in the capture space, the distance between each unmanned boat, the distance between each unmanned boat and the target, the distance between each unmanned boat and the obstacle, and the angle between two adjacent unmanned boats and the target are respectively obtained;

[0019] The collision avoidance benefit model in the capture benefit model is obtained according to the distances between the unmanned boats. The expression of the collision avoidance benefit model is:

[0020]

[0021] Among them, L i,j is the distance between the i-th unmanned boat and the j-th unmanned boat, is the collision avoidance benefit of the i-th unmanned boat, L2 is the safe distance between unmanned boats, and n is the total number of unmanned boats;

[0022] The distance benefit model in the capture benefit model is obtained according to the distances between each unmanned boat and the target. The expression of the distance benefit model is:

[0023]

[0024] Among them, L1 is the safe distance between the unmanned boat and the target, L i,goal is the distance between the i-th unmanned boat and the target, is the capture benefit generated by the distance between the i-th unmanned boat and the target;

[0025] The obstacle avoidance benefit model in the capture benefit model is obtained according to the distance between each unmanned boat and the obstacle. The expression of the obstacle avoidance benefit model can be:

[0026]

[0027] in, is the distance between the ith unmanned boat and the ith obstacle, is the obstacle avoidance benefit between the i-th unmanned boat and the i-th obstacle, and L4 is the safe distance between the unmanned boat and the obstacle;

[0028] The team benefit model in the capture benefit model is obtained according to the angle between the two adjacent unmanned boats and the target. The expression of the team benefit model can be:

[0029]

[0030] Among them, θ is the angle between the i-th unmanned boat and the adjacent unmanned boat, is the team profit of the i-th unmanned boat.

[0031] Preferably, the total revenue model of the unmanned boat is established through all the capture revenue models, specifically including:

[0032] Obtain the weights corresponding to different roundup profit models according to different roundup stages;

[0033] In different capture stages, the collision avoidance benefit model, distance benefit model, obstacle avoidance benefit model and team benefit model are multiplied by corresponding weights and summed up to obtain the total benefit model corresponding to different capture stages.

[0034] Preferably, the anti-encirclement benefit model between different strategic actions and anti-encirclement benefits in various strategic types of targets is established based on the coordinate values ​​of the unmanned boat, the target and the obstacle in the encirclement space, and further includes:

[0035] According to the coordinate values ​​of the unmanned boats, the targets and the obstacles in the capture space, the positions of the centroids of all the unmanned boats and the distances between the targets and each unmanned boat are respectively obtained;

[0036] The breakout benefit model in the anti-encirclement benefit model is obtained according to the distance between the target and the centroid of all unmanned boats. The expression of the breakout benefit model is:

[0037]

[0038] Among them, J2 is the target’s breakout benefit, (x i ,y i ) is the centroid coordinate of the unmanned boat formation, (x goal ,y goal ) is the coordinate of the target; n is the total number of unmanned boats;

[0039] The escape benefit model in the anti-encirclement benefit model is obtained according to the distance between the target and each unmanned boat. The expression of the escape benefit model can be:

[0040]

[0041] Among them, J1 is the target's escape benefit, k1 is the distance benefit coefficient, L i,goal is the distance between the ith unmanned boat and the target, and L2 is the safe distance between unmanned boats.

[0042] Preferably, the total escape risk model of the target is established through all anti-encirclement and capture benefit models, specifically including:

[0043] Obtain the weights corresponding to different anti-encirclement and capture profit models according to different encirclement and capture stages;

[0044] At different stages of encirclement and capture, the breakout benefit model and the escape benefit model are multiplied by corresponding weights and summed up to obtain the total escape risk model corresponding to different stages of encirclement and capture.

[0045] Preferably, the establishment of a whale algorithm model, solving the strategic actions of the unmanned boat through the whale algorithm model, and iteratively optimizing the solved strategic actions of the unmanned boat according to the total benefit model and the total escape risk model to obtain the optimal strategic actions, specifically includes:

[0046] In each round of iteration, the whale algorithm model is used to solve the solution of the unmanned boat's strategic actions corresponding to this round of iteration, and the unmanned boat's strategic actions are brought into the total benefit model and the total escape risk model to obtain the benefits of the solution corresponding to this round of iteration;

[0047] Compare the benefits of the solution corresponding to each iteration with the benefits of all solutions in the previous iterations, and take the solution with the highest benefit as the preferred strategy action;

[0048] When all iterations are completed, the obtained preferred strategy action is used as the optimal strategy action.

[0049] Preferably, the formula of the whale algorithm model is:

[0050]

[0051] in, is the position of the optimal individual, is the distance between the current individual and the optimal individual, and are all coefficient vectors, b is a constant, t is the current number of iterations, is the position vector of the optimal strategy action, is the position vector of the current solution.

[0052] In a second aspect, an unmanned boat collaborative capture device based on an improved whale algorithm is provided, comprising at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the unmanned boat collaborative capture method based on the improved whale algorithm.

[0053] In a third aspect, the present invention further provides a non-volatile computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are executed by one or more processors to complete the method described in the first aspect.

[0054] In a fourth aspect, a chip is provided, comprising: a processor and an interface, for calling and running a computer program stored in the memory from a memory, and executing the method of the first aspect.

[0055] According to a fifth aspect, a computer program product comprising instructions is provided. When the instructions are executed on a computer or a processor, the computer or the processor executes the method according to the first aspect.

[0056] In the sixth aspect, an unmanned boat collaborative capture system based on an improved whale algorithm is provided, including an unmanned boat collaborative capture device based on an improved whale algorithm as in the second aspect, and using an unmanned boat collaborative capture method based on an improved whale algorithm as in the first aspect.

[0057] The present invention provides an unmanned boat collaborative capture method and device based on an improved whale algorithm. According to the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space, capture benefit models of various strategic types of the unmanned boat and anti-capture benefit models of various strategic types of the target are established; a total benefit model of the unmanned boat is established through all the capture benefit models, and a total escape risk model of the target is established through all the anti-capture benefit models; a whale algorithm model is established, and the strategic actions of the unmanned boat are solved through the whale algorithm model, and the solved strategic actions of the unmanned boat are iteratively optimized according to the total benefit model and the total escape risk model to obtain the optimal strategic action; by applying the whale algorithm to the capture action of the unmanned boat, the capture efficiency and success rate of the unmanned boat are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 It is a method flow chart of an unmanned boat collaborative encirclement method based on an improved whale algorithm provided by an embodiment of the present invention;

[0060] Figure 2 It is a flow chart of a method for sending and receiving communication data of a communication node in an unmanned boat collaborative hunting method based on an improved whale algorithm provided by an embodiment of the present invention;

[0061] Figure 3 It is a flow chart of a method for a host node to determine whether a slave node has an abnormality in an unmanned boat collaborative capture method based on an improved whale algorithm provided by an embodiment of the present invention;

[0062] Figure 4 It is a flow chart of a method for a host node to maintain an abnormal situation in an unmanned boat collaborative hunting method based on an improved whale algorithm provided by an embodiment of the present invention;

[0063] Figure 5 It is a method block diagram of an unmanned boat collaborative encirclement method based on an improved whale algorithm provided by an embodiment of the present invention;

[0064] Figure 6 It is a method flow chart of an unmanned boat collaborative encirclement method based on an improved whale algorithm provided by an embodiment of the present invention;

[0065] Figure 7 is a method flow chart of another unmanned boat collaborative capture method based on an improved whale algorithm provided by an embodiment of the present invention;

[0066] Figure 8 It is a schematic diagram of an unmanned boat collaborative capture device based on an improved whale algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0068] In the description of the present invention, it is necessary to understand that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present disclosure.

[0069] In the description of the present invention, the terms "first" and "second" are used only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, the meaning of "multiple" is two or more. In addition, for example, the same type of nouns may be described as two independent individuals by adding "A" and "B" at the end. In this case, the corresponding features defined as "A" and "B" are only used to distinguish the same type of individuals for description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features.

[0070] As used herein, "about," "substantially," or "approximately" includes the stated value and an average value that is within an acceptable range of deviation from the particular value as determined by one of ordinary skill in the art taking into account the measurements in question and the errors associated with the measurement of the particular quantity (i.e., the limitations of the measurement system).

[0071] Unless the context requires otherwise, throughout the specification and claims, the term "including" is to be interpreted as open inclusion, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" and the like are intended to indicate that specific features, structures, materials or characteristics associated with the embodiment or example are included in at least one embodiment or example of the present disclosure. The schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, the specific features, structures, materials or characteristics may be included in any one or more embodiments or examples in any appropriate manner, that is, although they may be carried in the embodiments or examples of the above terms due to reasons such as the order and position of appearance, it is not limited to that they can be carried in combination by one embodiment or example.

[0072] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0073] Embodiment 1:

[0074] Embodiment 1 of the present invention provides an unmanned boat collaborative capture method based on an improved whale algorithm, such as Figure 1 As shown, the method flow includes:

[0075] In step 101, the capture space is projected into a coordinate system to represent the coordinate values ​​of different positions in the capture space.

[0076] In this embodiment, the capture space refers to the sea area where multiple unmanned boats capture the target. In this embodiment, in order to quantify the positions of the unmanned boats and the target, the capture space is flattened, the positions of the unmanned boats and the target are converted into coordinate values, and subsequent calculations are performed using the coordinate values.

[0077] In step 102, a capture benefit model between different strategic actions in various strategic types of the unmanned boat and the capture benefits is established based on the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space, and an anti-encirclement benefit model between different strategic actions in various strategic types of the target and the anti-encirclement benefits is established based on the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space.

[0078] In this embodiment, the obstacle refers to an entity in the sea area of ​​the capture space that will block the movement of the unmanned boat and the target. When the unmanned boat and the target are actually moving, they should keep a distance from the obstacle. When the distance between the unmanned boat and the target and the obstacle is greater than the safety distance, there is no risk of collision between the unmanned boat and the target and the obstacle. When the distance between the unmanned boat and the target and the obstacle is less than or equal to the safety distance, there is a risk of collision between the unmanned boat and the target and the obstacle.

[0079] In this embodiment, the strategy type of the unmanned boat refers to the various moving states of the unmanned boat compared with other reference objects. The strategy types of the unmanned boat include: the distance between each unmanned boat, the distance between each unmanned boat and the target, the distance between each unmanned boat and the obstacle, and the angle between two adjacent unmanned boats and the target. The strategic action of the unmanned boat represents the adjustment of the specific numerical values ​​of the above strategy types. By adjusting the movement direction and movement speed of the unmanned boat, the positional relationship between the unmanned boat and other objects is adjusted. The impact of the change in the positional relationship on the capture benefit of the unmanned boat is the capture benefit model, wherein the capture benefit is the score of whether the unmanned boat can successfully capture the target. When the unmanned boat is easier to successfully capture the target, the capture benefit is relatively higher. When the unmanned boat is more difficult to successfully capture the target, the capture benefit is relatively lower.

[0080] In this embodiment, the strategy type of the target refers to the movement state of the target compared with other reference objects. The strategy type of the target includes: the distance between the target and each drone and the distance between the target and the centroid of all unmanned boats. The strategic action of the target represents the adjustment of the specific numerical values ​​of the above strategy types. When the target has different movement directions and movement speeds, the positional relationship between the target and other objects changes. The relationship between this change and the anti-encirclement benefit of the target is the encirclement benefit model, wherein the anti-encirclement benefit is the score of whether the target can escape the encirclement of the unmanned boat. When the target is easier to escape, the anti-encirclement benefit is relatively higher, and when the target is more difficult to escape, the anti-encirclement benefit is relatively lower.

[0081] In step 103, a total benefit model of the unmanned boat is established through all the capture benefit models, and a total escape risk model of the target is established through all the anti-capture benefit models.

[0082] In this embodiment, all the encirclement and capture benefit models are integrated according to the corresponding weights to obtain the total benefit model, and the total benefit model is used to reflect the score of the unmanned boat capturing the target. All the anti-encirclement and capture benefit models are integrated according to the corresponding weights to obtain the total escape risk model, and the total escape risk model is used to reflect the score of the target escaping from the unmanned boat encirclement.

[0083] In step 104, a whale algorithm model is established, and the strategic actions of the unmanned boat are solved by the whale algorithm model. The solved strategic actions of the unmanned boat are iteratively optimized according to the total benefit model and the total escape risk model to obtain the optimal strategic actions.

[0084] The whale algorithm is an algorithmic model of the behavior pattern of whales when a group of whales are hunting. It plans and predicts the behavior strategy of whales with the capture target as the iteration direction. It is obtained by summarizing the behavior pattern of whales in nature. In this embodiment, each unmanned boat individual is regarded as a whale individual in the whale algorithm, and the action strategy of each unmanned boat is iteratively solved by imitating the hunting behavior of the whale group. The action strategy of each unmanned boat solved in each round of iteration is brought into the total benefit model and the total escape risk model, and the benefit corresponding to the action strategy solved in each round of iteration is calculated. In each round of iteration, the action strategy with the highest current benefit is obtained by comparing the benefits. After all iteration rounds are completed, the action strategy with the highest benefit at this time is used as the optimal strategy action, and the optimal strategy action is applied to the action of the unmanned boat, thereby achieving the highest benefit strategy action for the unmanned boat and improving the capture efficiency of the unmanned boat.

[0085] Furthermore, in this embodiment, in order to digitize different position areas in the capture space, and thus digitize the positions of each unmanned boat, obstacle and target in the capture space at different times for application in subsequent model calculation and iteration, this embodiment also involves the following designs, such as Figure 2 As shown:

[0086] In step 201, the capture space is planarized and discretized into a grid network.

[0087] In step 202, the grid network is projected into a coordinate system, and the vertices of each grid area in the grid network correspond to coordinate values ​​in the coordinate system.

[0088] In step 203, the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space are obtained respectively according to the positions of the unmanned boat, the target and the obstacle in the grid network.

[0089] In this embodiment, the capture space refers to the sea area where the unmanned boat captures the target, and the capture space is also used to determine whether the target is captured successfully. When the target is within the capture space, it cannot be determined that the target escapes successfully. When the target is outside the capture space, it can be determined that the target escapes successfully. The grid network is a grid distributed in an array. In this embodiment, the side length of each grid can be 1m.

[0090] Furthermore, in this embodiment, since the capture benefit of the unmanned boat for the target includes multiple factors, and it is also necessary to ensure that the unmanned boat will not collide with obstacles during the driving process, the capture benefit model corresponding to different strategy types needs to take the above factors into consideration. Therefore, this embodiment also involves the following designs, such as Figure 3 As shown:

[0091] In step 301, the distances between the unmanned boats, the distances between the unmanned boats and the targets, the distances between the unmanned boats and the obstacles, and the angles between two adjacent unmanned boats and the targets are obtained according to the coordinate values ​​of the unmanned boats, the targets, and the obstacles in the capture space.

[0092] In this embodiment, the distance between two different positions in the capture space can be expressed as:

[0093]

[0094] Among them, L i,j is the distance between point i and point j, where the coordinates of point i are (x i ,y i ), the coordinates of point j are (x j ,y j ).

[0095] In step 302, a collision avoidance benefit model in the encirclement benefit model is obtained according to the distance between the unmanned boats, a distance benefit model in the encirclement benefit model is obtained according to the distance between the unmanned boats and the target, an obstacle avoidance benefit model in the encirclement benefit model is obtained according to the distance between the unmanned boats and the obstacle, and a team benefit model in the encirclement benefit model is obtained according to the angle between the two adjacent unmanned boats and the target.

[0096] In this embodiment, when the distance between the target and m (3≤m≤n) unmanned boats is always kept within the preset capture range within the preset duration, it is considered that the target is captured successfully. When the target is out of the boundary range of the capture space, it is determined that the target capture fails. The preset duration and the preset capture range can be set by those skilled in the art. n is the total number of unmanned boats. The determination of the success of the target capture by n unmanned boats can be:

[0097] L i,goal (Δ×iter0) <R,i=1,2,....m;

[0098] Where iter0 is the time taken for a single iteration, Δ×iter0 is the preset duration, R is the preset capture range, and L i,goal (Δ×iter0) is the distance between the unmanned boat and the target during the preset duration.

[0099] In this embodiment, the distance benefit model needs to be designed according to the following method: when there is no collision between the unmanned boat and the target, the closer the unmanned boat is to the target, the higher the capture benefit and the more likely it is to be successful. In order to avoid a collision between the unmanned boat and the target, a safe distance is set between the unmanned boat and the target. When the distance between the unmanned boat and the target is less than or equal to the safe distance, the benefit is set to negative infinity to avoid a collision between the unmanned boat and the target. In summary, the expression of the distance benefit model can be:

[0100]

[0101] Among them, L1 is the safe distance between the unmanned boat and the target, L i,goal is the distance between the i-th unmanned boat and the target, is the capture benefit generated by the distance between the i-th unmanned boat and the target.

[0102] In this embodiment, the team benefit model needs to be designed according to the following method: In the process of encirclement and capture, in order to strengthen the team coordination of the unmanned boats and achieve more effective encirclement, the angles between two adjacent unmanned boats and the target are set to be the same. The difference is the distribution benefit. When the angle between two adjacent unmanned boats and the target is closer The higher the corresponding income, the higher the corresponding income. In summary, the expression of the team income model can be:

[0103]

[0104] Among them, θ is the angle between the i-th unmanned boat and the adjacent unmanned boat, is the team profit of the i-th unmanned boat.

[0105] In this embodiment, the collision avoidance benefit model needs to be designed according to the following method: in the process of rounding up, in order to avoid collision between unmanned boats, a safety distance L2 between unmanned boats is set. When the distance between the unmanned boats is less than the safety distance L2, the risk of collision between the unmanned boats is greatly increased. At this time, the benefit should be set to negative infinity to avoid the distance between the unmanned boats being less than or equal to the safety distance L2. When the distance between the unmanned boats is greater than the safety distance L2, the longer the distance between the unmanned boats, the higher the benefit. In summary, the expression of the collision avoidance benefit model can be:

[0106]

[0107] Among them, L i,j is the distance between the i-th unmanned boat and the j-th unmanned boat, is the collision avoidance benefit of the i-th unmanned boat.

[0108] In this embodiment, the obstacle avoidance profit model needs to be designed according to the following method: in the process of encirclement and capture, in order to avoid collision between the unmanned boat and the obstacle, a safety distance L4 is set between the unmanned boat and the obstacle. When the distance between the unmanned boat and the obstacle is less than or equal to the safety distance, the risk of collision is greatly increased. At this time, the profit value is set to negative infinity. When the distance between the unmanned boat and the obstacle is greater than the safety distance L4, the greater the distance between the unmanned boat and the obstacle, the higher the corresponding profit. In summary, the expression of the obstacle avoidance profit model can be:

[0109]

[0110] in, is the distance between the ith unmanned boat and the ith obstacle, is the obstacle avoidance benefit between the i-th unmanned boat and the i-th obstacle.

[0111] Furthermore, in this embodiment, the comprehensive benefit of a single unmanned boat needs to be integrated with the above-mentioned collision avoidance benefit, distance benefit, obstacle avoidance benefit and team benefit with corresponding proportions, so as to obtain the comprehensive benefit of each unmanned boat to judge the probability of successful capture of a single unmanned boat. Therefore, this embodiment also designs the following design:

[0112] Obtain weights corresponding to different capture benefit models according to different capture stages; in different capture stages, multiply the collision avoidance benefit model, distance benefit model, obstacle avoidance benefit model and team benefit model by corresponding weights and sum them up to obtain the total benefit model corresponding to different capture stages.

[0113] Correspondingly, the expression of the total revenue model is:

[0114]

[0115] Wherein, w1 is the weight corresponding to the distance benefit model, w2 is the weight corresponding to the team benefit model, w3 is the weight corresponding to the collision avoidance benefit model, and w4 is the weight corresponding to the obstacle avoidance benefit model. i is the total revenue of the i-th unmanned boat.

[0116] Further, in this embodiment, since the distance benefit model and the team benefit model should have different emphases at different stages of the roundup, for example, in the early stage of the roundup, each unmanned boat should try to get close to the target, so at this time, the distance benefit should be relatively emphasized, and in the roundup model, each drone should try to form a roundup formation, so at this time, the team benefit should be relatively emphasized. In summary, on this basis, the expressions of the weight w1 corresponding to the distance benefit model and the weight w2 corresponding to the team benefit model can be:

[0117]

[0118] Among them, w dis is the distance between the unmanned boat and the target, w ca is the weight or related parameter of the collision avoidance benefit model between unmanned boats, iter is the current iteration number, iter max is the maximum number of iterations.

[0119] Furthermore, in this embodiment, in addition to selecting a strategic action by calculating the capture benefit of the unmanned boat, it is also necessary to simulate the strategic action of the target to calculate the anti-capture benefit of the target, and formulate the strategic action of the unmanned boat by comprehensively considering the capture benefit and the anti-capture benefit, so as to further improve the convergence efficiency of the strategic action of the unmanned boat in the iterative generation process. Therefore, this embodiment also involves the following designs, such as Figure 4 As shown:

[0120] In step 401, the positions of the centroids of all the unmanned boats and the distances between the targets and each unmanned boat are respectively obtained according to the coordinate values ​​of the unmanned boats, the targets and the obstacles in the capture space.

[0121] In step 402, a breakout benefit model in the anti-encirclement benefit model is obtained according to the distance between the target and the centroids of all unmanned boats, and an escape benefit model in the anti-encirclement benefit model is obtained according to the distance between the target and each unmanned boat.

[0122] In this embodiment, the breakout benefit model needs to be designed according to the following method: during the capture process, the closer the target is to the centroid of the unmanned boat formation, the easier it is to capture the target; the farther the target is from the centroid of each unmanned boat formation, the easier it is to break out and escape. Therefore, the expression of the breakout benefit model can be:

[0123]

[0124] Among them, J2 is the target’s breakout benefit, (x i ,y i ) is the centroid coordinate of the unmanned boat formation, (x goal ,y goal ) are the coordinates of the target.

[0125] In this embodiment, the escape benefit model needs to be designed according to the following method: during the capture process, under the premise that the target and each unmanned boat do not collide, the closer the distance between the target and each unmanned boat is, the easier it is for the target to be captured successfully; when the distance between the target and each unmanned boat is farther, the easier it is for the target to escape successfully. Therefore, the expression of the escape benefit model can be:

[0126]

[0127] Among them, J1 is the target's escape benefit and k1 is the distance benefit coefficient.

[0128] Furthermore, in this embodiment, the total escape benefit for the target needs to be integrated with the above-mentioned breakout benefit and escape benefit with corresponding weights, so as to obtain the total escape benefit for the target to judge the probability of the target's successful escape. Therefore, this embodiment also designs the following design, such as Figure 5 As shown:

[0129] In step 501, weights corresponding to different anti-encirclement and capture profit models are obtained according to different encirclement and capture stages.

[0130] In step 502, at different capture stages, the breakout benefit model and the escape benefit model are multiplied by corresponding weights and summed to obtain a total escape risk model corresponding to different capture stages.

[0131] In this embodiment, the expression of the total escape risk model can be:

[0132] J i =w5J1+w6J2;

[0133] Among them, J i is the total escape risk of the target, w5 is the weight of the escape profit model of the target, and w6 is the weight of the breakout profit model of the target.

[0134] After obtaining the total benefit model and the total escape risk model, the strategic actions of each unmanned boat can be iterated through the whale algorithm. By allowing the unmanned boat to imitate the strategic actions of real whale groups to capture prey, the capture efficiency and success rate of the unmanned boat can be improved. On this basis, the strategic actions of the unmanned boat are solved and iterated for multiple rounds through the whale algorithm. The solution generated in each round of iteration is evaluated for benefits through the total benefit model and the total escape risk model to obtain the strategic action with the highest relative benefit, thereby improving the capture efficiency and success rate of the unmanned boat. Therefore, this embodiment also involves the following designs, such as Figure 6 As shown:

[0135] In step 601, in each round of iteration, the solution of the strategic action of the unmanned boat corresponding to this round of iteration is solved by the whale algorithm model, and the strategic action of the unmanned boat is brought into the total benefit model and the total escape risk model to obtain the benefit of the solution corresponding to this round of iteration.

[0136] In this embodiment, the whale algorithm includes two models: a shrinking and surrounding mechanism and a spiral model mechanism, wherein the expression of the shrinking and surrounding mechanism can be:

[0137]

[0138] in, is the position of the optimal individual, is the distance between the current individual and the optimal individual, and are all coefficient vectors, t is the current iteration number, is the position vector of the optimal strategy action in the tth iteration, is the position vector of the solution in the tth iteration.

[0139] in, and The expression is as follows:

[0140]

[0141] in, is a number that decreases linearly from 2 to 0 with each iteration, are two random numbers between 0 and 1.

[0142] In this embodiment, the expression of the spiral model mechanism in the whale algorithm is:

[0143]

[0144] in, is the position of the optimal individual, is the distance between the current individual and the optimal individual, and are all coefficient vectors, t is the current iteration number, is the position vector of the optimal strategy action in the tth iteration, is the position vector of the solution in the tth iteration.

[0145] In actual situations, for the two capture methods of shrinking and surrounding mechanism and spiral model mechanism, it is necessary to consider both of them when solving iterative solutions. Set p as a random number between [0, 1]. The size of p determines the selection of shrinking and surrounding mechanism and spiral model mechanism in the capture process, so as to update the position of the unmanned boat in the capture process. The corresponding mathematical model is:

[0146]

[0147] in, is the position of the optimal individual, is the distance between the current individual and the optimal individual, and are all coefficient vectors, t is the current iteration number, is the position vector of the optimal strategy action in the tth iteration, is the position vector of the solution in the tth iteration.

[0148] Furthermore, whale groups use random search to find prey, and individuals in the group will perform random searches based on the locations of their companions. When it is greater than or equal to 1, A value will be randomly selected to move away from the current prey and look for more suitable prey. By randomly selecting a search agent to update the position of the search phase, the whale algorithm can perform a global search, which is modeled as:

[0149]

[0150] in, is a random generation solution in the iterative process.

[0151] Furthermore, based on the above algorithm, the crossover and selection operations of the differential algorithm are introduced to analyze the solution space of the whale algorithm. The crossover operation in the differential algorithm is used to push the individual to the infeasible area to generate the test vector vi. The corresponding expression is:

[0152]

[0153] Among them, v i,j,t+1 It is the position vector of the i-th unmanned boat when it takes the j-th strategic action in the t+1-th iteration.

[0154] Furthermore, in order to prevent individuals from moving to the position where fitness crosses, the greedy selection operation in the differential algorithm is introduced, and the corresponding expression is:

[0155]

[0156] The test variables are updated for some dimensions. If the fitness of the test variables is worse than that of the individuals of the previous generation, they will not be updated, thereby further strengthening the better position of the previous generation during the iteration of the community.

[0157] In summary, the formula of the whale algorithm model is:

[0158]

[0159] in, is the position of the optimal individual, is the distance between the current individual and the optimal individual, and are all coefficient vectors, b is a constant, which can generally be set to 1, t is the current number of iterations, is the position vector of the optimal strategy action, is the position vector of the current solution.

[0160] In step 602, the benefit of the solution corresponding to each iteration is compared with the benefits of all solutions in the previous iteration rounds, and the solution with the highest benefit is taken as the preferred strategic action.

[0161] In step 603, when all iterations are completed, the obtained preferred strategy action is used as the optimal strategy action.

[0162] In summary, if Figure 7 , which is a flow chart of an unmanned boat collaborative capture method based on an improved whale algorithm according to an embodiment of the present invention.

[0163] Embodiment 2:

[0164] like Figure 8, which is a schematic diagram of an unmanned boat cooperative capture device based on an improved whale algorithm according to an embodiment of the present invention. The unmanned boat cooperative capture device based on an improved whale algorithm according to this embodiment includes one or more processors 41 and a memory 42.

[0165] The processor 41 and the memory 42 may be connected via a bus or other means. Figure 8 The example of connecting through bus is taken in the following.

[0166] The memory 42 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs and non-volatile computer executable programs, such as the unmanned boat collaborative capture method based on the improved whale algorithm in the above embodiment. The processor 41 executes the unmanned boat collaborative capture method based on the improved whale algorithm by running the non-volatile software programs and instructions stored in the memory 42.

[0167] The memory 42 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 42 may optionally include a memory remotely arranged relative to the processor 41, and these remote memories may be connected to the processor 41 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0168] The program instructions / modules are stored in the memory 42, and when executed by the one or more processors 41, the unmanned boat cooperative capture method based on the improved whale algorithm in the above embodiment is executed, for example, the above described Figure 1-Figure 7 The steps shown.

[0169] An embodiment of the present invention further provides a computer storage medium having computer program instructions stored thereon; when the computer program instructions are executed by a processor, the unmanned boat collaborative capture method based on the improved whale algorithm provided in an embodiment of the present invention is implemented.

[0170] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A cooperative capture method of unmanned boats based on an improved whale algorithm, characterized in that: include: Projecting the capture space into a coordinate system to represent coordinate values ​​of different positions in the capture space; According to the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space, a capture benefit model between different strategic actions and capture benefits in various strategic types of the unmanned boat is established; according to the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space, an anti-encirclement benefit model between different strategic actions and anti-encirclement benefits in various strategic types of the target is established; The total benefit model of the unmanned boat is established through all the capture benefit models, and the total escape risk model of the target is established through all the anti-capture benefit models; A whale algorithm model is established, and the strategic actions of the unmanned boat are solved by the whale algorithm model. The solved strategic actions of the unmanned boat are iteratively optimized according to the total benefit model and the total escape risk model to obtain the optimal strategic actions.

2. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 1 is characterized in that: The projecting of the capture space into the coordinate system to represent the coordinate values ​​of different positions in the capture space specifically includes: The capture space is flattened and discretized into a grid network; Projecting the grid network into a coordinate system, and corresponding the vertices of each grid area in the grid network to coordinate values ​​in the coordinate system; According to the positions of the unmanned boat, the target and the obstacle in the grid network, the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space are obtained respectively.

3. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 1 is characterized in that: The method of establishing a capture benefit model between different strategic actions and capture benefits in various strategy types of the unmanned boat based on the coordinate values ​​of the unmanned boat, the target and the obstacle in the capture space specifically includes: According to the coordinate values ​​of the unmanned boats, the target and the obstacle in the capture space, the distance between each unmanned boat, the distance between each unmanned boat and the target, the distance between each unmanned boat and the obstacle, and the angle between two adjacent unmanned boats and the target are respectively obtained; The collision avoidance benefit model in the capture benefit model is obtained according to the distances between the unmanned boats. The expression of the collision avoidance benefit model is: Among them, L i,j is the distance between the i-th unmanned boat and the j-th unmanned boat, is the collision avoidance benefit of the i-th unmanned boat, L2 is the safe distance between unmanned boats, and n is the total number of unmanned boats; The distance benefit model in the capture benefit model is obtained according to the distances between each unmanned boat and the target. The expression of the distance benefit model is: Among them, L1 is the safe distance between the unmanned boat and the target, L i,goal is the distance between the i-th unmanned boat and the target, is the capture benefit generated by the distance between the i-th unmanned boat and the target; The obstacle avoidance benefit model in the capture benefit model is obtained according to the distance between each unmanned boat and the obstacle. The expression of the obstacle avoidance benefit model can be: in, is the distance between the ith unmanned boat and the ith obstacle, is the obstacle avoidance benefit between the i-th unmanned boat and the i-th obstacle, and L4 is the safe distance between the unmanned boat and the obstacle; The team benefit model in the capture benefit model is obtained according to the angle between the two adjacent unmanned boats and the target. The expression of the team benefit model can be: Among them, θ is the angle between the i-th unmanned boat and the adjacent unmanned boat, is the team profit of the i-th unmanned boat.

4. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 3 is characterized in that: The total revenue model of the unmanned boat is established through all the capture revenue models, specifically including: Obtain the weights corresponding to different roundup profit models according to different roundup stages; In different capture stages, the collision avoidance benefit model, distance benefit model, obstacle avoidance benefit model and team benefit model are multiplied by corresponding weights and summed up to obtain the total benefit model corresponding to different capture stages.

5. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 1 is characterized in that: The anti-encirclement benefit model between different strategic actions and anti-encirclement benefits in various strategic types of targets is established based on the coordinate values ​​of the unmanned boat, the target and the obstacle in the encirclement space, and further includes: According to the coordinate values ​​of the unmanned boats, the targets and the obstacles in the capture space, the positions of the centroids of all the unmanned boats and the distances between the targets and each unmanned boat are respectively obtained; The breakout benefit model in the anti-encirclement benefit model is obtained according to the distance between the target and the centroid of all unmanned boats. The expression of the breakout benefit model is: Among them, J2 is the target’s breakout benefit, (x i ,y i ) is the centroid coordinate of the unmanned boat formation, (x goal ,y goal ) is the coordinate of the target; n is the total number of unmanned boats; The escape benefit model in the anti-encirclement benefit model is obtained according to the distance between the target and each unmanned boat. The expression of the escape benefit model can be: Among them, J1 is the target's escape benefit, k1 is the distance benefit coefficient, L i,goal is the distance between the ith unmanned boat and the target, and L2 is the safe distance between unmanned boats.

6. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 5 is characterized in that: The total escape risk model of the target is established through all anti-encirclement and capture benefit models, specifically including: Obtain the weights corresponding to different anti-encirclement and capture profit models according to different encirclement and capture stages; At different stages of encirclement and capture, the breakout benefit model and the escape benefit model are multiplied by corresponding weights and summed up to obtain the total escape risk model corresponding to different stages of encirclement and capture.

7. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 1 is characterized in that: The whale algorithm model is established, the strategy action of the unmanned boat is solved by the whale algorithm model, and the solved strategy action of the unmanned boat is iteratively optimized according to the total benefit model and the total escape risk model to obtain the optimal strategy action, which specifically includes: In each round of iteration, the whale algorithm model is used to solve the solution of the unmanned boat's strategic actions corresponding to this round of iteration, and the unmanned boat's strategic actions are brought into the total benefit model and the total escape risk model to obtain the benefits of the solution corresponding to this round of iteration; Compare the benefits of the solution corresponding to each iteration with the benefits of all solutions in the previous iterations, and take the solution with the highest benefit as the preferred strategy action; When all iterations are completed, the obtained preferred strategy action is used as the optimal strategy action.

8. The unmanned boat collaborative capture method based on the improved whale algorithm according to claim 7 is characterized in that: The formula of the whale algorithm model is: in, is the position of the optimal individual, is the distance between the current individual and the optimal individual, and are all coefficient vectors, b is a constant, t is the current number of iterations, is the position vector of the optimal strategy action, is the position vector of the current solution.

9. An unmanned boat cooperative capture device based on an improved whale algorithm, characterized in that: It includes at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the unmanned boat collaborative encirclement method based on the improved whale algorithm as described in any one of claims 1 to 8.

10. A non-volatile computer storage medium, characterized in that: The computer storage medium stores computer program instructions, which, when executed by one or more processors, implement the unmanned boat collaborative capture method based on the improved whale algorithm as described in any one of claims 1 to 8.

Images