An environmental detection method and system for underwater multi-autonomous robotic fish

Abstract

This invention discloses an environmental detection method for underwater multi-autonomous robotic fish. First, an interconnected, coupled, immune-cooperative detection model of underwater multi-autonomous robotic fish is constructed and its parameters are initialized. The robotic fish acquires environmental information about surrounding obstacles and the detected area, and identifies the original antigens of the obstacles. E o and the original antigens in the detected areas E d The invention identifies the decomposition antigens targeting obstacles and detected areas, thereby determining the antibody concentrations of the robotic fish and selecting antibodies based on these concentrations. Based on the behavior of the selected antibodies, the robotic fish completes the next step of detection until the entire detection process is finished. This invention uses the robotic fish's environment as an antigen, the robotic fish as a B cell, and its behavior as antibodies. It constructs an interconnected and coupled immune collaborative detection network targeting obstacles and detected areas, achieving a balance in the robotic fish's processing of two types of environmental information. This improves the coverage of environmental detection, reduces detection duplication, and enables the robotic fish to move rapidly and effectively in unknown and complex environments, thus improving detection efficiency.

Description

Technical Field

[0001] This invention belongs to the field of environmental detection for underwater robots, specifically relating to a method and system for environmental detection using multiple autonomous underwater robotic fish. Background Technology

[0002] In recent years, with the continuous development and utilization of marine resources, underwater target detection has attracted increasing attention. Bionic robotic fish possess advantages such as extremely high energy efficiency, high speed, strong maneuverability, low noise, and minimal environmental disturbance, overcoming the disadvantages of traditional propeller-driven underwater robots, such as high noise, low efficiency, and high energy consumption. Compared to single robotic fish, the cooperation between multiple robotic fish can significantly improve task execution capabilities and efficiency.

[0003] Improving the quality and efficiency of environmental detection in unknown and complex environments has always been a research challenge and hot topic for underwater autonomous robots. Although relevant research results have emerged in recent years, the following shortcomings still exist: (1) When designing environmental detection methods, robots are treated as point masses without considering their motion constraints, thus reducing the effectiveness and practicality of the methods; (2) The flexibility of robots' autonomous selection of detection behaviors in complex environments is insufficient; (3) The collaboration methods among multiple robots are relatively simple. Therefore, drawing on biomimetic intelligent information processing mechanisms to improve environmental detection methods, thereby enhancing the quality and efficiency of detection in unknown and complex environments, is of great significance for promoting the development of my country's underwater robot industry and contributing to my country's maritime power strategy. Summary of the Invention

[0004] To address the aforementioned problems, the present invention aims to provide an environmental detection method and system for underwater multi-autonomous robotic fish.

[0005] The specific technical solution for achieving the objective of this invention is as follows:

[0006] An environmental detection method for underwater multi-autonomous robotic fish includes the following steps:

[0007] Step 1: Construct an interconnected, coupled, immune-cooperative detection model for multiple autonomous underwater robotic fish;

[0008] Step 2: Initialize the parameters in the interconnected coupled immune cooperative detection model;

[0009] Step 3: The robotic fish acquires environmental information about surrounding obstacles and the detected area, and identifies the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas;

[0010] Step 4: Based on the information obtained in Step 3, determine the concentration of each antibody in the robotic fish, and select antibodies based on the concentration of each antibody.

[0011] Step 5: Based on the selected antibody behavior, complete the next step of the robotic fish's detection.

[0012] Step 6: Determine whether the detection task is complete. If yes, end the process; otherwise, proceed to step 3.

[0013] An underwater multi-autonomous robotic fish environmental detection system includes the following modules:

[0014] Interconnected Coupling Immune Collaborative Detection Model Construction Module: Used to construct an interconnected coupling immune collaborative detection model for underwater multi-autonomous robotic fish and initialize its parameters;

[0015] Information acquisition module: Used by the robotic fish to acquire environmental information about surrounding obstacles and the detected area, and to determine the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas;

[0016] Antibody selection module: Used to determine the concentration of each antibody in the robotic fish based on the acquired information, and to select antibodies based on the concentration of each antibody.

[0017] Detection module: Used to perform detection by the robotic fish based on the behavior of the selected antibody, until the detection task is completed.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0019] (1) The present invention uses the environment of the robotic fish as an antigen, the robotic fish as a B cell, and the behavior of the robotic fish as an antibody. It constructs an interconnected and coupled immune cooperative detection network for obstacles and detected areas, thereby achieving a balance when the robotic fish processes two kinds of environmental information, improving the coverage of environmental detection, and reducing the detection duplication rate.

[0020] (2) The present invention divides the detection distance of the robotic fish in three forward directions into three levels: far, medium and near, and encodes the environmental information antigen based on binary, which is conducive to the accurate calculation of the coefficients of mutual stimulation and inhibition between antigens and antibodies, and the precise selection of antibodies, thereby helping to improve the detection quality and efficiency of the robotic fish.

[0021] (3) This invention draws on the T-cell effect and uses the avoidance distance and avoidance angle of the robotic fish facing the obstacle or the detected area as the antigen decomposition. By defining the antibody concentration regulation factor for antigen decomposition by avoidance distance and the potential field guidance factor for antigen decomposition by avoidance angle, the robotic fish can be rapidly and effectively transferred in unknown and complex environments, thus improving the detection efficiency.

[0022] The present invention will be further described below with reference to specific embodiments. Attached Figure Description

[0023] Figure 1 This is a flowchart of an underwater multi-autonomous robotic fish environmental detection method according to the present invention.

[0024] Figure 2 This is a schematic diagram of the three-jointed bionic robotic fish structure in an embodiment of the present invention.

[0025] Figure 3 This is a schematic diagram of the robotic fish environmental detection in an embodiment of the present invention.

[0026] Figure 4 This is a schematic diagram of Jerne's unique immune network.

[0027] Figure 5 This is a schematic diagram of the BT cell immune model.

[0028] Figure 6 A schematic diagram of an interconnected and coupled immune collaborative detection network.

[0029] Figure 7 This is a schematic diagram of the artificial potential field guidance for the underwater autonomous robotic fish of the present invention.

[0030] Figure 8 This is the first detection environment in an embodiment of the present invention.

[0031] Figure 9 This is the second detection environment in an embodiment of the present invention.

[0032] Figure 10 This is the third detection environment in the embodiments of the present invention.

[0033] Figure 11 This is the fourth detection environment in the embodiments of the present invention.

[0034] Figure 12 This is a schematic diagram illustrating the results of four robotic fish using the CDMBIM method to detect environmental conditions in an embodiment of the present invention.

[0035] Figure 13 This is a schematic diagram illustrating the results of four robotic fish using the ICDMBBE method to detect environmental conditions in an embodiment of the present invention.

[0036] Figure 14 This is a schematic diagram illustrating the results of four robotic fish using the ICEDMBTCE method to detect environment 1 in an embodiment of the present invention.

[0037] Figure 15This is a schematic diagram illustrating the results of four robotic fish using the CDMBIM method to detect environment two in an embodiment of the present invention.

[0038] Figure 16 This is a schematic diagram illustrating the results of four robotic fish using the ICDMBBE method to detect environment two in an embodiment of the present invention.

[0039] Figure 17 This is a schematic diagram illustrating the results of four robotic fish using the ICEDMBTCE method to detect environment two in an embodiment of the present invention.

[0040] Figure 18 This is a schematic diagram illustrating the results of four robotic fish using the CDMBIM method to detect environmental condition 3 in an embodiment of the present invention.

[0041] Figure 19 This is a schematic diagram illustrating the results of four robotic fish using the ICDMBBE method to detect environmental conditions in an embodiment of the present invention.

[0042] Figure 20 This is a schematic diagram illustrating the results of four robotic fish using the ICEDMBTCE method to detect environmental condition 3 in an embodiment of the present invention.

[0043] Figure 21 This is a schematic diagram illustrating the results of four robotic fish using the CDMBIM method to detect environment four in an embodiment of the present invention.

[0044] Figure 22 This is a schematic diagram illustrating the results of four robotic fish using the ICDMBBE method to detect environment four in an embodiment of the present invention.

[0045] Figure 23 This is a schematic diagram illustrating the results of four robotic fish using the ICEDMBTCE method to detect environment four in an embodiment of the present invention. Detailed Implementation

[0046] Example

[0047] Combination Figure 1 An environmental detection method for underwater multi-autonomous robotic fish includes the following steps:

[0048] Step 1: Based on the T cell effect, construct an interconnected and coupled immune cooperative detection model for multiple autonomous underwater robotic fish:

[0049] like Figure 2As shown, the underwater robotic fish of the present invention for exploring unknown environments mainly includes three joints, each driven by a servo motor. The upper, lower, and tail parts of the robotic fish are equipped with a dorsal fin, a pectoral fin, and a caudal fin, respectively. The communication module, control system, and power module of the robotic fish are located in the head of the robotic fish. The movement of the robotic fish is mainly achieved by driving the three servo motors to swing the body and propel it. Therefore, the turning angle when swimming in the water is constrained.

[0050] The robotic fish detection primarily utilizes sensors to detect environmental information in the area in front. To facilitate the design of the detection method in this invention and improve the detection quality and efficiency of the robotic fish, the environment of the robotic fish is modeled using a grid. Based on the robotic fish's motion constraints, its forward-looking left and right detection angles are constrained to [-45°, 45°], i.e., the robotic fish's detection direction D = {directly forward, 45° left, 45° right}. The robotic fish's detection distance is divided into three levels: "far," "medium," and "near." The specific distances for these three levels can be determined according to actual needs. The robotic fish's motion direction M is consistent with the detection direction, i.e., M = {forward, 45° left movement, 45° right movement}, and is represented by l = {1, 2, 3}. Figure 3 As shown in the figure, black represents obstacles; the cross-section grid represents the detected area; and the blank grid represents the undetected area.

[0051] To enable underwater autonomous robotic fish to explore unknown and complex environments, this invention draws on technologies such as... Figure 4 The unique immune network hypothesis presented here uses the environment of the robotic fish as an antigen, the robotic fish as a B cell, and the robotic fish's behavior as an antibody. Based on the immune network dynamics model, the robotic fish selects antibodies (i.e., behaviors) to achieve environmental detection based on the stimulation and inhibition between antigens and antibodies. Therefore, the correctness of the robotic fish's antibody selection determines the quality and efficiency of its environmental detection. To further improve the detection quality and efficiency of multiple robotic fish in unknown and complex environments, this invention again draws on the T cell effect in the biological immune system to improve the design of the immune collaborative detection network model and method.

[0052] In the biological immune system, T cells primarily originate from lymphoid stem cells in the bone marrow. After differentiating and maturing in the thymus, they are distributed throughout the body's immune organs and tissues via the bloodstream. T cells themselves do not produce antibodies; they only communicate recognition results to B cells via epitopes. When antigens invade the biological immune system, T cells assist B cells in producing antibodies, preventing B cells from generating excessive antibodies and causing autoimmune diseases. Figure 5As shown, when antigens invade the biological immune system, some act on phagocytes, while others act directly on B cells. Phagocytes engulf the antigens, then break them down into antigen 1 and antigen 2 under the action of enzymes. The decomposed antigens are then displayed on the surface of macrophages and transmitted to T cells. T cells react with the antigens, notifying the immune system of antigen invasion. In the early stages of immunity, T cells assist B cells in producing antibodies. Once the antigens are eliminated, T cells inhibit B cells from continuing to produce antibodies, thereby maintaining the balance of the entire immune system.

[0053] As can be seen from the T-cell effect, antigens broken down by phagocytes, upon further interaction with T cells, help B cells produce antibodies. During environmental exploration, the optimal behavior (antibody) of the robotic fish is directly related not only to obstacle information and area detection status within its detection range, but also to its optimal avoidance strategy relative to obstacles or already detected areas. Therefore, this invention defines the robotic fish sensor as a phagocyte, and defines the avoidance distance and angle when avoiding obstacles or already detected areas as two types of decomposed antigens. This allows for the application of the T-cell effect to improve the artificial immune collaborative detection network model and detection method. To avoid conceptual confusion in the design of the detection method, this invention refers to the original environmental information detected by the robotic fish—i.e., obstacle information and already detected areas—as the original antigen.

[0054] The robotic fish primarily selects its detection behavior based on environmental information, namely obstacle information and the detected area. To facilitate better balance in processing these two types of environmental information, and drawing on the T-cell effect, this invention constructs a system as follows: Figure 6 The interconnected and coupled immune cooperative detection network shown:

[0055] The surrounding environment of the underwater robotic fish is used as an antigen, obstacles and the detected area are used as the original antigen, the underwater robotic fish is used as B cells, and the detection behavior of the underwater robotic fish is used as antibodies.

[0056] Based on the T-cell effect, the robotic fish sensor is used as a phagocytic cell, and the avoidance distance and avoidance angle when the robotic fish avoids obstacles or the already detected area during the detection process are used as two types of decomposing antigens.

[0057] Thus, under the stimulation of the original antigen composed of obstacles and detected areas, and the decomposition antigen composed of avoidance distance and avoidance angle, and simultaneously under the stimulation and inhibition of the antibodies composed of the underwater robotic fish's detection behavior, the underwater robotic fish can avoid obstacles and detected areas through optimal antibody selection, and approach undetected areas, thereby achieving underwater environmental detection.

[0058] Step 2: Initialize the parameters in the interconnected immune cooperative detection model, including the initial position of the underwater robotic fish, the initial antibody stimulation value and concentration value, the concentration regulation coefficient δ, the stimulation regulation coefficient β, the equilibrium coefficient γ, etc.

[0059] Step 3: The robotic fish acquires environmental information about surrounding obstacles and the detected area, and identifies the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposition antigens facing obstacles and detected areas:

[0060] The barrier primitive antigen E o and the original antigen E in the detected area d Together they constitute the original antigen E:

[0061]

[0062] Among them, the barrier primitive antigen E o Determined based on the underwater robotic fish's detection direction and detection distance:

[0063] ;

[0064] Among them, E o A 6-bit binary encoding is used, with each two bits representing obstacle information in the l-th detection direction of the underwater robotic fish, where l∈[1, 3]; ∀ l∈[1, 3], e o (l)∈{00, 01, 10, 11}, where the elements of the set represent no obstacles, distant obstacles, medium obstacles, and near obstacles in the grid within the detection range and in the l-th detection direction, respectively;

[0065] Original antigen E in the detected region d Determined based on the underwater robotic fish's detection direction and detection distance:

[0066]

[0067] In the formula, E d A 6-bit binary encoding is used, with each two bits representing the detected area information in the l-th detection direction of the underwater robotic fish, where l∈[1, 3]; ∀ l∈[1, 3], e d (l)∈{00, 01, 10, 11}, the set elements respectively represent no detected grid cells, 1 detected grid cell, 2 detected grid cells, and 3 detected grid cells in the l-th detection direction and within the detection range from far to near.

[0068] The decomposing antigen of the avoidance distance is:

[0069]

[0070]

[0071] in, δ is an antibody concentration regulator for the decomposition of antigens targeting obstacles. o This represents the antibody concentration regulation coefficient for obstacle-oriented targets. It is the obstacle antigen excitation value at time t, which is the reciprocal of the distance between the underwater robotic fish and the obstacle in the direction of antibody behavior. Let be the excitation value of antibody i at time t-1 in an obstacle-oriented immune detection network; δ is an antibody concentration regulator targeting the decomposition antigen in the detected region. d This is the antibody concentration adjustment coefficient for the detected region. It is the antigen excitation value of the detected area, and is the reciprocal of the distance between the underwater robotic fish and the detected area in the direction of antibody behavior; Let be the excitation value of antibody i at time t-1 in the immune detection network facing the detected region.

[0072] As can be seen from the above formula, the closer the robotic fish is to the obstacle, the smaller the antibody concentration regulating factor, thereby reducing the antibody concentration to avoid the probability of the robotic fish colliding with the obstacle; conversely, it helps to increase the antibody concentration and improve the antibody execution efficiency when facing obstacles at a distance.

[0073] Similarly, the closer the robotic fish is to the already detected area, the smaller the antibody concentration regulating factor becomes, thereby reducing the antibody concentration and preventing the robotic fish from repeatedly detecting the already detected area; conversely, it helps to increase the antibody concentration and improve the antibody execution efficiency when the detected area is detected.

[0074] The antigen-decomposing angles for avoidance are also divided into two categories: those facing obstacles and those targeting already detected areas. The stimulatory effect of the antigen-decomposing angles on antibodies is primarily reflected by defining a potential field guiding factor. For example... Figure 7 As shown in the figure, X r X obs X dt X udt The coordinates represent the robotic fish, the obstacle, the detected area, and the undetected area, respectively. The robotic fish experiences a virtual repulsive force from the obstacle (detected area). ( ), and the virtual gravity F of the unexplored region closest to the robotic fish. att The mechanism, under the combined effect of virtual attraction and repulsion, prioritizes antibody behavior that matches the direction of the combined force to achieve rapid detection. Specifically:

[0075] According to the artificial potential field method:

[0076] The virtual gravity of the robotic fish is:

[0077]

[0078] According to the artificial potential field method, the underwater robotic fish experiences a virtual repulsive force F towards obstacles within its detection range. o rep Obstacle avoidance angle λ o Sum of potential field steering factor σ o They are respectively:

[0079] ;

[0080] ;

[0081]

[0082] Among them, F att For the virtual gravity of the robotic fish, X r X udt These represent the coordinates of the robotic fish and the unexplored area, respectively, where ║·║ represents the Euclidean distance, and k is the coordinate of the unexplored area. att X is the gravitational scaling factor. obs Let ρ be the coordinates of the obstacle. o (X r , X obs )=║X r - X obs ║,ρ r For the robotic fish to detect distance, k o η is the obstacle repulsion force scaling factor. o Let ν be the angle between the direction of a certain antibody's behavior towards an obstacle on the robotic fish and the y-axis. o The obstacle potential field guidance adjustment coefficient;

[0083] Similarly, the underwater robotic fish exerts a virtual repulsive force F towards the detected area within its detection range. d rep Obstacle avoidance angle λ d Sum of potential field steering factor σ d They are respectively:

[0084]

[0085] ;

[0086]

[0087]

[0088] Among them, F att For the virtual gravity of the robotic fish, X dt ρ represents the coordinates of the detected area. d (Xr , X dt )=║X r – X dt ║,k d η is the repulsive force scaling factor for the detected region. d Let ν be the angle between the direction of the robotic fish's behavior toward a detected area and the y-axis. d This is the potential field guidance adjustment coefficient for the detected region.

[0089] Step 4: Based on the information obtained in Step 3, determine the concentration of each antibody in the robotic fish, and select antibodies based on the concentration of each antibody.

[0090] First, we define antibody alignments for obstacles and antibody alignments for detected regions.

[0091] Antibodies represent the behavior of robotic fish in processing environmental information, and their alignment mainly consists of the preconditions and behaviors for obstacle avoidance (of already detected areas) during the detection process. Based on the interconnected coupled immune cooperative detection network, the antibodies of this invention also include two types: obstacle-oriented and obstacle-oriented.

[0092] In this embodiment, we defined a total of nine behaviors, corresponding to nine antibodies, and in this step, we determined the antibody concentration of each antibody one by one.

[0093] The antibody alignment encoding for the obstacle-facing component is determined according to the following table:

[0094] Antibody number <![CDATA[Avoidance precondition (P o )]]> Behavior 45° left straight ahead Right 45° 1 ## 00 ## go ahead 2 ## 01 ## go ahead 3 11 10 11 go ahead 4 00 10 ## Turn left 45° 5 ## 10 00 Turn right 45° 6 00 11 ## Turn left 45° 7 ## 11 11 Turn left 45° 8 ## 11 00 Turn right 45° 9 11 11 ## Turn right 45°

[0095] Each antibody number corresponds to a robot fish's action, and each action corresponds to an obstacle avoidance precondition P. o That is, its corresponding bitwise encoding, where # represents either 0 or 1, which can be chosen arbitrarily; P o A 6-bit binary encoding is used, with each two bits representing obstacle information in the l-th detection direction of the underwater robotic fish, where l∈[1, 3]; ∀ l∈[1, 3], e o (l)∈{00, 01,10, 11}, where the elements of the set represent no obstacles, long-distance obstacles, medium-distance obstacles, and short-distance obstacles in the grid within the detection direction and detection range of the l-th detection area, respectively;

[0096] The antibody alignment encoding for the detected region is determined according to the following table:

[0097] Antibody number <![CDATA[Precondition for avoiding detected areas (P d )]]> Behavior 45° left straight ahead Right 45° 1 ## 00 ## go ahead 2 ## 01 ## go ahead 3 1# 10 1# go ahead 4 00 10 ## Turn left 45° 5 ## 10 00 Turn right 45° 6 0# 11 ## Turn left 45° 7 00 11 1# Turn left 45° 8 ## 11 0# Turn right 45° 9 1# 11 00 Turn right 45°

[0098] Each antibody number corresponds to a robot fish action, and each action corresponds to a precondition P for avoiding a detected area. d That is, its corresponding bitwise encoding, where # represents either 0 or 1, which can be chosen arbitrarily; P dA 6-bit binary encoding is used, with each two bits representing the detected area information in the l-th detection direction of the underwater robotic fish, where l∈[1, 3]; ∀ l∈[1, 3], e d (l)∈{00, 01, 10, 11}, the set elements respectively represent no detected grid cells, 1 detected grid cell, 2 detected grid cells, and 3 detected grid cells in the l-th detection direction and within the detection range from far to near.

[0099] Based on the principle of the interconnected coupled immune cooperative detection network model, the antibodies of this invention include two categories: those facing obstacles and those facing detected regions. Therefore, when calculating antibody concentration, it is necessary to calculate the antibody concentration in both cases separately before performing a weighted summation. Furthermore, the antibody concentration calculation incorporates an antibody concentration regulation factor based on antigen decomposition due to avoidance distance, and a potential field guidance regulation factor based on antigen decomposition due to avoidance angle.

[0100] The concentration of antibody i is:

[0101]

[0102]

[0103]

[0104]

[0105]

[0106] ;

[0107] ;

[0108] ;

[0109] .

[0110] .

[0111] .

[0112] Wherein, γ is the balance coefficient of the two immune networks. , , where are the antibody i activation values ​​facing the obstacle at times t and t-1, respectively; N is the number of antibodies; , and These represent the stimulation coefficient of antibody j against antibody i, the inhibition coefficient of antibody i against antibody j, and the stimulation coefficient of antigen against antibody i, respectively; β o This represents the regulatory coefficient of antigen to antibody stimulation. , where i is the natural mortality rate of antibodies against barrier antigens, and ⊕ is the XOR operation; and These are the alignment codes for antibodies j and i in an obstacle-oriented immune cooperative detection network, respectively; E o (s) encodes the original antigen of the barrier;

[0113] , , where are the antibody i activation values ​​facing the detected region at times t and t-1, respectively; N is the number of antibodies; , and These represent the stimulation coefficient of antibody j on antibody i, the inhibition coefficient of antibody i on antibody j, and the stimulation coefficient of antigen on antibody i, respectively, for the detected region. d This represents the regulatory coefficient of antigen to antibody stimulation. The natural mortality rate of antibodies targeting antigens in the detected regions. and These represent the alignment codes of antibodies j and i in the immune cooperative detection network facing the detected region, respectively; E d (s) represents the original antigen encoding of the detected region;

[0114] After determining the concentration of each antibody (antibody number 1 to antibody number 9), antibody selection is performed based on the roulette wheel algorithm, thus completing the behavior selection of the underwater robotic fish in this step.

[0115] Step 5: Based on the selected antibody behavior, complete the next step of the robotic fish's detection.

[0116] Step 6: Determine whether the detection task is complete. If yes, end the process; otherwise, proceed to step 3.

[0117] An underwater multi-autonomous robotic fish environmental detection system includes the following modules:

[0118] Interconnected Coupling Immune Collaborative Detection Model Construction Module: Used to construct an interconnected coupling immune collaborative detection model for underwater multi-autonomous robotic fish and initialize its parameters;

[0119] Information acquisition module: Used by the robotic fish to acquire environmental information about surrounding obstacles and the detected area, and to determine the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas;

[0120] Antibody selection module: Used to determine the concentration of each antibody in the robotic fish based on the acquired information, and to select antibodies based on the concentration of each antibody.

[0121] Detection module: Used to perform detection by the robotic fish based on the behavior of the selected antibody, until the detection task is completed.

[0122] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the following steps:

[0123] Step 1: Construct an interconnected, coupled, immune-cooperative detection model for multiple autonomous underwater robotic fish;

[0124] Step 2: Initialize the parameters in the interconnected coupled immune cooperative detection model;

[0125] Step 3: The robotic fish acquires environmental information about surrounding obstacles and the detected area, and identifies the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas;

[0126] Step 4: Based on the information obtained in Step 3, determine the concentration of each antibody in the robotic fish, and select antibodies based on the concentration of each antibody.

[0127] Step 5: Based on the selected antibody behavior, complete the next step of the robotic fish's detection.

[0128] Step 6: Determine whether the detection task is complete. If yes, end the process; otherwise, proceed to step 3.

[0129] A computer-storable medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0130] Step 1: Construct an interconnected, coupled, immune-cooperative detection model for multiple autonomous underwater robotic fish;

[0131] Step 2: Initialize the parameters in the interconnected coupled immune cooperative detection model;

[0132] Step 3: The robotic fish acquires environmental information about surrounding obstacles and the detected area, and identifies the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas;

[0133] Step 4: Based on the information obtained in Step 3, determine the concentration of each antibody in the robotic fish, and select antibodies based on the concentration of each antibody.

[0134] Step 5: Based on the selected antibody behavior, complete the next step of the robotic fish's detection.

[0135] Step 6: Determine whether the detection task is complete. If yes, end the process; otherwise, proceed to step 3.

[0136] To verify the effectiveness and superiority of the underwater multi-autonomous robotic fish environmental detection method (Immune Collaborative Environment Detection Method Based on T Cell Effect, ICEDMBTCE) of this invention, targeting, for example... Figures 8 to 11 Numerical detection tests were conducted on four environments with different numbers of robotic fish. The test results were compared with those of the Complete Detection Method Based on Immune Mechanism [1] (CDMBIM) and the Immune Cooperative Detection Method Based on Biological Entropy [2] (ICDMBBE) based on performance indicators such as average detection steps, average repeated detection steps, coverage, and detection repetition rate. The detection comparison results of the three methods for four different environments and different numbers of robotic fish are shown in Tables 3 and 4.

[0137] Table 3

[0138]

[0139] Table 4:

[0140]

[0141] As shown in Tables 3 and 4, when using two robotic fish for four environmental detections, the ICEDMBTCE method of this invention improved coverage by 2.9%, reduced the repetition rate by 15.5%, reduced the average number of detection steps by 6.1%, and reduced the average number of repetition detection steps by 12.8%. When using three robotic fish for four environmental detections, the ICEDMBTCE method of this invention improved coverage by 7.7%, reduced the repetition rate by 33.4%, reduced the average number of detection steps by 13.1%, and reduced the average number of repetition detection steps by 27.7%. When using four robotic fish for four environmental detections, the ICEDMBTCE method of this invention improved coverage by 12.3%, reduced the repetition rate by 49.7%, reduced the average number of detection steps by 15.6%, and reduced the average number of repetition detection steps by 33.8%. This not only verifies the effectiveness of the method of this invention but also demonstrates that its detection performance is significantly better than the other two methods.

[0142] Further comparison Figures 8 to 11 It can be seen that, Figures 8 to 11 The complexity of the four environments increases sequentially, especially environment four, which contains numerous obstacles and two concave obstacles, increasing the detection difficulty for the robotic fish. Further comparison of the data in Tables 3 and 4 shows that the performance advantage of the ICEDMBTCE method of this invention becomes more prominent as the complexity of the detection environment and the number of robotic fish used for detection increases. This demonstrates its superior ability to handle complex and unknown environments and its enhanced collaboration among multiple robotic fish. This is primarily due to the interconnected and coupled immune collaborative detection network constructed for obstacles and detected areas, effectively balancing the processing of environmental information during detection, improving coverage, reducing repetitive detection, and enhancing detection quality. Secondly, the design of avoidance distance and avoidance angle decomposition antigen regulatory factors for obstacles / detected areas improves the obstacle avoidance efficiency of the robotic fish, thereby reducing the average number of detection steps and the average number of repeated detection steps, thus increasing detection efficiency. In addition, the classification of detection distances (far, medium, and near) in the three forward directions of the robotic fish, as well as the antigen encoding of the detected environmental information, not only improves the accuracy of environmental detection but also enables precise calculation of antibody concentration. This not only helps the robotic fish to achieve accurate antibody selection but also improves the detection quality and efficiency of multiple robotic fish.

[0143] Figures 12 to 23 The results of three immune detection methods for four different environments, using four robotic fish, are presented. As shown in the figure, although all three methods achieved detection in different environments, the detection coverage of the method presented in this invention (Immune Collaborative Environment Detection Method Based on T Cell Effect, ICEDMBBE) is significantly higher, followed by the Immune Cooperative Detection Method Based on Biological Entropy (ICDMBBE).

[0144] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for environmental detection using underwater multi-autonomous robotic fish, characterized in that, Includes the following steps: Step 1: Construct an interconnected, coupled, immune-cooperative detection model for multiple autonomous underwater robotic fish: The surrounding environment of the underwater robotic fish is used as an antigen, obstacles and the detected area are used as the original antigen, the underwater robotic fish is used as B cells, and the detection behavior of the underwater robotic fish is used as antibodies. Based on the T-cell effect, the robotic fish sensor is used as a phagocytic cell, and the avoidance distance and avoidance angle when the robotic fish avoids obstacles or the already detected area during the detection process are used as two types of decomposing antigens. Thus, under the stimulation of the original antigen composed of obstacles and the detected area, and the decomposition antigen composed of avoidance distance and avoidance angle, and simultaneously under the stimulation and inhibition between the antibodies composed of the underwater robotic fish's detection behavior, the underwater robotic fish can avoid obstacles and the detected area through optimal antibody selection, and approach the undetected area, thereby realizing underwater environment detection. Step 2: Initialize the parameters in the interconnected coupled immune cooperative detection model; Step 3: The robotic fish acquires environmental information about surrounding obstacles and the detected area, and identifies the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas; Step 4: Based on the information obtained in Step 3, determine the concentration of each antibody in the robotic fish, and select antibodies based on the concentration of each antibody. Step 5: Based on the selected antibody behavior, complete the next step of the robotic fish's detection. Step 6: Determine whether the detection task is complete. If yes, end the process; otherwise, proceed to step 3.

2. The underwater multi-autonomous robotic fish environmental detection method according to claim 1, characterized in that, The obstacle primitive antigen E in step 3 o and the original antigen E in the detected area d Together they constitute the original antigen E: And=(And o ,AND d ) Among them, the barrier primitive antigen E o Determined based on the underwater robotic fish's detection direction and detection distance: Among them, E o A 6-bit binary code is used, and each two bits represent obstacle information in the l-th detection direction of the underwater robotic fish, l∈[1,3]; e o (l)∈{00,01,10,11}, the elements of the set represent no obstacles, long-distance obstacles, medium-distance obstacles and short-distance obstacles in the grid within the detection range of the l-th detection direction; Original antigen E in the detected region d Determined based on the underwater robotic fish's detection direction and detection distance: In the formula, E d A 6-bit binary code is used, and each two bits represent the information of the detected area in the l-th detection direction of the underwater robotic fish, l∈[1,3]; e d (l)∈{00,01,10,11}, the set elements respectively represent the l-th detection direction and the range from far to near where the grid is not detected, 1 grid is detected, 2 grids are detected and 3 grids are detected.

3. The underwater multi-autonomous robotic fish environmental detection method according to claim 2, characterized in that, The decomposing antigen of the avoidance distance is: in, δ is an antibody concentration regulator for the decomposition of antigens targeting obstacles. o This represents the antibody concentration regulation coefficient for obstacle-oriented targets. It is the obstacle antigen excitation value at time t, which is the reciprocal of the distance between the underwater robotic fish and the obstacle in the direction of antibody behavior. Let be the excitation value of antibody i at time t-1 in an obstacle-oriented immune detection network; δ is an antibody concentration regulator targeting the decomposition antigen in the detected region. d This is the antibody concentration adjustment coefficient for the detected region. It is the antigen excitation value of the detected area, and is the reciprocal of the distance between the underwater robotic fish and the detected area in the direction of antibody behavior; Let be the excitation value of antibody i at time t-1 in the immune detection network facing the detected region.

4. The underwater multi-autonomous robotic fish environmental detection method according to claim 2, characterized in that, The antigen that is evaded is: The virtual repulsive force F of the underwater robotic fish towards obstacles within its detection range o rep Obstacle avoidance angle λ o Sum of potential field steering factor σ o They are respectively: F att =-k att ||X r -X udt || Among them, F att For the virtual gravity of the robotic fish, X r X udt These represent the coordinates of the robotic fish and the unexplored area, respectively, where ║·║ represents the Euclidean distance, and k is the coordinate of the unexplored area. att X is the gravitational scaling factor. obs Let ρ be the coordinates of the obstacle. o (X r ,X obs )=║X r -X obs ║,ρ r For the robotic fish to detect distance, k o η is the obstacle repulsion force scaling factor. o Let ν be the angle between the direction of a certain antibody's behavior towards an obstacle on the robotic fish and the y-axis. o This is the obstacle potential field guidance adjustment coefficient; The underwater robotic fish exerts a virtual repulsive force F towards the detected area within its detection range. d rep Obstacle avoidance angle λ d Sum of potential field steering factor σ d They are respectively: F att =-k att ||X r -X udt || Among them, F att For the virtual gravity of the robotic fish, X dt ρ represents the coordinates of the detected area. d (X r ,X dt )=║X r –X dt ║,k d η is the repulsive force scaling factor for the detected region. d Let ν be the angle between the direction of the robotic fish's behavior toward a detected area and the y-axis. d This is the potential field guidance adjustment coefficient for the detected region.

5. The underwater multi-autonomous robotic fish environmental detection method according to claim 1, characterized in that: The antibody concentration in step 4 is: Where γ is the balance coefficient. , where are the antibody i activation values ​​facing the obstacle at times t and t-1, respectively; N is the number of antibodies; and These represent the stimulation coefficient of antibody j against antibody i, the inhibition coefficient of antibody i against antibody j, and the stimulation coefficient of antigen against antibody i, respectively; β o This represents the regulatory coefficient of antigen to antibody stimulation. For antibodies targeting barrier antigens, the natural mortality rate For XOR operation; and These are the alignment codes for antibodies j and i in an obstacle-oriented immune cooperative detection network, respectively; E o (s) encodes the original antigen of the barrier; , where are the antibody i activation values ​​facing the detected region at times t and t-1, respectively; N is the number of antibodies; and These represent the stimulation coefficient of antibody j on antibody i, the inhibition coefficient of antibody i on antibody j, and the stimulation coefficient of antigen on antibody i, respectively, for the detected region. d This represents the regulatory coefficient of antigen to antibody stimulation. The natural mortality rate of antibodies targeting antigens in the detected regions. and These represent the alignment codes of antibodies j and i in the immune cooperative detection network facing the detected region, respectively; E d (s) represents the original antigen encoding of the detected region; After determining the concentration of each antibody, antibody selection is performed based on the roulette wheel algorithm, which is the behavior selection of the underwater robotic fish.

6. The underwater multi-autonomous robotic fish environmental detection method according to claim 5, characterized in that: The antibody targeting the obstacle is encoded as follows: Antibody alignment encoding for obstacles is determined by obstacle avoidance precondition P. o It consists of behavior, where each antibody number corresponds to an action of the robotic fish, and each action corresponds to an obstacle avoidance precondition P. o That is, its corresponding bitwise encoding, where # represents either 0 or 1, which can be chosen arbitrarily; P o It uses 6-bit binary encoding, with each two bits representing obstacle information in the underwater robotic fish's detection directions of 45° to the left, directly in front, and 45° to the right; The nine antibody pair codes for the obstacle are as follows: {(##00## forward), (##01## forward), (11 10 11 forward), (00 10## turn left 45°), (##10 00 turn right 45°), (00 11## turn left 45°), (##11 11 turn left 45°), (##1100 turn right 45°), (11 11## turn right 45°)}; The antibody alignment encoding for the detected region is as follows: Each antibody number corresponds to a robot fish action, and each action corresponds to a precondition P for avoiding a detected region. d That is, its corresponding bitwise encoding, where # represents either 0 or 1, which can be chosen arbitrarily; P d It uses 6-bit binary encoding, with each two bits representing the detected area information in the underwater robotic fish's detection directions of 45° to the left, directly in front, and 45° to the right. Its behavior includes moving forward, turning 45° to the left, and turning 45° to the right. The nine antibody pairs facing the detected region are defined as follows: {(##00## forward), (##01## forward), (1#10 1# forward), (00 10## turn left 45°), (##10 00 turn right 45°), (0#11## turn left 45°), (00 11 1# turn left 45°), (##11 0# turn right 45°), (1#11 00 turn right 45°)}.

7. An underwater multi-autonomous robotic fish environmental detection system, characterized in that, Includes the following modules: Interconnected Coupling Immune Cooperative Detection Model Construction Module: Used to construct an interconnected coupling immune cooperative detection model for underwater multi-autonomous robotic fish and initialize its parameters. The surrounding environment of the underwater robotic fish is used as an antigen, obstacles and the detected area are used as the original antigen, the underwater robotic fish is used as B cells, and the detection behavior of the underwater robotic fish is used as antibodies. Based on the T-cell effect, the robotic fish sensor is used as a phagocytic cell, and the avoidance distance and avoidance angle when the robotic fish avoids obstacles or the already detected area during the detection process are used as two types of decomposing antigens. Thus, under the stimulation of the original antigen composed of obstacles and the detected area, and the decomposition antigen composed of avoidance distance and avoidance angle, and simultaneously under the stimulation and inhibition between the antibodies composed of the underwater robotic fish's detection behavior, the underwater robotic fish can avoid obstacles and the detected area through optimal antibody selection, and approach the undetected area, thereby realizing underwater environment detection. Information acquisition module: Used by the robotic fish to acquire environmental information about surrounding obstacles and the detected area, and to determine the original antigen E of the obstacles. o and the original antigen E in the detected area d And identify the decomposing antigens facing obstacles and detected areas; Antibody selection module: Used to determine the concentration of each antibody in the robotic fish based on the acquired information, and to select antibodies based on the concentration of each antibody. Detection module: Used to perform detection by the robotic fish based on the behavior of the selected antibody, until the detection task is completed.

8. A computer 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 computer program, it implements the steps of the method according to any one of claims 1-6.

9. A computer-storable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-6.

Images