A Method for Enhancing Coverage of Directed Sensor Networks Based on AUV

Abstract

A method for enhancing the coverage of a directed sensor network based on AUV. The present invention relates to a method for enhancing the coverage of a directed sensor network based on AUV. In the existing coverage enhancement method based on the virtual force algorithm, the present invention introduces a repulsive force to maintain a safe distance, considers the angle constraint that the path turning angle cannot be obtuse when planning the AUV movement path, and introduces a virtual moment of the unmonitored area when adjusting the sensor sensing angle of the AUV. The above improvements solve the problems that the existing methods do not consider the maneuverability characteristics and collision avoidance problems during the AUV navigation, improve the network coverage effect on the premise of ensuring the safety of the AUV and the rationality of the path planning point selection, and at the same time accelerate the convergence speed and reduce the AUV movement energy consumption, thereby improving the AUV deployment scheme for the two-dimensional underwater directed sensor network area coverage. The present invention belongs to the field of AUV-assisted underwater sensor network area coverage.

Description

Technical Field

[0001] The present invention relates to a method for enhancing the coverage of a directed sensor network based on AUV. The present invention belongs to the field of area coverage of underwater sensor networks. Background Art

[0002] The problem of area coverage is a hot issue in sensor networks, which is of great significance for efficiently completing the perception and collection of environmental information and improving the monitoring ability of regional targets. At present, there are few studies on the area coverage problem of underwater robot swarms. The relevant research mainly focuses on unmanned aerial vehicle swarms and land mobile robot swarms. Therefore, how to apply the area coverage enhancement algorithm to the underwater robot network and consider the particularity of the underwater environment at the same time to realize the enhancement of the coverage of the sensor network based on underwater robots has important research significance.

[0003] In a sensor network, sensors can be classified into omnidirectional sensors and directed sensors according to the sensing model. When performing a monitoring task, an AUV can be regarded as a moving directed sensor. Therefore, the area coverage algorithm of the directed sensor network can be applied to the AUV swarm to realize the deployment of the target area coverage task of the AUV swarm.

[0004] When applying the existing directed sensor area coverage enhancement algorithm to the AUV swarm, the maneuvering constraints and collision avoidance problems during the AUV navigation are not considered. Therefore, how to improve the existing enhancement algorithm and apply it to the AUV swarm task has important application value. On the other hand, for a directed sensor network based on AUV, how to reduce the movement of nodes while increasing the coverage effect of the network plays an important role. By reducing the energy consumption for node movement under limited resources and using more energy for network monitoring and coverage tasks, the network lifetime can be extended. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem that when the existing directed sensor area coverage enhancement algorithm is applied to the AUV swarm, the maneuvering constraints and collision avoidance problems during the AUV navigation are not considered, and a method for enhancing the coverage of a directed sensor network based on AUV is proposed.

[0006] The specific process of a method for enhancing the coverage of a directed sensor network based on AUV is as follows:

[0007] Step 1: Select a network monitoring area, discretize the network monitoring area, and randomly deploy n directed mobile sensor nodes S = {s1, s2,..., s n} in this monitoring area. Given the initial state information of each directed sensor node, determine the maximum number of iterations K of the algorithm, the initial number of iterations k = 0, where the i-th directed mobile sensor node si The state information at the k-th iteration includes the position coordinates (x i (k), y i (k)), the opening angle δ of the sensing sector, the sensing radius r s and the sensing direction θ i (k);

[0008] The directed mobile sensor node is an AUV;

[0009] Step 2: Node s i communicates with neighbor nodes within the communication radius r i and establishes a neighbor node list list c of node s i , where m represents the number of neighbor nodes of node s i ; i The number of neighbor nodes;

[0010] Step 3: Node s i calculates the Euclidean distance d i between node s i and the nodes in the neighbor node list s j ; ij (k);

[0011]

[0012] Among them, (x j (k), y j (k)) is the position coordinate of node s j ;

[0013] Step 4: Determine whether there is d ij (k) < l sth , j = 1, 2, 3... m. If so, record the neighbor node s j , and then proceed to Step 5. If not, jump to Step 7;

[0014] Among them, l sth is the safety distance threshold between nodes;

[0015] Step 5: Calculate the virtual repulsive force j of node s i to maintain a safe distance from node s

[0016]

[0017] Among them, ω r is the proportionality coefficient; is the distance vector from node s j to node s i ;

[0018] Step Six: Obtain node s i The virtual resultant force received represents the virtual resultant force received by node s at the k-th iteration i Jump to and execute Step Twelve;

[0019] Step Seven: Establish node s i and s i the set list of neighbor nodes i corresponding centroid-based virtual omnidirectional sensor node model;

[0020]

[0021] Among them, for node s i the state parameters of node s' in the corresponding virtual omnidirectional sensor at the k-th iteration i are the position coordinates (x i '(k), y i '(k)) and the sensing radius r s ';

[0022] Step Eight: Based on the virtual omnidirectional sensor node model, calculate the virtual repulsive force between node s i ' and s i 's neighbor node s' j

[0023] Step Nine: Calculate the virtual attractive force of the unmonitored area around node s i ' on node s i '

[0024] Step Ten: Calculate the virtual boundary acting force received by node s i '

[0025] Step Eleven: Calculate the resultant force i of node s That is, the resultant force received by the virtual node s i ' of node s i

[0026]

[0027] Step Twelve: Divide the resultant force into a component on the x-axis and a component on the y-axis:

[0028]

[0029] Among them, and​​ Unit vectors on the x-axis and y-axis respectively; F ix (k) is the resultant force A component on the x-axis, F iy (k) is the resultant force A component on the y-axis;

[0030] Step Thirteen: Obtain the direction angle α i (k) of the resultant force;

[0031] Step Fourteen: According to the direction angle α i (k) of the resultant force and the moving direction β i (k - 1) of the (k - 1)-th iteration, obtain the moving direction β i of the k-th iteration of node s i (k);

[0032] Step Fifteen: Obtain the movement vector of node s i according to the moving direction β i (k);

[0033] Step Sixteen: Update the position of the node according to the movement vector i of node s ;

[0034] Step Seventeen: Calculate the virtual moment i generated by the unmonitored area around node s i on node s

[0035] Step Eighteen: Update the sensing angle of the node according to the virtual moment ;

[0036] Step Nineteen: Determine whether k = K - 1 is satisfied. If not, update the iteration number k = k + 1 and return to Step Two; if satisfied, end the loop and execute Step Twenty;

[0037] Step Twenty: Obtain the final position information (x i (K), y i (K)) and the sensing direction θ i (K) of node s, that is, the position coordinates and heading angle of the AUV numbered i, and perform the area coverage monitoring task until the monitoring task ends. i (K), that is, the position coordinates and heading angle of the AUV numbered i, and perform the area coverage monitoring task until the monitoring task ends.

[0038] The beneficial effects of the present invention are:

[0039] The present invention proposes a method for enhancing the coverage of a directed sensor network based on an AUV. By adding a repulsive force to maintain a safe distance to the virtual force algorithm in the existing coverage enhancement method and considering the constraint that the planned trajectory angle of the AUV cannot be obtuse during actual navigation to guide the movement of the AUV, and introducing a virtual moment of the unmonitored area to adjust the sensor sensing angle of the AUV, it solves the problems that the existing methods do not consider the maneuvering constraints and collision avoidance problems during the navigation of the AUV, ensures the safety of the AUV and the rationality of the selection of its planned points, improves the network coverage effect, simultaneously speeds up the convergence speed, reduces the energy consumption of the AUV movement, and optimizes the AUV deployment problem for the two-dimensional underwater directed sensor network area coverage. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the unmonitored area around a node provided by an embodiment of the present invention, where s i ′ is the node, s1′, s′2, s3′, s′4 are the i neighbor nodes of s 11 , M 12 , M 13 are the unmonitored sub-areas around the node s i ′, Z1 is the centroid of the unmonitored area around the node s i ′, r s ′ is the sensing radius;

[0041] Figure 2 It is a schematic diagram of the node position update provided by an embodiment of the present invention. represents the virtual resultant force received by the node s i at the k-th iteration, α i (k) represents the direction angle of the resultant force, β i (k) represents the moving direction. represents the moving vector of the node s i at the k-th iteration, represents the moving vector of the node s i at the (k - 1)-th iteration, p i (k) is the node position coordinate at the k-th iteration, p i (k + 1) is the node position coordinate at the (k + 1)-th iteration;

[0042] Figure 3 It is a schematic diagram of the virtual moment based on the unmonitored area around a node provided by an embodiment of the present invention. The sector OAB is the left adjacent area of the node s i , the sector OCD is the right adjacent area of the node s i , θ i is the sensing direction, S + is the area of the unmonitored part in the left adjacent sector area of the node sector sensing area.- is the area of the unmonitored part in the adjacent sector area on the right side of the node's fan-shaped sensing area, is the adjacent sector arc, is the torque, s i (x i , y i ) is the final position information of node s i , where O, A, B, C, and D are the side points of the sector;

[0043] Figure 4 is the initial deployment diagram of the nodes in the AUV-based directed sensor network. X is the abscissa of the target area, and Y is the abscissa of the target area. Each black dot in the figure represents a directed mobile sensor node, and the fan-shaped area represents the sensing area of the directed mobile sensor node. Since the initial deployment of the directed mobile sensor nodes is relatively dense, there will be many overlapping parts between the sensing areas of the directed mobile sensor nodes;

[0044] Figure 5 is the final deployment diagram of the nodes in the AUV-based directed sensor network. Each black dot in the figure represents a directed mobile sensor node, and the fan-shaped area represents the sensing area of the directed mobile sensor node;

[0045] Figure 6 is the comparison diagram of the network coverage rate change between the present invention and two other methods. Specific implementation manner

[0046] Specific implementation manner 1: The specific process of a method for enhancing the coverage of an AUV-based directed sensor network in this implementation manner is as follows:

[0047] Step 1: Select the network monitoring area, discretize the network monitoring area, and randomly deploy n directed mobile sensor nodes S = {s1, s2,..., s n} in this monitoring area. Given the initial state information of each directed sensor node, determine the maximum number of iterations K of the algorithm, and the initial number of iterations k = 0. Among them, the state information of the i-th directed mobile sensor node s i at the k-th iteration includes the position coordinates (x i (k), y i (k)), the opening angle δ of the sensing fan, the sensing radius r s and the sensing direction θ i (k);

[0048] The directed mobile sensor node is an AUV;

[0049] Step 2: Node s i communicates with s i within the communication radius r cPerform information interaction with neighbor nodes within it, and establish the neighbor node list list of node s i where m represents the number of neighbor nodes of node s i ; i

[0050] Step 3: Node s i Calculate the Euclidean distance d i between node s i and the nodes in the neighbor node list j (k); ij (k);

[0051]

[0052] where (x j (k), y j (k)) is the position coordinate of node s j ;

[0053] Step 4: Determine whether there exists d ij (k) < l sth , j = 1, 2, 3... m. If so, record neighbor node s j , and then proceed to Step 5. If not, jump to and execute Step 7;

[0054] where l sth is the safety distance threshold between nodes;

[0055] Step 5: Calculate the virtual repulsive force j for node s i to maintain a safe distance

[0056]

[0057] where ω r is the repulsive coefficient; is the distance vector from node s j to node s i ;

[0058] Step 6: Obtain the virtual resultant force i received by node s represents the virtual resultant force received by node s i at the k-th iteration. Jump to and execute Step 12;

[0059] Step 7: Establish the centroid-based virtual omnidirectional sensor node model i corresponding to node s i and its neighbor node set list i ; ​

[0060]

[0061] Among them, node s i The state parameters of node s at the k-th iteration in the corresponding virtual omnidirectional sensor i ′ are the position coordinates (x i ′(k), y i ′(k)) and the sensing radius r s ′;

[0062] Step Eight: Based on the virtual omnidirectional sensor node model, calculate the virtual repulsive force between node s i ′ and the neighbor node s i ′, s′ j of node s

[0063] Step Nine: Calculate the virtual attractive force of the unmonitored area around node s i ′ on node s i ′

[0064] Step Ten: Calculate the virtual boundary force acting on node s i ′

[0065] Step Eleven: Calculate the resultant force i of node s That is, the resultant force received by the virtual node s i of node s i ′,

[0066]

[0067] Step Twelve: Divide the resultant force into a component on the x-axis and a component on the y-axis:

[0068]

[0069] Among them, and are the unit vectors on the x-axis and y-axis respectively; F ix (k) is a component of the resultant force on the x-axis, and F iy (k) is a component of the resultant force on the y-axis;

[0070] Step Thirteen: Obtain the direction angle α i (k) of the resultant force according to the resultant force components;

[0071] Step Fourteen: According to the direction angle α i(k) and the movement direction β of the (k - 1)-th iteration i The (k - 1)-th iteration obtains node s i The movement direction β of the k-th iteration i (k);

[0072] Step Fifteen: Obtain the movement vector of node s according to the movement direction β i (k) i of

[0073] Step Sixteen: Update the position of the node according to the movement vector of node s i of ;

[0074] Step Seventeen: Calculate the virtual torque generated by the unmonitored area around node s i on node s i ;

[0075] Step Eighteen: Update the sensing angle of the node according to the virtual torque ;

[0076] Step Nineteen: Determine whether k = K - 1 is satisfied. If not, update the iteration number k = k + 1 and return to Step Two; if satisfied, end the loop and execute Step Twenty;

[0077] Step Twenty: Obtain the final position information (x i (K), y i (K)) and the sensing direction θ i (K) of node s i , that is, the position coordinates and heading angle of the AUV numbered i, and perform the area coverage monitoring task until the monitoring task ends.

[0078] Specific Embodiment Two: The difference between this embodiment and Specific Embodiment One is that in Step Eight, based on the virtual omnidirectional sensor node model, calculate the virtual repulsive force between node s i ' and its neighbor node s i ' j ;

[0079] [[ID=6|1]]

[0080] where is the distance vector from node s' j to node s i ', and l th is the distance threshold.

[0081] Other steps and parameters are the same as those in Specific Embodiment One.

[0082] Embodiment 3: The difference between this embodiment and Embodiment 1 or 2 is that in Step 9, the virtual attraction of the unmonitored area around the computing node s i ′ to the node s i ′ is

[0083]

[0084] where ω a is the attraction coefficient, is the distance vector from the node s i ′ to the centroid Z i of the unmonitored area around s i ′.

[0085] Other steps and parameters are the same as those in Embodiment 1 or 2.

[0086] Embodiment 4: The difference between this embodiment and any one of Embodiments 1 to 3 is that in Step 10, the virtual boundary acting force i received by the computing node s

[0087]

[0088] where ω b is the monitoring area boundary attraction coefficient, is the distance vector from the monitoring area boundary to the node s i ′.

[0089] Other steps and parameters are the same as those in any one of Embodiments 1 to 3.

[0090] Embodiment 5: The difference between this embodiment and any one of Embodiments 1 to 4 is that in Step 13, the direction angle α i (k) of the resultant force is obtained according to the resultant force components;

[0091]

[0092] where 0 ≤ α i (k) < 2π.

[0093] Other steps and parameters are the same as those in any one of Embodiments 1 to 4.

[0094] Embodiment 6: The difference between this embodiment and any one of Embodiments 1 to 5 is that in Step 14, because it is necessary to consider the angle constraint that the planned trajectory angle of the AUV cannot be obtuse during actual navigation, so according to the direction angle α i (k) of the resultant force and the moving direction β i (k - 1) of the (k - 1)-th iteration, the node si The moving direction β of the k-th iteration i (k);

[0095] When the iteration number k = 0:

[0096]

[0097] When the iteration number k ≥ 1:

[0098]

[0099] Other steps and parameters are the same as those in any one of the first to fifth specific embodiments.

[0100] Specific embodiment seven: The difference between this embodiment and any one of the first to sixth specific embodiments is that in step fifteen, the node s is obtained according to the moving direction β i (k) i The moving vector of

[0101]

[0102] where ρ is the fixed step size of the node movement in each iteration.

[0103] Other steps and parameters are the same as those in any one of the first to sixth specific embodiments.

[0104] Specific embodiment eight: The difference between this embodiment and any one of the first to seventh specific embodiments is that in step sixteen, the node is updated in position according to the moving vector of node s i The moving vector of where p

[0105]

[0106] where p i (k) is the node position coordinate of the k-th iteration, and p i (k + 1) is the node position coordinate of the (k + 1)-th iteration.

[0107] Other steps and parameters are the same as those in any one of the first to seventh specific embodiments.

[0108] Specific embodiment nine: The difference between this embodiment and any one of the first to eighth specific embodiments is that in step seventeen, the virtual torque generated by the unmonitored area around node s on node s i around node s i is calculated

[0109]

[0110] where S +(k) is the area of the unmonitored part in the adjacent sector area on the left side of the node's sector sensing area, S + (k) acts on node s i The torque on it is in the counterclockwise direction; S - (k) is the area of the unmonitored part in the adjacent sector area on the right side of the node's sector sensing area, S - (k) acts on node s i The torque on it is in the clockwise direction.

[0111] Other steps and parameters are the same as those in any one of the first to eighth specific embodiments.

[0112] Specific embodiment ten: The difference between this embodiment and any one of the first to ninth specific embodiments is that in step eighteen, the sensing angle of the node is updated according to the virtual torque where θ

[0113]

[0114] where, θ i (k) is the sensing angle of the node at the k-th iteration, θ i (k + 1) is the sensing angle of the node at the (k + 1)-th iteration.

[0115] Other steps and parameters are the same as those in any one of the first to ninth specific embodiments.

[0116] Example:

[0117] Step 1: Construct a directed sensor network model based on AUVs. Select a 3000m * 3000m square area as the network monitoring area, and discretize the network monitoring area into 150 * 150 grids (grid side length 2)0m), and randomly deploy 90 directed mobile sensor nodes S = {s1, s2,..., s 90} in this monitoring area as shown Figure 4 In the figure, each black dot represents a directed mobile sensor node, and the sector area represents the sensing area of the directed mobile sensor node. Given the initial state information of each directed sensor node, determine the maximum number of algorithm iterations K(200), the initial iteration number k = 0, where the state information of the i-th node s i at the k-th iteration includes the position coordinates (x i (k), y i (k)), the opening angle δ(120°) of the sensing sector, the sensing radius r s (350m), and the sensing direction θ i (k);

[0118] Step 2: Node s i communicates with those within its communication radius r cPerform information interaction with neighbor nodes within (600m) and establish the neighbor node list list of node s i where m represents the number of neighbor nodes of node s i ; i

[0119] Step 3: Node s i Calculate the Euclidean distance d i (k) between node s i and the nodes in the neighbor node list j ; ij (k);

[0120]

[0121] Step 4: Determine whether there exists d ij (k) < l sth , j = 1, 2, 3... m, where l sth is the safety distance threshold between nodes, l sth = 110. If it exists, record the neighbor node s j , and then proceed to Step 5. If not, jump to and execute Step 7;

[0122] Step 5: Calculate the virtual repulsive force j to maintain a safe distance for node s i ;

[0123]

[0124] where ω r is the repulsive coefficient, ω r = 5; is the distance vector from node s j to node s i ;

[0125] Step 6: Obtain the virtual resultant force i received by node s where k is the current iteration number, representing the virtual resultant force received by node s i at the k-th iteration, and then jump to and execute Step 12;

[0126] Step 7: Establish the centroid-based virtual omnidirectional sensor node model i corresponding to node s i and its neighbor node set list

[0127]

[0128] where the virtual omnidirectional sensor s i corresponding to node s​i The state parameter of ′ is the position coordinate (x i ′(k), y i ′(k)) and the sensing radius r s ′;

[0129] Step Eight: Based on the virtual omnidirectional sensor model, calculate the virtual repulsive force between node s i ′ and s i ′s neighbor node s′ j where

[0130]

[0131] where is the distance vector from node s′ j to node s i ′, l th is the distance threshold,

[0132] Step Nine: Calculate the virtual attractive force of the unmonitored area around node s i ′

[0133]

[0134] where ω a is the attraction coefficient, ω a = 1.2, is the distance vector from node s i ′ to the centroid Z i of its surrounding unmonitored area, as shown in Figure 1 the unmonitored area around node s i ′ is M = M 11 ∪M 12 ∪M 13 ;

[0135] Step Ten: Calculate the virtual boundary force on node s i ′

[0136]

[0137] where ω b is the boundary attraction coefficient of the monitoring area, ω b = 100, is the distance vector from the boundary of the monitoring area to node s i ′;

[0138] Step Eleven: Calculate the resultant force of node s i ′ That is, node s iThe virtual node s i ′ The resultant force received by;

[0139]

[0140] Step Twelve: Divide the resultant force into a component on the x-axis and a component on the y-axis;

[0141]

[0142] Among them, and are the unit vectors on the x-axis and y-axis respectively; F ix (k) is a component of the resultant force on the x-axis, and F iy (k) is a component of the resultant force on the y-axis;

[0143] Step Thirteen: Obtain the direction angle α i (k) of the resultant force according to the resultant force components;

[0144]

[0145] Among them, 0 ≤ α i (k) < 2π;

[0146] Step Fourteen: Because it is necessary to consider the angle constraint that the planned trajectory angle of the AUV cannot be obtuse during actual navigation, so according to the resultant force direction α i (k) and the moving direction β i (k - 1) of the (k - 1)-th iteration, obtain the moving direction β i of the node s i (k) of the k-th iteration;

[0147] When the iteration number k = 0:

[0148]

[0149] When the iteration number k ≥ 1:

[0150]

[0151] Step Fifteen: Obtain the movement vector i (k) of the node s i according to the moving direction β

[0152]

[0153] Among them, ρ is the fixed step size of the node movement for each iteration, ρ = 20;

[0154] Step Sixteen: Obtain the movement vector of node s i (k) according to the movement direction β i and update the position of the node as shown in Figure 2 where p

[0155]

[0156] (k) is the node position coordinate at the k-th iteration, and p i (k + 1) is the node position coordinate at the (k + 1)-th iteration; i (k+1) is the node position coordinate at the (k + 1)-th iteration;

[0157] Step Seventeen: Calculate the virtual moment generated by the unmonitored area around node s i where, as shown

[0158]

[0159] in Figure 3 S + (k) is the area of the unmonitored part in the left adjacent sector area of the node fan-shaped sensing area, and the moment generated by S + (k) acting on node s i is in the counterclockwise direction. S - (k) is the area of the unmonitored part in the right adjacent sector area of the node fan-shaped sensing area, and the moment generated by S - (k) acting on node s i is in the clockwise direction. The adjacent sector arc is 20°;

[0160] Step Eighteen: Update the sensing angle of the node according to the virtual moment as shown

[0161]

[0162] Step Nineteen: Determine whether k = K - 1 is satisfied. If not, update the iteration number k = k + 1 and return to Step Two; if satisfied, end the loop and execute Step Twenty;

[0163] Step Twenty: Obtain the final position information (x i (200), y i (200)) and the sensing direction θ i (200) of node s i , that is, the position coordinate and the heading angle of the AUV numbered i. As shown Figure 5 in this is the final deployment diagram of the node, where each black dot in the figure represents a directed movement sensor node, and the fan-shaped area represents the sensing area of the directed movement sensor node.

[0164] Figure 6 This is the curve graph showing the change of network coverage rate with the number of iterations for the present invention and two other algorithms. It can be seen from the graph that after 45 iterations of the present invention, the network coverage rate of the present invention is basically in a converged state, and the convergence speed is very fast. After 200 iterations, the network coverage rate is increased from 61.98% to 97.85%.

[0165] However, the algorithm combining virtual force and virtual boundary moment requires at least 80 iterations, and the required mobile energy consumption is greater for the network coverage rate to tend to a converged state. After 200 iterations, the network coverage rate is increased from 61.98% to 93.44%.

[0166] However, the algorithm combining virtual force and virtual centripetal moment requires at least 80 iterations, and the required mobile energy consumption is greater for the network coverage rate to tend to a converged state. After 200 iterations, the network coverage rate is increased from 61.98% to 93.17%.

[0167] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and deformations according to the present invention, but these corresponding changes and deformations should fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for enhancing the coverage of a directed sensor network based on AUV, characterized in that: The specific process of the method is as follows: Step 1: Select a network monitoring area, discretize the network monitoring area, and randomly deploy n directed mobile sensor nodes S = {s1, s2,..., s n} in this monitoring area. Given the initial state information of each directed sensor node, determine the maximum number of iterations K, and the initial number of iterations k = 0. The state information of the i-th directed mobile sensor node s i at the k-th iteration includes the position coordinates (x i (k), y i (k)), the opening angle δ of the sensing sector, the sensing radius r s and the sensing direction θ i (k); The directed mobile sensor node is an AUV; Step 2: Node s i communicates with neighbor nodes within the communication radius r of s i and establishes a neighbor node list list of node s c ; i i ​​ Step 3: Node s i According to the neighbor node list list i , calculate node s i and the Euclidean distance d j between node s in the neighbor node list ij (k); Among them, (x j (k), y j (k)) is the position coordinate of node s j Position coordinate; Step 4: Determine whether there is d ij (k) < l sth where j = 1, 2, 3... m. If it exists, record the neighbor node s j Then proceed to Step 5. If not, jump to and execute Step 7; where l sth is the safety distance threshold between nodes; m represents the number of neighbor nodes of node s i ; Step Five: Calculate node s j For node s i Virtual repulsive force to maintain a safe distance Among them, ω r is the rejection coefficient; is the distance vector from node s j to node s i ; Step Six: Obtain node s i The virtual resultant force received Represents node s at the k-th iteration i The virtual resultant force received, jump to and execute Step Twelve; Step 7: Establish node s i and s i neighbor node set list i corresponding centroid-based virtual omnidirectional sensor node model; Among them, node s i The state parameters of node s' at the k-th iteration in the corresponding virtual omnidirectional sensor i are the position coordinates (x' i (k), y' i (k)) and the sensing radius r' s ; Step Eight: Based on the virtual omnidirectional sensor node model, calculate the node s′ i and s′ i neighbor node s′ j inter-node virtual repulsive force Step Nine: Calculate node s′ i The virtual attraction of the unmonitored area around node s′ i to node s′ Step Ten: Calculate the virtual boundary force on node s′ i received Step Eleven: Calculate the resultant force of node s i That is, the resultant force of node s i is the resultant force received by the virtual node s' i of node s Step Twelve: Divide the resultant force into one component on the x-axis and one component on the y-axis: wherein, and are unit vectors on the x-axis and y-axis respectively; F ix (k) is a component of the resultant force on the x-axis, and F iy (k) is a component of the resultant force on the y-axis; Step Thirteen: Obtain the direction angle α of the resultant force based on the resultant force components i (k); Step 14: Based on the angle constraint that the planned trajectory angle cannot be obtuse when the AUV is actually sailing, according to the direction angle α of the resultant force i (k) and the moving direction β i (k - 1) of the (k - 1)-th iteration, obtain the moving direction β i of node s i (k) for the k-th iteration; Step 15: According to the moving direction β i (k) Obtain the movement vector of node s i ​ Step Sixteen: Update the position of the node according to the motion vector of node s i of the node; Step 17: Calculate node s i The unmonitored area around i the virtual moment generated by node s Step Eighteen: Update the sensing angle of the node according to the virtual moment ​ Step 19: Determine whether k = K - 1 is satisfied. If not, update the iteration number k = k + 1 and return to Step 2; if satisfied, end the loop and execute Step 20; Step Twenty: Obtain node s i The final position information (x i (K), y i (K)) and the sensing direction θ i (K), that is, the position coordinates and heading angle of the AUV numbered i, perform the area coverage monitoring task until the monitoring task ends.

2. The method for enhancing the coverage of a directed sensor network based on an AUV according to claim 1, characterized in that: In the eighth step, based on the virtual omnidirectional sensor node model, calculate the node s′ i and s′ i the neighbor node s′ j the virtual repulsive force between nodes Among them, is the distance vector from node s' j to node s' i , and l th is the distance threshold.

3. A method for enhancing the coverage of a directed sensor network based on AUV according to claim 2, characterized in that: In step nine, calculate node s′ i The virtual attraction of the unmonitored area around node s′ i to node s′ Among them, ω a is the attraction coefficient, is the node s′ i to s′ i the centroid Z of the unmonitored area around i the distance vector of.

4. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 3, characterized in that: The calculation node s' in step ten i The virtual boundary force received where ω b is the boundary attraction coefficient of the monitoring area, is the distance vector from the boundary of the monitoring area to the node s′ i ​ 5. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 4, characterized in that: In step thirteen, the direction angle α of the resultant force is obtained based on the components of the resultant force i (k); where 0 ≤ α i (k) < 2π.

6. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 5, characterized in that: In Step 14, based on the angle constraint that the planned trajectory angle cannot be obtuse during the actual navigation of the AUV, according to the direction angle α of the resultant force i (k) and the moving direction β i (k - 1) of the (k - 1)-th iteration, the moving direction β i (k) of the k-th iteration of node s i (k) is obtained; When the iteration number k = 0: When the iteration number k ≥ 1:

7. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 6, characterized in that: In step fifteen, according to the moving direction β i (k) obtain the node s i of the moving vector Where ρ is the fixed step size of node movement for each iteration.

8. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 7, characterized in that: In step sixteen, according to the motion vector of node s i the position of the node is updated according to its motion vector Among them, p i (k) is the node position coordinate at the k-th iteration, and p i (k + 1) is the node position coordinate at the (k + 1)-th iteration.

9. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 8, characterized in that: The calculation node s in step seventeen i The unmonitored area around i The virtual torque generated on node s Among them, S + (k) is the area of the unmonitored part in the adjacent sector area on the left side of the node's sector sensing area, and the moment of S + (k) acting on the node s i is in the counterclockwise direction; S - (k) is the area of the unmonitored part in the adjacent sector area on the right side of the node's sector sensing area, and the moment of S - (k) acting on the node s i is in the clockwise direction.

10. A method for enhancing the coverage of a directed sensor network based on an AUV according to claim 9, characterized in that: In step eighteen, according to the virtual moment update the perception angle of the node Among them, θ i (k) is the node sensing angle at the k-th iteration, and θ i (k + 1) is the node sensing angle at the (k + 1)-th iteration.

Images