An Optimization Construction Method for the Communication Network of a Formation of Water Quality Monitoring Robots

Abstract

The present invention relates to the technical field of water quality monitoring robots, and specifically to an optimized construction method for the formation communication network of water quality monitoring robots, including the following steps: Step S1: Determine the water area of the water quality monitoring task; Step S2: Determine the coordinates of all water quality monitoring points within the water area of the water quality monitoring task; Step S3: Input the formation situation of the water quality monitoring robot formation; Step S4: Assign tasks to the water quality detection robot formation and establish a task assignment model; Step S5: Use the improved osprey optimization algorithm to solve the task assignment model; Step S6: Establish a fully connected model for the communication of the water quality monitoring robot formation; Step S7: Determine the full connectivity of the formation communication network of the water quality monitoring robot; Step S8: Construct and optimize the communication links in the fully connected network state and the non-fully connected network state. It ensures the connectivity of the formation communication network of the water quality monitoring robot and solves the problem of low monitoring efficiency in large water areas.

Description

Technical Field

[0001] The present invention relates to the technical field of water quality monitoring robots, and specifically to an optimized construction method for a formation communication network of water quality monitoring robots. Background Art

[0002] Water quality monitoring robots have broad application prospects in the fields of water environment monitoring, fishery management, etc. With the continuous expansion of the monitoring scope and the diversification of monitoring scenarios, meeting the requirements of long-term, dynamic, and efficient applications has become the key. To improve the monitoring efficiency, multiple water quality monitoring robots usually need to cooperate, and the monitoring process is optimized through reasonable task allocation.

[0003] During the water quality monitoring process, data needs to be transmitted between water quality monitoring robots through a communication network. Currently, common water quality monitoring methods include randomly deploying fixed anchor nodes on the water surface or directly transmitting data from water quality monitoring robots to the base station. However, these methods have problems such as poor communication effect, high cost, and easy node disconnection. To solve these problems, it is necessary to design a dynamic, efficient, and low-cost communication network construction method according to the formation distribution of water quality monitoring robots to adapt to the complex water quality monitoring environment. Summary of the Invention

[0004] The present invention proposes an optimized construction method for a formation communication network of water quality monitoring robots, based on an improved osprey optimization algorithm, aiming to improve the data transmission efficiency and task execution efficiency of the formation of water quality monitoring robots.

[0005] The water quality monitoring robot has a boat-shaped design and is equipped with a variety of water quality parameter detection sensors, including a pH sensor, a conductivity sensor, a TDS sensor, and a water temperature sensor. The water quality monitoring robot is powered by solar energy, packs the collected water quality data, and transmits it through its own communication module. According to the actual position of the robot, the data can be sent to the base station by direct transmission or multi-hop transmission.

[0006] The present invention is implemented by the following technical solutions:

[0007] An optimized construction method for a formation communication network of water quality monitoring robots, which is implemented by the following steps:

[0008] Step S1: Determine the water area of the water quality monitoring task, and establish a water area map and a coordinate system;

[0009] Step S2: Determine the coordinates of all water quality monitoring points C = {c i (x i , y i )|i = 1, 2,..., N c}, where N c is the number of water quality monitoring points;

[0010] Step S3: Input the formation situation of water quality monitoring robots. The number of water quality monitoring robots is N S , and the formation of water quality robots is expressed as S = {s k | k = 1, 2,.., N S};

[0011] Step S4: Perform task allocation for the formation of water quality detection robots and establish a task allocation model;

[0012] Step S5: Use the Latin hypercube sampling strategy and combine the chaotic mapping mechanism to improve the osprey optimization algorithm to enhance the algorithm performance, and use the improved osprey optimization algorithm to solve the task allocation model;

[0013] Step S6: Establish a fully connected communication model for water quality monitoring robots;

[0014] Step S7: Perform a full connectivity determination on the communication network of the water quality monitoring robot formation;

[0015] Step S8: Construct and optimize the communication links in the fully connected network state;

[0016] Step S9: Construct and optimize the communication links in the non-fully connected network state.

[0017] Furthermore, in the said Step S4: Perform task allocation for the formation of water quality detection robots and establish a task allocation model; specifically as follows:

[0018] Step 4.1: The distance between any two water quality monitoring points i and j is denoted as:

[0019] d ij = ||c i - c j || (1);

[0020] Step 4.2: When the water quality monitoring robot moves forward at a constant speed with a speed of v opt , the instantaneous energy consumption E opt is the smallest. When moving between two water quality monitoring points i and j, the corresponding energy consumption is:

[0021]

[0022] Step 4.3: Establish decision variables When the water quality monitoring robot k moves from water quality monitoring point i to water quality monitoring point j, let Otherwise

[0023] Step 4.4: Establish decision variables When the water quality monitoring robot k takes the water quality monitoring point i as its task point Otherwise

[0024] Step 4.5: Establish decision variables Used to control the water quality monitoring robot k to determine the order of water quality detection at water quality monitoring points

[0025] Step 4.6: Establish a task allocation model for the formation of water quality monitoring robots

[0026]

[0027] s.t.

[0028]

[0029] Among them is the water quality monitoring task point corresponding to the k-th water quality monitoring robot

[0030] Constraint 1 restricts that each water quality monitoring point can only be detected by one water quality monitoring robot

[0031] Constraint 2 restricts that if the water quality monitoring robot has completed the detection of water quality monitoring point i, it must enter the next monitoring point

[0032] Constraint 3 restricts that the task route of each water quality monitoring robot is a complete loop

[0033] Constraint 4 defines the fairness of task allocation, where λ is the energy consumption difference adjustment coefficient, ensuring that the energy consumption difference between water quality monitoring robots is limited within an acceptable range

[0034] Furthermore, in step S5: Use the Latin hypercube sampling strategy and combine it with the chaos mapping mechanism to improve the osprey optimization algorithm to enhance the algorithm performance, and use the improved osprey optimization algorithm to solve the task allocation model; specifically as follows

[0035] Step 5.1: Use the Latin hypercube sampling strategy to improve the population diversity of the osprey optimization algorithm: In the osprey optimization algorithm, the initialization of the osprey population is random. After improvement by the Latin hypercube sampling strategy, the osprey population initialization formula can be rewritten as

[0036] X = LHS(n, D) (5);

[0037]

[0038] where n is the population size; D is the dimension of the problem; LHS is the process of generating n D-dimensional samples using Latin hypercube sampling; X gh is the value of the g-th sample in the h-th dimension. L h and U h are the lower and upper bounds of the h-th dimensional variable respectively; P gh is a random sequence on the interval [0, 1);

[0039] Step 5.2: Further perturb the osprey population using the chaotic mapping mechanism to enhance search diversity and form a new sequence;

[0040] X gh = X gh (1 - φx t ) (7);

[0041] where x t is the value generated by the chaotic mapping, and φ is the mapping control parameter.

[0042] Step 5.3: Take the formation task allocation model of the water quality monitoring robot as the optimization object, and let

[0043]

[0044] For each osprey individual X g , calculate the fitness value f(X g );

[0045] Step 5.4: The osprey hovers for global search, and its position update formula is as follows:

[0046] X g (t + 1) = X b (t) + R 1 sin(2πR 2 )(X b (t) - X g )(t)) (9);

[0047] where X b (t) is the optimal position of the current osprey individual; R 1 , R 2 ∈[0, 1];

[0048] Step 5.5: The osprey approaches the prey for local search, and update the position of the osprey individual as follows:

[0049] X g (t + 1) = X b (t) + R 3 (X b (t) - X g (t)) (10);

[0050] Among them, R 3 is a random number generated from [0, 1], which is used to adjust the degree of proximity of osprey individuals to the optimal solution;

[0051] Step 5.6: After each update, continue to use the chaotic mapping to further perturb the positions of osprey individuals near the optimal solution to avoid local optimality of the algorithm:

[0052] X gh (t) = X gh (t + 1)(1 - φx t ) (11);

[0053] Step 5.7: Determine whether the maximum number of iterations is reached or the fitness value meets the accuracy requirement. If so, output the optimal solution; otherwise, return to Step 5.4.

[0054] Furthermore, in the said Step S6: Establish a full - connectivity communication model for the water quality monitoring robot; specifically as follows:

[0055] Step 6.1: At any moment ε, for any two water quality monitoring robots k, l ∈ 1, 2,... N S , with corresponding positions p k (ε), p l (ε) respectively, the distance between them is obtained as:

[0056] d kl (ε) = ||p k (ε) - p l (ε)|| (12);

[0057] Step 6.2: At any moment ε, the distance from the water quality monitoring robot k to the base station p B (x B , y B ) is:

[0058] d kB (ε) = ||p k (ε) - p B || (13);

[0059] Step 6.3: Define d θ as the maximum distance for establishing communication between two water quality monitoring robots and between a water quality monitoring robot and the base station; then, when the distances between any two water quality monitoring robots S k , S l and between a water quality monitoring robot and the base station satisfy d kl (ε) ≤ d θ , d kB (ε) ≤ d θWhen a connection relationship is established between the two;

[0060] Use the symbol ψ kl = 1 to represent the water quality monitoring robot S k , S l Establish a connection. Conversely, ψ kl = 0; ψ kB = 1 indicates that the water quality monitoring robot S k Establishes a connection with the base station. Conversely, ψ kB = 0;

[0061] Step 6.4: At any moment ε, if S k Wants to transmit a data packet to the base station, then the communication network link of the water quality monitoring robot needs to meet:

[0062]

[0063] Equation (14) means that in the water quality monitoring robots on the island platform, any one water quality monitoring robot establishes a connection relationship with at least one other water quality monitoring robot, and at least one water quality monitoring robot can establish a connection with the base station. The above network is called a fully connected network.

[0064] Furthermore, in step S7: perform a full connectivity determination on the formation communication network of the water quality monitoring robots; specifically as follows:

[0065] Step 7.1: In step S4, the task point matrix corresponding to each water quality monitoring robot is obtained. Find the water quality monitoring robot S k The corresponding task point matrix T k Among them, the distance between each water quality monitoring point and the task point matrix T l Corresponding to the water quality monitoring robot S l Among each water quality monitoring point, construct a distance matrix For storing the corresponding distance information;

[0066] Step 7.2: If the maximum distance between the task points of the water quality monitoring robots S k and S l Satisfies max(D kl ) ≤ d θ , it means that S k and S l Can always remain connected during the task process, that is

[0067]

[0068] Among them, T k , T l Is the time taken for the water quality monitoring robot to complete a round of monitoring tasks;

[0069] Step 7.3: According to the definition of the fully connected network in step S6 and combined with equation (15), if at any time all water quality monitoring robots have at least one data transmission link to the base station, then at all times, the formation of water quality monitoring robots constitutes a fully connected network, which is called an all-time - fully connected network;

[0070] Step 7.4: If the distance matrix D kl does not satisfy max(D kl ) ≤ d θ , but there exist α ∈ {1, 2,..., N k}, β ∈ {1, 2,..., N l} such that D αβ ≤ d θ ; in other words, there is a distance D kl between two water quality monitoring points in the distance matrix D αβ that is less than or equal to the connection threshold d θ , which means that when the water quality monitoring robots S k and S l are at these two water quality monitoring points respectively, data forwarding can be completed, and correspondingly, it is called a part-time - fully connected network;

[0071] Step 7.5: If in the distance matrix D kl , min(D kl ) > d θ > 0, it means that the water quality monitoring robots S k and S l cannot establish a connection throughout the entire mission, which is called a non-fully connected network.

[0072] Furthermore, the step S8: Construct and optimize the communication link in the fully connected network state; specifically as follows:

[0073] Step 8.1: For the all-time - fully connected network, if the water quality monitoring robot S k has a data transmission requirement, by constructing the line connecting its own position S k (x k , y k ) and the base station position p B (x B , y B ), the straight-line equation of S k p B , S k p B is:

[0074]

[0075] S kTo query the water quality monitoring robots connected at the current moment, such as S l (x l , y l ) can be connected to it, then S l to the straight line S k p B The foot of the perpendicular is calculated as follows:

[0076]

[0077] Furthermore, obtain to the base station p B (x B , y B ) distance:

[0078]

[0079] Take as the evaluation index for relay selection, The smaller the value of, the greater the effective distance for the data packet of the water quality monitoring robot S k to advance, and the greater the probability that the water quality monitoring robot is selected as a relay;

[0080] S k Select The water quality monitoring robot S with the smallest l as the next-hop information relay; Similarly, S l Continue to find the next relay node according to the above method until the data packet is sent to the base station;

[0081] Step 8.2: For the time-division full-connected network, it is necessary to optimize the waiting time of the water quality monitoring robot. In step S7.4, there exists α ∈ {1, 2,..., N k}}, β ∈ {1, 2,..., N l}, such that D αβ ≤ d θ , at this time, the water quality monitoring robots S k and S l can establish a connection. If the water quality monitoring robot S k has a transmission requirement, it can wait for S at the water quality monitoring point and wait for S l to reach its water quality monitoring point for data transmission; The waiting transmission time of S k can be reduced by optimizing the initial positions and speeds of S k and S l ;

[0082] When S k sends to S lAfter the transmission of data is completed, S l According to the method in step S8.1 or by waiting, continue to forward the data packet to the next water quality monitoring robot node until it reaches the base station.

[0083] Further, the step S9: construct and optimize the communication link in the non-fully connected network state; specifically as follows:

[0084] Step 9.1: In the water quality monitoring robot formation network, if some water quality monitoring robot nodes cannot join the network, new waypoints need to be added to the route of this node or the surrounding water quality monitoring robot nodes to make it join the data transmission network.

[0085] According to the method in step S7, determine that the water quality monitoring robot node S k cannot join the network, and at the same time obtain S k The corresponding water quality monitoring point T k Among them, there is a point The closest to the water quality monitoring robot S l The corresponding water quality monitoring point is recorded as And S l can join the network, S k Select S l As its relay communication node; otherwise, find the next water quality monitoring robot that meets the above conditions.

[0086] Step 9.2: The water quality monitoring robot S k and S l Each find a point So that these two points meet the connectivity requirements, and the total distance traveled by the water quality monitoring robots S k and S l is the smallest, and the corresponding model is as follows:

[0087]

[0088] The above formula (19) can be solved by using the improved osprey optimization algorithm in step S5; in addition, when multiple nodes need S l To act as a relay, the above formula also applies, as long as the dimension of the solution is increased.

[0089] Step 9.3: Insert As a task point into the matrix T k In the element After that, as the waypoint of the water quality monitoring robot S k ; Insert As a task point into the matrix T l In the element After that, as the water quality monitoring robot S l of the waypoint;

[0090] Step 9.4: The water quality monitoring robot S k Performs data transmission according to step S8.

[0091] Furthermore, the value of the d θ can be determined according to the channel fading parameter, communication transmission power, and antenna gain of the water quality monitoring task water area.

[0092] The structure design of the present invention is reasonable and reliable. Compared with the prior art, the present invention has the following advantages:

[0093] 1. An optimization construction method for the communication network of a water quality monitoring robot formation solves the problem of low monitoring efficiency of a single water quality monitoring robot in a large area of water area. A scheme for the collaborative task of a water quality monitoring robot formation is designed, and a formation task allocation algorithm is designed. An improved osprey algorithm is proposed to improve the task allocation efficiency.

[0094] 2. An optimization construction method for the communication network of a water quality monitoring robot formation, aiming at the problem of constructing the communication network of a water quality monitoring robot formation, deduces a fully connected network model, and optimizes the positions of the water quality monitoring robots to ensure the connectivity of the communication network. Under the optimized network, all water quality monitoring robots can transmit water quality data to the base station, improving the reliability of water quality monitoring; applicable to the deployment and construction of a water surface communication network. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 is a schematic diagram of the system framework of the present invention.

[0096] Figure 2 is the implementation flowchart of the present invention.

[0097] Figure 3 is the implementation flowchart of the improved osprey optimization algorithm based on the Latin hypercube sampling strategy combined with the chaotic mapping mechanism in the present invention.

[0098] Figure 4 is a schematic diagram of the relay selection process of the water quality monitoring robot of the present invention.

[0099] Figure 5 is the deployment result of the water quality robot communication network nodes obtained by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0100] An optimization construction method for the communication network of a water quality monitoring robot formation. First, for the water quality monitoring of large areas of water, a scheme for the collaborative tasks of a water quality monitoring robot formation is designed. Second, a modified osprey optimization algorithm based on the Latin hypercube sampling strategy combined with the chaotic mapping mechanism is developed for task allocation to the water quality monitoring robot formation. Then, a fully connected model of the communication network of the water quality monitoring robot is derived and established to evaluate the communication network of the water quality monitoring robot. Finally, corresponding network connectivity optimization and link optimization models are designed for different network states.

[0101] An optimization construction method for the communication network of a water quality monitoring robot formation, as shown in Figure 1 , Figure 2 shown, is implemented by the following steps:

[0102] Step S1: Determine the water area for water quality monitoring tasks, and establish a water area map and coordinate system;

[0103] Step S2: Determine the coordinates of all water quality monitoring points C = {c i (x i , y i )|i = 1, 2,..., N c ), where N c is the number of water quality monitoring points;

[0104] Step S3: Input the situation of the water quality monitoring robot formation. The number of water quality monitoring robots is N S , and the water quality robot formation is expressed as S = {s k |k = 1, 2,.., N S};

[0105] Step S4: Perform task allocation for the water quality detection robot formation and establish a task allocation model; specifically as follows:

[0106] Step 4.1: The distance between any two water quality monitoring points i and j is denoted as:

[0107] d ij = ||c i - c j || (1);

[0108] Step 4.2: When the water quality monitoring robot moves forward at a constant speed of v opt , the instantaneous energy consumption E opt is the smallest. When moving between two water quality monitoring points i and j, the corresponding energy consumption is:

[0109]

[0110] Step 4.3: Establish decision variables When the water quality monitoring robot k moves from the water quality monitoring point i to the water quality monitoring point j, let Otherwise

[0111] Step 4.4: Establish decision variables When the water quality monitoring robot k takes the water quality monitoring point i as its own task point, Otherwise

[0112] Step 4.5: Establish decision variables Used to control the order in which the water quality monitoring robot k performs water quality detection on water quality monitoring points;

[0113] Step 4.6: Establish a water quality monitoring robot formation task allocation model;

[0114]

[0115] s.t.

[0116]

[0117] The above formula means that the water quality monitoring task points with the quantity of N c are allocated to N S water quality monitoring robots, and determine the order in which each water quality monitoring robot performs water quality detection on water quality monitoring points, so as to minimize the energy consumption of the water quality monitoring robot formation; among them, is the water quality monitoring task point corresponding to the k-th water quality monitoring robot;

[0118] Constraint 1 restricts that each water quality monitoring point can only be detected by one water quality monitoring robot;

[0119] Constraint 2 restricts that if the water quality monitoring robot has completed the detection of the water quality monitoring point i, it must enter the next monitoring point;

[0120] Constraint 3 restricts that the task route of each water quality monitoring robot is a complete loop;

[0121] Constraint 4 defines the fairness of task allocation, where λ is the energy consumption difference adjustment coefficient, ensuring that the energy consumption difference between water quality monitoring robots is limited within an acceptable range.

[0122] Step S5: Use the Latin hypercube sampling strategy and combine the chaotic mapping mechanism to improve the osprey optimization algorithm to enhance the algorithm performance, and use the improved osprey optimization algorithm to solve the task allocation model; specifically as follows:

[0123] Step 5.1: Using the Latin hypercube sampling strategy to improve the population diversity of the osprey optimization algorithm: In the osprey optimization algorithm, the initialization of the osprey population is random. After improvement by the Latin hypercube sampling strategy, the osprey population initialization formula can be rewritten as:

[0124] X = LHS(n, D) (5);

[0125]

[0126] where n is the population size; D is the dimension of the problem; LHS is the process of generating n D-dimensional samples using Latin hypercube sampling; X gh is the value of the g-th sample in the h-th dimension. L h and U h are the lower and upper bounds of the h-th dimensional variable respectively; P gh is a random sequence on the interval [0, 1);

[0127] Step 5.2: Using the chaos mapping mechanism to further perturb the osprey population to improve the search diversity and form a new sequence;

[0128] X gh = X gh (1 - φx t ) (7);

[0129] where x t is the value generated by the chaos mapping, and φ is the mapping control parameter.

[0130] Step 5.3: Taking the formation task allocation model of the water quality monitoring robot as the optimization object, let

[0131]

[0132] For each osprey individual X g , calculate the fitness value f(X g );

[0133] Step 5.4: The osprey hovers for global search, and its position update formula is as follows:

[0134] X g (t + 1) = X b (t) + R 1 sin(2πR 2 )(X b (t) - X g )(t)) (9);

[0135] where X b (t) is the current optimal position of the osprey individual; R 1 , R 2∈[0, 1];

[0136] Step 5.5: The osprey approaches the prey for local search, and updates the position of the osprey individual as follows:

[0137] X g (t + 1) = X b (t) + R 3 (X b (t) - X g (t)) × (10);

[0138] where R 3 ∈[0, 1] is a randomly generated number used to adjust the closeness of the osprey individual to the optimal solution;

[0139] Step 5.6: After each update, continue to use the chaotic map to further perturb the positions of the osprey individuals near the optimal solution to avoid local optimality of the algorithm:

[0140] X gh (t) = X gh (t + 1) × (1 - φx t ) × (11);

[0141] Step 5.7: Determine whether the maximum number of iterations is reached or the fitness value meets the accuracy requirement. If so, output the optimal solution; otherwise, return to Step 5.4.

[0142] The implementation flowchart of the improved osprey optimization algorithm is as shown in the appendix Figure 3 and through the above process, the task allocation of the water quality monitoring formation and the monitoring task points and corresponding task paths of each water quality monitoring robot are obtained

[0143] Step S6: Establish a fully connected communication model for water quality monitoring robots; specifically as follows:

[0144] Step 6.1: At any moment ε, for any two water quality monitoring robots k, l ∈ 1, 2,..., N S , with corresponding positions p k (ε) and p l (ε) respectively, the distance between them is obtained as:

[0145] d kl (ε) = ||p k (ε) - p l (ε)|| × (12);

[0146] Step 6.2: At any moment ε, the distance from water quality monitoring robot k to the base station p B (x B , y B ) is:

[0147] d kB (ε) = ||p k (ε) - p B || (13);

[0148] Step 6.3: Define d θ as the maximum distance for establishing communication between two water quality monitoring robots and between a water quality monitoring robot and a base station; the d θ value can be determined according to the channel fading parameter, communication transmission power, and antenna gain of the water quality monitoring task water area; then, when the distance between any two water quality monitoring robots S k , S l and between a water quality monitoring robot and a base station satisfies d kl (ε) ≤ d θ , d kB (ε) ≤ d θ , a connection relationship is established between them;

[0149] Use the symbol ψ kl = 1 to indicate that the water quality monitoring robots S k , S l establish a connection, otherwise ψ kl = 0; ψ kB = 1 to indicate that the water quality monitoring robot S k establishes a connection with the base station, otherwise ψ kB = 0;

[0150] Step 6.4: At any time ε, if S k wants to transmit a data packet to the base station, the communication network link of the water quality monitoring robot needs to satisfy:

[0151]

[0152] Equation ( 14 ) means that in the water quality monitoring robots on the island platform, any one water quality monitoring robot establishes a connection relationship with at least one other water quality monitoring robot, and at least one water quality monitoring robot can establish a connection with the base station. The above network is called a fully connected network.

[0153] Step S7: Perform a full connectivity determination on the communication network of the water quality monitoring robot formation; specifically as follows:

[0154] Step 7.1: In Step S4, the task point matrix corresponding to each water quality monitoring robot is obtained, and the task point matrix T k corresponding to the water quality monitoring robot S kIn it, each water quality monitoring point and the water quality monitoring robot S l The corresponding task point matrix T l The distances between each pair of water quality monitoring points in it are used to construct a distance matrix For storing the corresponding distance information;

[0155] Step 7.2: If the maximum distance between the task points of the water quality monitoring robots S k and S l satisfies max(D kl ) ≤ d θ , it means that S k and S l can always remain connected during the task process, that is

[0156]

[0157] where, T k , T l is the time taken for the water quality monitoring robot to complete a round of monitoring tasks;

[0158] Step 7.3: According to the definition of a fully connected network in step S6 and combined with equation (15), if at any time all water quality monitoring robots have at least one data transmission link to the base station, then at all times, the formation of water quality monitoring robots constitutes a fully connected network, called an all-time - fully connected network;

[0159] Step 7.4: If the distance matrix D kl does not satisfy max(D kl ) ≤ d θ , but there exist α ∈ {1, 2,..., N k}, β ∈ {1, 2,..., N l} such that D αβ ≤ d θ ; in other words, there is a distance D kl between two water quality monitoring points in the distance matrix D αβ less than or equal to the connection threshold d θ , which means that when the water quality monitoring robots S k and S l are at these two water quality monitoring points respectively, data forwarding can be completed, and correspondingly, it is called a time - segmented - fully connected network;

[0160] Step 7.5: If in the distance matrix D kl , min(D kl ) > d θ > 0, it means that the water quality monitoring robots S k and S l cannot establish a connection throughout the task process, called a non - fully connected network.

[0161] Step S8: Construct and optimize the communication link in the fully connected network state; as shown in the appendix Figure 4 as follows:

[0162] Step 8.1: For the all-time - fully connected network, if the water quality monitoring robot S k has a data transmission requirement, by constructing the connection line S k (x k , y k ) with the base station location p B (x B , y B ), the straight line equation of S k p B , S k p B is:

[0163]

[0164] S k is the water quality monitoring robot connected to it at the current moment. For example, if S l (x l , y l ) can be connected to it, then the foot of the perpendicular from S l to the straight line S k p B is calculated as follows: as follows:

[0165]

[0166] Furthermore, the distance from to the base station p B (x B , y B ) is obtained:

[0167]

[0168] Taking as the evaluation index for relay selection, the smaller the value of, the greater the effective distance for the data packet of the water quality monitoring robot S k to advance, and the greater the probability for the water quality monitoring robot to be selected as a relay;

[0169] S k selects the water quality monitoring robot S l with the smallest l as the next-hop information relay; similarly, S l continues to find the next relay node according to the above method until the data packet is sent to the base station;

[0170] Step 8.2: For the time - segmented full - connected network, it is necessary to optimize the waiting time of the water quality monitoring robot. In step S7.4, there exist α ∈ {1, 2,..., N k}, β ∈ {1, 2,..., N l}, such that D αβ ≤d θ . At this time, the water quality monitoring robots S k and S l can establish a connection. If the water quality monitoring robot S k has a sending requirement, when there are water quality monitoring points, it can wait for S l to reach its water quality monitoring point for data transmission; the waiting transmission time of S k can be reduced by optimizing the initial positions and speeds of S k and S l .

[0171] When the data transmission from S k to S l is completed, S l continues to forward the data packet to the next water quality monitoring robot node according to the method in step S8.1 or by waiting until it reaches the base station.

[0172] Step S9: Construct and optimize the communication link in the non - fully - connected network state; specifically as follows:

[0173] Step 9.1: In the water quality monitoring robot formation network, if some water quality monitoring robot nodes cannot join the network, new waypoints need to be added to the route of this node or the surrounding water quality monitoring robot nodes to make it join the data transmission network.

[0174] According to the method in step S7, it is determined that the water quality monitoring robot node S k cannot join the network, and at the same time, among the water quality monitoring points T k corresponding to S k , there is a point that is the closest to the water quality monitoring point l corresponding to the water quality monitoring robot S , denoted as and S l can join the network. S k selects S l as its relay communication node; otherwise, look for the next water quality monitoring robot that meets the above conditions.

[0175] Step 9.2: Each of the water quality monitoring robots S k and S l finds a point Make these two points meet the connectivity requirement, and the water quality monitoring robot S k and S l The total moving distance is minimized, and the corresponding model is as follows:

[0176]

[0177] The above formula (19) can be solved by using the improved osprey optimization algorithm in step S5; in addition, when multiple nodes require S l For relaying, the above formula also applies, as long as the dimension of the solution is increased;

[0178] Step 9.3: Insert As a task point into the matrix T k In the element After that, it is used as the waypoint of the water quality monitoring robot S k ; Insert As a task point into the matrix T l In the element After that, it is used as the waypoint of the water quality monitoring robot S l ;

[0179] Step 9.4: The water quality monitoring robot S k Performs data transmission according to step S8.

[0180] As shown in the appendix Figure 5 As shown, according to the above process, the present invention obtains a formation composed of 9 water quality monitoring robots within a range of 10 km × 10 km, and the corresponding task allocation result and communication link schematic diagram.

[0181] From the above results, it can be seen that the method of the present invention can realize the task planning of the water quality monitoring robot formation, improve the connectivity of the network, and ensure the smooth transmission of water quality monitoring data. This has positive significance for the field of water quality monitoring.

[0182] In the description of the present invention, it should be understood that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying 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 construed as a limitation of the present invention.

[0183] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing and constructing a communication network for a water quality monitoring robot formation, characterized in that: This method is implemented using the following steps: Step S1: determine the water quality monitoring task water area and establish the water area map and coordinate system; Step S2: Determine the coordinates of all water quality monitoring points in the water quality monitoring task area C = {c i (x i ,y i )|i=1,2,...,N c }, where N c is the number of water quality monitoring points; Step S3: Input the water quality monitoring robot formation situation, the number of water quality monitoring robots is N S , the water quality monitoring robot formation is represented by S = {s k |k=1,2,..,N S }; Step S4: assigning tasks to the water quality monitoring robot formation and establishing a task assignment model; Step S5: using the Latin hypercube sampling strategy and combining the chaotic mapping mechanism to improve the Osprey optimization algorithm to improve the algorithm performance, and using the improved Osprey optimization algorithm to solve the task allocation model; Step S6: Establishing a fully connected communication model of the water quality monitoring robot; Step S7: Conducting full connectivity determination on the communication network of the water quality monitoring robot formation; Step S8: constructing and optimizing the communication link in the fully connected network state; Step S9: constructing and optimizing the communication link in the non-fully connected network state; The step S5: using the Latin hypercube sampling strategy and combining the chaotic mapping mechanism to improve the Osprey optimization algorithm to improve the algorithm performance, and using the improved Osprey optimization algorithm to solve the task allocation model; the details are as follows: Step 5.1: Use the Latin hypercube sampling strategy to improve the population diversity of the osprey optimization algorithm: In the osprey optimization algorithm, the osprey population initialization is performed randomly. After the improvement of the Latin hypercube sampling strategy, the osprey population initialization formula can be rewritten as: X = LHS(n,D)(5); Where n is the population size; D is the dimension of the problem; LHS is the process of generating n D-dimensional samples using Latin hypercube sampling; X gh is the value of the g-th sample in the h-th dimension; L h and U h are the lower and upper bounds of the h-th dimension variable respectively; P gh is a random sequence on the interval [0,1); Step 5.2: Use the chaotic mapping mechanism to further perturb the osprey population, improve search diversity, and form a new sequence; X gh =X gh (1-φx t ) (7); Among them, x t is the value generated by the chaotic mapping, φ is the mapping control parameter; Step 5.3: Take the water quality monitoring robot formation task allocation model as the optimization object, let For each Osprey individual X g , calculate the fitness value f(X g ); Step 5.4: The osprey hovers to perform a global search, and its position update formula is as follows: X g (t+1)=X b (t)+R1sin(2πR2)(X b (t)-X g (t)) (9); Among them, X b (t) is the optimal position of the current osprey individual; R1, R2∈[0,1] are random number coefficients generated in the closed interval from 0 to 1. R1 is used to control the amplitude of the osprey update, and R2 is used to control the direction of the osprey position update; Step 5.5: The osprey approaches the prey for a local search, and the position of the individual osprey is updated as follows: X g (t+1)=X b (t)+R3(X b (t)-X g (t)) (10); Among them, the random number generated by R3∈[0,1] is used to adjust the degree of proximity of the osprey individual to the optimal solution; Step 5.6: After each update, continue to use chaotic mapping to further perturb the individual positions of the ospreys near the optimal solution to avoid the local optimality of the algorithm: X gh (t)=X gh (t+1)(1-φx t ) (11); Step 5.7: Determine whether the maximum number of iterations is reached or the fitness value meets the accuracy requirement. If so, output the optimal solution, otherwise return to step 5.4; The step S6: establishing a fully connected communication model for the water quality monitoring robot; the details are as follows: Step 6.1: At any time ε, any two water quality monitoring robots k,l∈1,2,...,N S , the corresponding positions are p k (ε),p l (ε), the distance between the two is obtained as: d kl (e)=||p k (e)-p l (e)|| (12); Step 6.2: At any time ε, water quality monitoring robot k arrives at base station p B (x B ,y B ) is: d kB (e)=||p k (e)-p B || (13); Step 6.3: Define d θ is the maximum distance for establishing communication between two water quality monitoring robots and between a water quality monitoring robot and a base station; then, When any two water quality monitoring robots S k ,S l The distance between the water quality monitoring robot and the base station satisfies d kl (ε)≤d θ ,d kB (ε)≤d θ When , a connection relationship is established between the two; Using the symbol ψ kl =1 means water quality monitoring robot S k ,S l Establish a connection, otherwise ψ kl =0;ψ kB =1 means water quality monitoring robot S k Establish a connection with the base station, otherwise ψ kB =0; Step 6.4: At any time ε, if S k To transmit data packets to the base station, the water quality monitoring robot communication network link needs to meet the following requirements: Formula (14) shows that in N S Among the water quality monitoring robots, any one of them can establish a connection with at least one other water quality monitoring robot, and at least one of them can establish a connection with the base station. The above network is called a fully connected network. The step S7: performing full connectivity determination on the communication network of the water quality monitoring robot formation; specifically as follows: Step 7.1: In step S4, the task point matrix corresponding to each water quality monitoring robot is obtained. Obtain the water quality monitoring robot S k The corresponding task point matrix T k In each water quality monitoring point and water quality monitoring robot S l The corresponding task point matrix T l The distance between each water quality monitoring point in the Used to store corresponding distance information; Step 7.2: If the water quality monitoring robot S k and S l The maximum distance between the task points satisfies max(D kl )≤d θ , which means S k and S l The connection can always be maintained during the task, that is, Among them, T k ,T l The time it takes for the water quality monitoring robot to complete a round of monitoring tasks; T s It represents the maximum time taken by two water quality monitoring robots to complete their respective rounds of monitoring tasks; Step 7.3: According to the definition of the fully connected network in step S6, combined with equation (15), if at any time, all water quality monitoring robots have at least one data transmission link to the base station, then at all times, the water quality monitoring robot formation constitutes a fully connected network, which is called a full-time fully connected network; Step 7.4: If the distance matrix D kl Does not satisfy max(D kl )≤d θ , but there exists α∈{1,2,...,N k },β∈{1,2,...,N l }, so that D αβ ≤d θ ; In other words, the distance matrix D kl There are two water quality monitoring points at a distance D αβ Less than or equal to the connectivity threshold d θ , which means that the water quality monitoring robot S k and S l When the two water quality monitoring points are respectively located, data forwarding can be completed, and accordingly, it is called a time-sharing-full-connectivity network; Step 7.5: If the distance matrix D kl In, min(D kl )>d θ >0, means water quality monitoring robot S k and S l The inability to establish a connection during the entire mission is called a non-fully connected network; The step S8: constructing and optimizing the communication link in the fully connected network state; the details are as follows: Step 8.1: For the full-time and full-connection network, if the water quality monitoring robot S k There is a need to send data, by building a local location S k (x k ,y k ) and the base station location p B (x B ,y B ) k p B , S k p B The equation of the straight line is: S k To query the water quality monitoring robot currently connected to it, S l (x l ,y l ) can be connected with it, then S l To the straight line S k p B The vertical foot The calculation is as follows: Further seek To base station p B (x B ,y B ) distance: Will As an evaluation metric for relay selection, The smaller the value of, the better the water quality monitoring robot S k The greater the effective distance that the data packet travels, the greater the probability that the water quality monitoring robot will be selected as a relay; S k choose The smallest water quality monitoring robot l As the next hop information relay; similarly, S l Continue to search for the next relay node according to the above method until the data packet is sent to the base station; Step 8.2: For the time-segmented and fully connected network, it is necessary to optimize the waiting time of the water quality monitoring robot. In step 7.4, there exists α∈{1,2,...,N k },β∈{1,2,...,N l }, so that D αβ ≤d θ , at this time the water quality monitoring robot S k and S l A connection can be established if the water quality monitoring robot S k If you need to send, you can When there are water quality monitoring points, wait for S l Arrive at water quality monitoring points for data transmission; S k The waiting time for transmission can be optimized by k and S l The initial position and speed are reduced; When S k To S l After the data transmission is completed, S l Continue forwarding the data packet to the next water quality monitoring robot node according to the method of step 8.1 or by waiting until it reaches the base station; The step S9: constructing and optimizing the communication link in the non-fully connected network state; the details are as follows: Step 9.1: In the water quality monitoring robot formation network, if some water quality monitoring robot nodes cannot join the network, it is necessary to add new track points to the route of the node or the surrounding water quality monitoring robot nodes to join them into the data transmission network; According to the method of step S7, determine the water quality monitoring robot node S k Unable to join the network and obtain S k Corresponding water quality monitoring point T k In, there is a little Distance water quality monitoring robot S l Corresponding water quality monitoring points Recently, it is recorded as And S l Can join the network, S k Select S l As its relay communication node; otherwise, find the next water quality monitoring robot that meets the above conditions; Step 9.2: Water Quality Monitoring Robot S k and S l Find a point each This makes these two points meet the connectivity requirements, and the water quality monitoring robot S k and S l The total distance moved is the smallest, and the corresponding model is as follows: The above formula (19) can be solved using the improved Osprey optimization algorithm in step S5; in addition, when there are multiple nodes that need S l When relaying, the above formula also applies, just increase the dimension of the solution; Step 9.3: Insert it into the matrix T as a task point k Medium Element Later, as a water quality monitoring robot S k The track point of Insert it into the matrix T as a task point l Medium Element Afterwards, as a water quality monitoring robot S l Track points; Step 9.4: Water Quality Monitoring Robot S k According to step S8, data transmission is performed.

2. A method for optimizing and constructing a communication network for a water quality monitoring robot formation according to claim 1, characterized in that: The step S4: assigning tasks to the water quality monitoring robot formation and establishing a task assignment model; the details are as follows: Step 4.1: The distance between any two water quality monitoring points i and j is recorded as: d ij =||c i -c j || (1); Step 4.2: Water quality monitoring robot with v opt When the speed is moving at a constant speed, the instantaneous energy consumption E opt Minimum, to complete the movement between two water quality monitoring points i, j, the corresponding energy consumption is: Step 4.3: Create decision variables When the water quality monitoring robot k moves from the water quality monitoring point i to the water quality monitoring point j, let otherwise Step 4.4: Create decision variables When the water quality monitoring robot k takes the water quality monitoring point i as its task point, otherwise Step 4.5: Create decision variables It is used to control the water quality monitoring robot k to perform water quality testing on the water quality monitoring points in order; Step 4.6: Establish a water quality monitoring robot formation task allocation model; in, is the water quality monitoring task point corresponding to the kth water quality monitoring robot; N s Represents the total number of water quality monitoring points, N c represents the total number of water quality monitoring robots; are binary variables, Indicates that the water quality monitoring robot k is assigned to task point i, otherwise it is 0; E k ,E l represents the instantaneous energy consumption of water quality monitoring robot k and water quality monitoring robot l; Constraint 1 limits each water quality monitoring point to only one water quality monitoring robot for testing; Constraint 2 restricts that if the water quality monitoring robot completes the detection of water quality monitoring point i, it must enter the next monitoring point; Constraint 3 limits the task route of each water quality monitoring robot to a complete loop; Constraint 4 limits the fairness of task allocation, where λ is the energy consumption difference adjustment coefficient, which ensures that the energy consumption difference between water quality monitoring robots is limited to an acceptable range.

3. The method for optimizing and constructing a communication network for a water quality monitoring robot formation according to claim 1, characterized in that: The d θ The value of can be determined based on the channel fading parameters, communication transmission power and antenna gain of the water quality monitoring mission area.

Images