A multi-inspection robot combined cycle pipe corridor inspection planning system and method

Abstract

This invention proposes a multi-robot joint periodic utility tunnel inspection planning system and method. The invention constructs arrays for robot numbers, robot order, maximum speed, number of monitoring points, data interaction capabilities, interaction data volume, monitoring point numbers, and sensor node numbers. It then sequentially calculates the total interaction data volume at each monitoring point, the inspection time of each robot at a monitoring point, and the single inspection cycle of each robot. Next, it establishes constraints on the number of monitoring points, the inspection cycle of the utility tunnel inspection system, and optimization objectives, thus establishing an optimization problem for the utility tunnel inspection system. Finally, it solves this optimization problem using an iterative particle swarm optimization algorithm to obtain the optimal planning scheme for the utility tunnel inspection system. This invention maximizes the inspection capabilities of multiple robots while minimizing the single inspection cycle of the periodic inspection system, all while meeting the periodic inspection requirements of the utility tunnel sensor system.

Description

Technical Field

[0001] This invention belongs to the field of unmanned inspection technology, and in particular relates to a multi-inspection robot joint periodic pipe gallery inspection planning system and method. Background Technology

[0002] With the continuous development of science and technology, small electronic devices, represented by various sensors, are gradually becoming miniaturized and intelligent. This allows sensors to be flexibly deployed in various complex environments. Furthermore, by endowing sensor nodes with networking capabilities, sensors can form an Internet of Things (IoT) through wireless communication, collecting various environmental parameters in a timely manner. Therefore, sensors have been widely installed and deployed in urban underground utility tunnel networks. Utility tunnels centrally lay various pipelines, such as natural gas pipelines, power cables, water supply pipelines, heating pipelines, and sewage treatment pipelines, providing a large amount of basic services to the city and becoming its lifeline. The construction and maintenance costs of utility tunnels are enormous, and disasters could cause significant losses and severe damage. Real-time monitoring of the status of utility tunnels has high economic value, and typically, a dense network of various types of sensors is deployed within them, forming a large-scale sensor network. The environment within utility tunnels is complex, requiring real-time monitoring of numerous data indicators, resulting in a massive amount of data transmitted by the sensors. Traditionally, sensor data is collected and summarized through manual inspections. However, utility tunnels can be tens of kilometers long, with complex environments, and the space within the tunnels becomes relatively cramped after wiring, hindering personnel movement. Manually collecting and recording data from a large number of sensors is costly, unsafe, and extremely inefficient.

[0003] To improve the efficiency and reduce the cost of periodic inspections, many scholars have proposed various solutions to address the aforementioned problems: reducing the number of sensors or the amount of data collected, improving sensor location distribution, transmitting data via wireless communication, and using inspection robots. Reducing the number of sensors or the amount of data collected, and improving sensor location distribution, often come at the cost of sacrificing sensor network accuracy, which defeats the purpose of using a utility tunnel sensor system. If data is wirelessly transmitted between sensors, the massive amount of data will generate enormous power demands, negatively impacting the lifespan of the sensor system. By using inspection robots for periodic inspections, combined with wireless data transmission technology, the robots only need to collect data from nearby sensors, minimizing power consumption. Furthermore, the robots can move at high speeds along the utility tunnel inspection track, efficiently completing periodic inspection tasks. Therefore, in the application scenario of urban underground utility tunnels, using inspection robots to collect sensor data is the best choice for improving the periodic inspection system.

[0004] The sensor network deployed in utility tunnels is characterized by several features: First, a high number of sensors are densely distributed. If a periodic, fixed-point, timed data collection method is used—that is, the inspection robot interacts with all sensor nodes for a predetermined period before leaving—the periodic inspection would be too lengthy, hindering the overall inspection efficiency of the sensor network. Second, utility tunnels typically house various types of sensors, such as temperature sensors, humidity sensors, flammable and explosive gas sensors, and toxic gas sensors, or some nodes may use integrated sensors. This results in significant differences in the amount of data collected by different sensors, making it unreasonable to use the same data interaction time. Therefore, for sensor networks in utility tunnels, there is an urgent need to propose a multi-robot periodic inspection scheme that allows for on-demand interaction between densely distributed, multi-type sensors while minimizing the time cycle.

[0005] Traditional strategies for periodically training multi-type sensor systems using inspection robots typically involve the robot finding the nearest sensor, moving to that location, stopping, and then interacting with it. Once that sensor has collected the data, the interaction stops, and the robot moves on to the next sensor. This approach doesn't consider the robot's data interaction with sensors during movement, resulting in significant additional inspection time. Furthermore, due to varying robot performance (e.g., different speed limits, data interaction capabilities), when multiple robots with different performance characteristics form a periodic inspection system, the impact of these performance differences on inspection efficiency must be considered, further increasing the complexity of the problem.

[0006] To address the aforementioned issues, this invention proposes a multi-inspection robot joint periodic pipe gallery inspection planning system based on an improved particle swarm optimization algorithm. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention proposes a multi-inspection robot joint periodic pipe gallery inspection planning system and method.

[0008] The technical solution of this invention is a multi-inspection robot combined periodic utility tunnel inspection planning system, comprising:

[0009] Multiple inspection robots and multiple sensor nodes;

[0010] An inspection track is laid along the inside of the pipe gallery, and the multiple inspection robots are placed on the inspection track in sequence to move.

[0011] Multiple sensor nodes are evenly fixed at different locations inside the pipe gallery;

[0012] Set the start and end points of the inspection track, and evenly divide the area between the start and end points of the inspection track into multiple monitoring points.

[0013] Each inspection robot interacts wirelessly with multiple sensor nodes at the monitoring point in sequence.

[0014] Construct an array of inspection robot numbers, a maximum speed array for each inspection robot, an array of the number of initial monitoring points each inspection robot is responsible for, an array of data interaction capabilities for each inspection robot, an array of monitoring points, an array of interaction data volume for all types of sensor nodes, and an array of the number of each type of sensor node near each monitoring point. Construct the initial setup order for the inspection robots, and sequentially calculate the total interaction data volume for each monitoring point, the inspection time of the inspection robot at each monitoring point, and the single inspection cycle of each inspection robot. Sequentially construct the constraints on the number of monitoring points, the inspection cycle constraints of the utility tunnel inspection system, and the optimization objective of the utility tunnel inspection system, and establish the optimization problem of the utility tunnel inspection system. Solve the optimization problem of the utility tunnel inspection system using an improved particle swarm optimization algorithm to obtain the optimal periodic inspection system design scheme.

[0015] The technical solution of this invention is a method for planning periodic pipe gallery inspections using multiple inspection robots, and the specific steps are as follows:

[0016] Step 1: Using the midline between two adjacent monitoring points as the boundary between them, set the number and type of sensor nodes for each monitoring point, obtain the distance between two adjacent monitoring points, obtain the amount of interactive data for each type of sensor node, obtain the number of inspection robots, and obtain the maximum speed and data interaction speed of each inspection robot.

[0017] Step 2: Construct an array of inspection robot numbers, construct the initialization setting order of the inspection robots based on the inspection robot numbers, construct an array of maximum speed for each inspection robot, an array of the number of initial monitoring points each inspection robot is responsible for, an array of data interaction capabilities for each inspection robot based on the initialization setting order of the inspection robots, construct an array of monitoring points, construct an array of the interaction data volume of all types of sensor nodes based on the interaction data volume of each type of sensor node, and construct an array of the number of each type of sensor node at each monitoring point;

[0018] Step 3: Calculate the total amount of interactive data at each monitoring point, the inspection time of the inspection robot at each monitoring point, and the single inspection cycle of each inspection robot in sequence, using the setting order of the inspection robots and the number of monitoring points that each inspection robot is responsible for as variables.

[0019] Step 4: Sequentially construct the constraints on the number of monitoring points, the inspection cycle of the utility tunnel inspection system, and the optimization objective of the utility tunnel inspection system to establish the optimization problem of the utility tunnel inspection system;

[0020] Step 5: Solve the optimization problem of the pipe gallery inspection system using the successive iterative particle swarm optimization algorithm to obtain the optimal setting order of the inspection robots and the optimal number of monitoring points each inspection robot is responsible for. Each inspection robot moves on the pipe gallery inspection track according to the optimal setting order and the optimal number of monitoring points it is responsible for, and performs periodic inspections of the sensor nodes.

[0021] Preferably, the inspection robot number array in step 2 is defined as follows:

[0022]

[0023] in, For the inspection robot's number array, The inspection robot is numbered, and K represents the number of inspection robots. The initialization setup sequence for the inspection robots in step 2 is defined as follows:

[0024] Δk (0) =[1,2,..,k,..,K]

[0025] Where, Δk (0) Set the initialization sequence array for the inspection robots, where k is the sequence number of the inspection robots and K represents the number of inspection robots;

[0026] The maximum speed array for each inspection robot described in step 2 is defined as follows:

[0027] Δv max =[v 1,max ,v 2,max ...,v k,max ,..,v K,max ]

[0028] Where, Δv max The array of maximum speeds for each inspection robot, v k,max The maximum speed of the inspection robot in the kth setting sequence, where k is the setting sequence number of the inspection robot and K represents the number of inspection robots;

[0029] The array of initial monitoring points for each inspection robot described in step 2 is defined as follows:

[0030] ΔL (0) =[L1,L2,..,L k ,..,L K ]

[0031] Where, ΔL (0) An array containing the number of initial monitoring points for each inspection robot, L. kThe number of monitoring points that the inspection robot in the kth set sequence is responsible for, where k is the set sequence number of the inspection robot and K represents the number of inspection robots;

[0032] The data interaction capability array for each inspection robot described in step 2 is defined as follows:

[0033] Δp=[p1,p2,..,p k ,..,p K ]

[0034] Where Δp is the data interaction capability array of the inspection robot, p k The data interaction capability for the kth inspection robot with the set order, where k is the order number of the inspection robot and K represents the number of inspection robots;

[0035] The monitoring point array mentioned in step 2 is defined as follows:

[0036] Δm=[1,2,..,m,..,M]

[0037] Where Δm is the array of monitoring points, m is the monitoring point number, and M is the number of monitoring points.

[0038] The array of interactive data volume for all types of sensor nodes mentioned in step 2 is defined as follows:

[0039] Δd=[d1,d2,..,d n ,..,d N ]

[0040] Where Δd is the array of interactive data volume of all sensor nodes, n is the sensor node type number, N is the number of sensor node types, and d n Let be the amount of interactive data from the nth sensor.

[0041] The array of the number of each type of sensor node near each monitoring point in step 2 is defined as follows:

[0042]

[0043] Where Δa is an array representing the number of each type of sensor node near each monitoring point, n is the sensor node type number, N is the number of sensor node types, m is the monitoring point number, M is the number of monitoring points, and a n,m This represents the number of nth type of sensor nodes near the mth monitoring point.

[0044] Preferably, the total amount of interactive data for each monitoring point in step 3 is defined as follows:

[0045]

[0046] Among them, Q mLet d be the total amount of interactive data at the m-th monitoring point. n Let a be the amount of interactive data from the nth sensor. n,m Let N be the number of sensor nodes of type n near the m-th monitoring point, where n is the sensor node type number, N is the number of sensor node types, m is the monitoring point number, and M is the number of monitoring points.

[0047] The inspection time of the inspection robot at each monitoring point in step 3 is defined as follows:

[0048]

[0049] Among them, t m Let Q be the inspection time of the inspection robot at the m-th monitoring point. m p represents the total amount of interactive data at the m-th monitoring point. k For the data interaction capability of the k-th inspection robot with set order, L k The number of monitoring points handled by the inspection robot in the k-th order, where Δl is the distance between monitoring points, and v k,max This represents the maximum speed of the inspection robot in the k-th setting sequence, where k is the robot setting sequence number, K represents the number of inspection robots, m is the monitoring point number, and M is the number of monitoring points. For from L1 to L k The summation operation is called `max{*,*}`, which is the operation to find the maximum value.

[0050] The single inspection cycle of each inspection robot mentioned in step 3 is defined as follows:

[0051]

[0052] Among them, T k Let t be the single inspection cycle of the k-th inspection robot. m+a Let L be the inspection time of the inspection robot at the (m+a)th monitoring point. k-1 This represents the number of monitoring points handled by the (k-1)th inspection robot in the assigned sequence, where m is the monitoring point number, k is the robot's assigned sequence number, and K represents the total number of inspection robots. For array [L1,L2,..,L] k Summation operation;

[0053] The constraint on the number of monitoring points mentioned in step 4 is defined as follows:

[0054]

[0055] Among them, L kThis represents the number of monitoring points that the inspection robot in the k-th setup sequence is responsible for, where k is the robot setup sequence number, K represents the number of inspection robots, and M is the number of monitoring points. For from L1 to L k Summation operation;

[0056] The inspection cycle constraint of the utility tunnel inspection system described in step 4 is defined as follows:

[0057]

[0058] Among them, T k Let k be the single inspection cycle of the inspection robot. The single inspection cycle of the utility tunnel inspection system is defined as k, where k is the sequence number of the inspection robot and K represents the number of inspection robots.

[0059] The optimization objective of the utility tunnel inspection system described in step 4 is defined as minimizing the single inspection cycle of the utility tunnel inspection system by optimizing the setting sequence of inspection robots and the number of monitoring points each inspection robot is responsible for:

[0060]

[0061] in, Let Δk be the single inspection cycle of the utility tunnel inspection system, Δk be the sequence array for setting the inspection robots, and ΔL be the array representing the number of monitoring points each inspection robot is responsible for. To maximize b by optimizing a;

[0062] The optimization problem of the utility tunnel inspection system described in step 4 is established as follows:

[0063] By optimizing the order in which the inspection robots are set up and the number of monitoring points each robot is responsible for, the single inspection cycle of the utility tunnel inspection system is minimized to:

[0064]

[0065]

[0066]

[0067] Where k is the inspection robot number, K is the number of inspection robots, M is the number of monitoring points, and L is the number of monitoring points. k The number of monitoring points that the k-th inspection robot is responsible for. For a single inspection cycle of the utility tunnel inspection system, T k Let k be the single inspection cycle of the k-th robot. For from L1 to L k The summation operation is performed, where Δk is the sequence array for the inspection robots, and ΔL is the array representing the number of monitoring points each inspection robot is responsible for. The goal is to maximize the operation of b by optimizing a.

[0068] Step 5, which describes solving the optimization problem of the pipe gallery inspection system using the successive iterative particle swarm optimization algorithm, includes the following steps:

[0069] Step 5.1: Initialize the position, velocity, and fitness value of each particle; calculate the initial iterative optimal fitness value and the initial iterative optimal particle position; calculate the global optimal fitness value and the global optimal particle position; and generate the mapping matrix.

[0070] Step 5.2: Calculate the particle position array and fitness value for the next iteration, update the iterative optimal fitness value and iterative optimal particle position, update the global optimal fitness value and global optimal particle position, and update the velocity of each particle for the next iteration;

[0071] Step 5.3: Determine the stopping condition for iteration. If the iteration stops, output the final value; otherwise, return to step 5.2.

[0072] The specific implementation of the iterative particle swarm optimization algorithm described in step 5 is as follows:

[0073] The position of each initialized particle described in step 5.1 is defined as:

[0074] A p (0) =[ΔK (0) ,ΔL (0) ],p=1,2,..,P

[0075] Among them, A p (0) Let p be the initial position of the p-th particle, where p is the particle number, P is the number of particles, and ΔK is the initial position of the particle. (0) To initialize the inspection robot, set a sequence array, ΔL (0) An array representing the number of initial monitoring points each inspection robot is responsible for, where [*,*] is the constituent array;

[0076] The initial particle velocity described in step 5.1 is defined as:

[0077] V p (0) =[Rand p (1)*K! ,Rand p (K-1),M-∑Rand p [(K-1)], p = 1, 2, ..., P

[0078] in, Let P be the initial velocity of the p-th particle, where p is the particle number, P is the number of particles, M is the number of monitoring points, K is the number of inspection robots, and Rand is the initial velocity of the particle. p(*) represents the array of length * generated randomly in the p-th iteration, ∑* represents the summation operation on array *, [*,*] represents the combined array operation, and *! represents the order multiplication calculation with respect to *.

[0079] The initial fitness value of each particle in step 5.1 is defined as follows:

[0080]

[0081] in, For located The initial fitness value of the particles. Let p be the initial position of the p-th particle, where p is the particle number, P is the number of particles, k is the inspection robot number, and K is the number of inspection robots. To initialize the single inspection cycle of the utility tunnel inspection system, Let Ω be the initial single inspection cycle for the k-th robot, and let Ω be a constant maximum parameter. For array [L1,L2,..,L] k Summation operation;

[0082] Step 5.1, which initializes the iterative optimal fitness value and the iterative optimal particle position, is defined as follows:

[0083]

[0084] Among them, F (0) * represents the initial iterative optimal fitness value. For located The initial fitness value of the particles. Let p be the initial position of the p-th particle. Let F be the initial iteration optimal fitness. (0) The initial particle position corresponding to * is also the initial position of the p*th particle, where p is the particle number and P is the number of particles. This is for calculating the maximum value.

[0085] The global optimal fitness value and the global optimal particle position mentioned in step 5.1 are defined as follows:

[0086]

[0087] Where F* is the global optimal fitness value, A* is the global optimal particle position, and F( 0 * represents the initial iteration optimal fitness value. Let F be the initial iteration optimal fitness. (0) The initial particle position corresponding to * is also the initial position of the p*th particle, where p is the particle number;

[0088] The mapping matrix generated in step 5.1 is defined as follows:

[0089]

[0090] Where Q is a mapping matrix of size K! * K, i is the row count of mapping matrix Q, k is the column count of mapping matrix Q, K is the number of inspection robots, Q(i,k) is the element in the i-th row and k-th column of mapping matrix Q, Q(i,q) is the element in the i-th row and q-th column of mapping matrix Q, and (*)! represents the multiplication calculation.

[0091] The first element of the particle position array for the next iteration, as described in step 5.2, is defined as follows:

[0092]

[0093] Where, ΔA p (r+1) Let ΔA be the position of the p-th particle in the (r+1)-th iteration. p (r) Let ΔA be the position of the p-th particle in the r-th iteration, where r is the iteration number, p is the particle number, P is the number of particles, and ΔA is the position of the p-th particle in the r-th iteration. p (r+1) (1) represents the first element of the position of the p-th particle in the (r+1)-th iteration, Q(*,:) represents the *-th row of the array that makes up the mapping matrix Q, and Δt is a constant time parameter. Let p be the velocity of the p-th particle in the r-th iteration. This indicates the rounding up operation;

[0094] The second to the Kth elements of the particle position array for the next iteration, as described in step 5.2, are defined as follows:

[0095]

[0096] Where, ΔA p (r+1) Let P be the position of the p-th particle in the (r+1)-th iteration, where r is the iteration number, p is the particle number, P is the number of particles, and ΔA is the position of the p-th particle in the (r+1)-th iteration. p (r+1) (2:K) represents an array consisting of the 2nd to the Kth elements of the position of the p-th particle in the (r+1)-th iteration, ΔA p (r) (2:K) represents an array consisting of the 2nd to the Kth elements of the position of the p-th particle in the r-th iteration, where Δt is a constant time parameter. Let p be the velocity of the p-th particle in the r-th iteration. This indicates the rounding up operation.

[0097] The (K+1)th element of the particle position array for the next iteration, as described in step 5.2, is defined as follows:

[0098] ΔA p (r+1) (K+1)=M-∑ΔA p (r+1) (2:K), p=1,2,..,P

[0099] Where, ΔA p (r+1) Let P be the position of the p-th particle in the (r+1)-th iteration, where r is the iteration number, p is the particle number, P is the number of particles, M is the number of monitoring points, and ΔA is the position of the p-th particle in the (r+1)-th iteration. p (r+1) (K+1) represents the (K+1)th element at the position of the p-th particle in the (r+1)-th iteration, ΔA p (r+1) (2:K) represents an array consisting of the 2nd to the Kth elements of the position of the p-th particle in the (r+1)-th iteration, ∑ * This represents the summation operation on all elements of array *.

[0100] The particle position array for the next iteration, as described in step 5.2, is defined as follows:

[0101] ΔA p (r+1) =[ΔA p (r+1) (1),ΔA p (r+1) (2:K),ΔA p (r+1) [(K+1)], p = 1, 2, ..., P

[0102] Where, ΔA p (r+1) Let ΔA be the position of the p-th particle in the (r+1)-th iteration. p (r+1) (1) represents the first element of the position of the p-th particle in the (r+1)-th iteration, ΔA p (r+1) (2:K) represents an array consisting of the 2nd to the Kth elements of the position of the p-th particle in the (r+1)-th iteration, ΔA p (r+1) (K+1) represents the (K+1)th element at the position of the p-th particle in the (r+1)-th iteration, and [*,*] represents the combined array operation.

[0103] Step 5.2 defines the fitness value of each particle for the next iteration as follows:

[0104]

[0105] in, For located The fitness value of the particles, Let p be the position of the p-th particle in the r-th iteration, where p is the particle ID, P is the number of particles, r is the iteration number, k is the inspection robot ID, and K is the number of inspection robots. Let r be the single inspection cycle of the utility tunnel inspection system in the r-th iteration. Let Ω be the single inspection cycle of the k-th robot in the r-th iteration, and let Ω be a constant maximum parameter. For from L1 to L k The summation operation.

[0106] The optimal fitness value and optimal particle position in step 5.2 are defined as follows:

[0107]

[0108] Among them, F (r) * represents the optimal fitness value in the r-th iteration. For located The particle fitness value, Let be the position of the p-th particle in the r-th iteration. For located The particle fitness value, The optimal fitness for the r-th iteration is F. (r) The * represents the particle position at time *, i.e., the initial position of the p*-th particle in the r-th iteration, where p is the particle number, P is the number of particles, and r is the iteration number. This is for calculating the maximum value.

[0109] Step 5.2, which updates the global optimal fitness value and the global optimal particle position, is defined as follows:

[0110]

[0111] Where F* is the global optimal fitness value, F (r) * represents the optimal fitness value in the r-th iteration, where r is the iteration number and p is the particle number. For located The particle fitness value, This represents the particle position when the global optimal fitness is F*, i.e., the position of the p*th particle in the r*th iteration. This is for calculating the maximum value.

[0112] Step 5.2 defines the update of the velocity of each particle in the next iteration as follows:

[0113]

[0114] in, Let p be the velocity of the p-th particle in the r-th iteration. Let ω be the velocity of the p-th particle in the (r+1)-th iteration, where r is the iteration number, p is the particle number, P is the number of particles, ω is the constant inertia coefficient, C1 is the constant individual learning factor, and C2 is the constant sociological coefficient factor. The first random number is within the interval [0,1]. F is a random number within the second interval [0,1]. (r) * represents the optimal fitness value in the r-th iteration, and F* represents the global optimal fitness value. For located The particle fitness value.

[0115] The convergence condition described in step 5.3 is defined as follows: [The condition is defined as follows] Where r is the number of iterations. The maximum number of iterations is given; if so, output the final value. The position of the particle corresponding to the global optimal fitness of F* is also the position of the p*th particle in the r*th iteration. According to step 5.2, restore Δk to set the order array for the inspection robot and ΔL to set the number of monitoring points for each inspection robot. The iteration ends; otherwise, update r = r + 1 and return to step 5.2.

[0116] The advantages of this invention are:

[0117] To address the shortcomings of existing periodic inspection strategies for utility tunnel sensor systems, and considering the periodic inspection requirements of utility tunnel sensors, this invention provides a periodic inspection strategy based on multiple utility tunnel inspection robots. This aims to maximize the performance of the inspection robots and improve the efficiency of the periodic inspection system. This invention proposes a method for quantifying inspection requirements based on the amount of data interaction between sensor nodes, a targeted joint design method for inspection schemes using multiple inspection robots, and a specific solution scheme based on an improved particle swarm optimization algorithm. The goal is to minimize the single inspection task cycle of the periodic inspection system while meeting the periodic inspection requirements of the utility tunnel sensor system.

[0118] This invention proposes a periodic inspection strategy for large-scale sensor networks in urban utility tunnels based on multiple inspection robots. This strategy involves inspection robots collecting data from sensor nodes via wireless data transmission. By acquiring system parameters of different types of sensors and inspection robots, a model is built for the periodic inspection system composed of sensors and inspection robots. Combined with a proposed method for quantifying inspection requirements, periodic inspection of sensor nodes is achieved. This problem is solved by optimizing the robot deployment order and the number of monitoring points each robot is responsible for, considering the constraint of the number of monitoring points, with the goal of minimizing the single inspection cycle of the utility tunnel inspection system. The solution yields an efficient allocation scheme for the single inspection task cycle of the periodic inspection system. This robot periodic inspection task scheme minimizes power overflow while meeting the periodic inspection requirements of sensor nodes, and simultaneously minimizes the single transmission task cycle of the inspection robots. It maximizes the inspection capabilities of multiple inspection robots while meeting the periodic inspection requirements of the utility tunnel sensor system, and simultaneously minimizes the single inspection task cycle of the periodic inspection system. Attached Figure Description

[0119] Figure 1 A simplified flowchart of the steps in an embodiment of the present invention;

[0120] Figure 2 : A schematic diagram of the monitoring boundary delineation according to an embodiment of the present invention;

[0121] Figure 3 : A schematic diagram illustrating the movement time of the inspection robot at different monitoring points according to an embodiment of the present invention;

[0122] Figure 4 : A schematic diagram illustrating the division of monitoring points by the inspection robot in this embodiment of the invention;

[0123] Figure 5 : Flowchart of the iterative particle swarm optimization algorithm according to an embodiment of the present invention. Detailed Implementation

[0124] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0125] In specific implementation, the method proposed in the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of this invention and computer equipment including the computer program running the corresponding computer program, should also be within the protection scope of this invention.

[0126] The technical solution of the system in the embodiment of the present invention is described below as a multi-inspection robot combined periodic pipe gallery inspection planning system, including:

[0127] Multiple inspection robots and multiple sensor nodes;

[0128] An inspection track is laid along the inside of the pipe gallery, and the multiple inspection robots are placed on the inspection track in sequence to move.

[0129] Multiple sensor nodes are evenly fixed at different locations inside the pipe gallery;

[0130] Set the start and end points of the inspection track, and evenly divide the area between the start and end points of the inspection track into multiple monitoring points.

[0131] Each inspection robot interacts wirelessly with multiple sensor nodes at the monitoring point in sequence.

[0132] Construct an array of inspection robot numbers, a maximum speed array for each inspection robot, an array of the number of initial monitoring points each inspection robot is responsible for, an array of data interaction capabilities for each inspection robot, an array of monitoring points, an array of interaction data volume for all types of sensor nodes, and an array of the number of each type of sensor node near each monitoring point. Construct the initial setup order for the inspection robots, and sequentially calculate the total interaction data volume for each monitoring point, the inspection time of the inspection robot at each monitoring point, and the single inspection cycle of each inspection robot. Sequentially construct the constraints on the number of monitoring points, the inspection cycle constraints of the utility tunnel inspection system, and the optimization objective of the utility tunnel inspection system, and establish the optimization problem of the utility tunnel inspection system. Solve the optimization problem of the utility tunnel inspection system using an improved particle swarm optimization algorithm to obtain the optimal periodic inspection system design scheme.

[0133] The inspection robot is a USR urban underground utility tunnel robot.

[0134] The sensor node is a BWM826 high-precision tilt sensor;

[0135] The following is combined Figures 1 to 5 The technical solution of the method in the embodiments of the present invention is a method for planning periodic pipe gallery inspections using multiple inspection robots. The specific steps are as follows:

[0136] Figure 1The diagram shows a simplified flowchart of the steps in an embodiment of the present invention. Step 1 includes obtaining the start and end points, dividing the monitoring points, and obtaining the number and type of sensors around each monitoring point, the distance between monitoring points, the amount of sensor data interaction, the number of inspection robots, their speed, and the data interaction speed. Step 2 includes constructing the setting order of the inspection robots, constructing the maximum speed set of the inspection robots, the set of the number of initial monitoring points they are responsible for, the set of data interaction capabilities, constructing the monitoring point set, and the set of sensor types and quantities near the monitoring points. Step 3 includes calculating the amount of data interaction at the monitoring points, calculating the inspection time at the monitoring points, and calculating the single inspection cycle of the inspection robots. Step 4 includes constructing constraints on the number of monitoring points, constructing system cycle constraints, constructing system optimization objectives, and establishing the problem. Step 5 includes using an improved particle swarm optimization algorithm to solve and restore the obtained results as the system cycle inspection scheme.

[0137] Step 1: Using the midline between two adjacent monitoring points as the boundary between them, set the number and type of sensor nodes for each monitoring point, obtain the distance between two adjacent monitoring points, obtain the amount of interactive data for each type of sensor node, obtain the number of inspection robots, and obtain the maximum speed and data interaction speed of each inspection robot.

[0138] Figure 2 The diagram shown is a schematic diagram of the monitoring boundary division according to an embodiment of the present invention. M = 10,000 monitoring points are equidistantly distributed on the inspection robot track. The midline between two adjacent monitoring points is used as the boundary between the two adjacent monitoring points. Four different types of sensor nodes are distributed on the inner wall of the pipe gallery parallel to the inspection robot track. Each sensor node is assigned to a unique monitoring point range.

[0139] Step 2: Construct an array of inspection robot numbers, construct the initialization setting order of the inspection robots based on the inspection robot numbers, construct an array of maximum speed for each inspection robot, an array of the number of initial monitoring points each inspection robot is responsible for, an array of data interaction capabilities for each inspection robot based on the initialization setting order of the inspection robots, construct an array of monitoring points, construct an array of the interaction data volume of all types of sensor nodes based on the interaction data volume of each type of sensor node, and construct an array of the number of each type of sensor node at each monitoring point;

[0140] Figure 3 The diagram shows the movement time of the inspection robot at different monitoring points according to an embodiment of the present invention. K=5 inspection robots are deployed sequentially on the inspection robot track, and the inspection robots spend different amounts of time moving within different ranges of M=10000 monitoring points.

[0141] Figure 4The diagram shows the division of monitoring points by the inspection robots according to an embodiment of the present invention. K=5 inspection robots are deployed sequentially on the inspection robot track, and each inspection robot is responsible for inspecting a corresponding number of monitoring points.

[0142] The inspection robot number array mentioned in step 2 is defined as follows:

[0143]

[0144] in, For the inspection robot's number array, The inspection robots are numbered, with K=5 representing the number of inspection robots. The initialization setup sequence for the inspection robots in step 2 is defined as follows:

[0145] Δk (0) =[1,2,..,k,..,K]

[0146] Where, Δk (0) Set the initialization sequence array for the inspection robots, where k is the sequence number of the inspection robots, and K=5 indicates the number of inspection robots;

[0147] The maximum speed array for each inspection robot described in step 2 is defined as follows:

[0148] Δv max =[v 1,max ,v 2,max ...,v k,max ,..,v K,max ]

[0149] Where, Δv max The array of maximum speeds for each inspection robot, v k,max The maximum speed of the inspection robot in the kth setting sequence, where k is the setting sequence number of the inspection robot, and K=5 indicates the number of inspection robots;

[0150] The array of initial monitoring points for each inspection robot described in step 2 is defined as follows:

[0151] ΔL (0) =[L1,L2,..,L k ,..,L K ]

[0152] Where, ΔL (0) An array containing the number of initial monitoring points for each inspection robot, L. k The number of monitoring points that the inspection robot in the kth setting sequence is responsible for, where k is the setting sequence number of the inspection robot, and K=5 indicates the number of inspection robots;

[0153] The data interaction capability array for each inspection robot described in step 2 is defined as follows:

[0154] Δp=[p1,p2,..,p k ,..,p K ]

[0155] Where Δp is the data interaction capability array of the inspection robot, p k The data interaction capability for the kth inspection robot with the set order, where k is the inspection robot's set order number, and K=5 indicates the number of inspection robots;

[0156] The monitoring point array mentioned in step 2 is defined as follows:

[0157] Δm=[1,2,..,m,..,M]

[0158] Where Δm is the array of monitoring points, m is the monitoring point number, and M = 10000 is the number of monitoring points.

[0159] The array of interactive data volume for all types of sensor nodes mentioned in step 2 is defined as follows:

[0160] Δd=[d1,d2,..,d n ,..,d N ]

[0161] Where Δd is the array of interactive data volume of all sensor nodes, n is the sensor node type number, N=50 is the number of sensor node types, and d n Let be the amount of interactive data from the nth sensor.

[0162] The array of the number of each type of sensor node near each monitoring point in step 2 is defined as follows:

[0163]

[0164] Where Δa is an array representing the number of each type of sensor node near each monitoring point, n is the sensor node type number, N=50 is the number of sensor node types, m is the monitoring point number, M=10000 is the number of monitoring points, and a n,m This represents the number of nth type of sensor nodes near the mth monitoring point.

[0165] Step 3: Calculate the total amount of interactive data at each monitoring point, the inspection time of the inspection robot at each monitoring point, and the single inspection cycle of each inspection robot in sequence, using the setting order of the inspection robots and the number of monitoring points that each inspection robot is responsible for as variables.

[0166] The total amount of interactive data for each monitoring point in step 3 is defined as follows:

[0167]

[0168] Among them, Q m Let d be the total amount of interactive data at the m-th monitoring point. n Let a be the amount of interactive data from the nth sensor. n,m Let N = 50, where N is the number of sensor nodes of type n near the m-th monitoring point, n is the sensor node type number, N = 50 is the number of sensor node types, m is the monitoring point number, and M = 10000 is the number of monitoring points.

[0169] The inspection time of the inspection robot at each monitoring point in step 3 is defined as follows:

[0170]

[0171] Among them, t m Let Q be the inspection time of the inspection robot at the m-th monitoring point. m p represents the total amount of interactive data at the m-th monitoring point. k For the data interaction capability of the k-th inspection robot with set order, L k The number of monitoring points handled by the inspection robot in the k-th order, where Δl is the distance between monitoring points, and v k,max This represents the maximum speed of the inspection robot in the k-th setting sequence, where k is the robot setting sequence number, K=5 represents the number of inspection robots, m is the monitoring point number, and M=10000 is the number of monitoring points. For from L1 to L k Summation operation, This is for calculating the maximum value.

[0172] The single inspection cycle of each inspection robot mentioned in step 3 is defined as follows:

[0173]

[0174] Among them, T k Let t be the single inspection cycle of the k-th inspection robot. m+a Let L be the inspection time of the inspection robot at the (m+a)th monitoring point. k-1 This represents the number of monitoring points handled by the (k-1)th inspection robot in the setup sequence, where m is the monitoring point number, k is the inspection robot setup sequence number, and K = 5 indicates the number of inspection robots. For array [L1,L2,..,L] k Summation operation;

[0175] Step 4: Sequentially construct the constraints on the number of monitoring points, the inspection cycle of the utility tunnel inspection system, and the optimization objective of the utility tunnel inspection system to establish the optimization problem of the utility tunnel inspection system;

[0176] The constraint on the number of monitoring points mentioned in step 4 is defined as follows:

[0177]

[0178] Among them, L k This represents the number of monitoring points that the inspection robot in the k-th setup sequence is responsible for, where k is the robot setup sequence number, K=5 represents the number of inspection robots, and M=10000 represents the number of monitoring points. For from L1 to L k Summation operation;

[0179] The inspection cycle constraint of the utility tunnel inspection system described in step 4 is defined as follows:

[0180]

[0181] Among them, T k Let k be the single inspection cycle of the inspection robot. The single inspection cycle of the utility tunnel inspection system is defined by k, which is the sequence number of the inspection robot. K = 5 indicates the number of inspection robots.

[0182] The optimization objective of the utility tunnel inspection system described in step 4 is defined as minimizing the single inspection cycle of the utility tunnel inspection system by optimizing the setting sequence of inspection robots and the number of monitoring points each inspection robot is responsible for:

[0183]

[0184] in, Let Δk be the single inspection cycle of the utility tunnel inspection system, Δk be the sequence array for setting the inspection robots, and ΔL be the array representing the number of monitoring points each inspection robot is responsible for. To maximize b by optimizing a;

[0185] The optimization problem of the utility tunnel inspection system described in step 4 is established as follows:

[0186] By optimizing the order in which the inspection robots are set up and the number of monitoring points each robot is responsible for, the single inspection cycle of the utility tunnel inspection system is minimized to:

[0187]

[0188]

[0189]

[0190] Where k is the inspection robot number, K=5 is the number of inspection robots, M=10000 is the number of monitoring points, and L k The number of monitoring points that the k-th inspection robot is responsible for. For a single inspection cycle of the utility tunnel inspection system, T k Let k be the single inspection cycle of the k-th robot. For from L1 to L k The summation operation is performed, where Δk is the sequence array for the inspection robots, and ΔL is the array representing the number of monitoring points each inspection robot is responsible for. The goal is to maximize the operation of b by optimizing a.

[0191] Step 5: Solve the optimization problem of the pipe gallery inspection system using the successive iterative particle swarm optimization algorithm to obtain the optimal setting order of the inspection robots and the optimal number of monitoring points each inspection robot is responsible for. Each inspection robot moves on the pipe gallery inspection track according to the optimal setting order and the optimal number of monitoring points it is responsible for, and performs periodic inspections of the sensor nodes.

[0192] Figure 5 The diagram shows the flowchart of the iterative particle swarm optimization algorithm according to an embodiment of the present invention. The initialization process includes: initializing the iteration count r = 0, initializing the position and velocity of each particle, calculating the initial fitness of each particle, updating the local optimal fitness and particle position, and updating the global optimal fitness and particle position. The processing process includes: calculating the position of each particle in the (r+1)th iteration, calculating the particle fitness in the (r+1)th iteration, updating the local optimal fitness and particle position, updating the global optimal fitness and particle position, and updating the particle velocity in the (r+1)th iteration. The judgment process includes: determining whether the following conditions are met. Where r is the number of iterations. The maximum number of iterations is given; the final output process includes outputting the optimal particle position, which is then restored to the inspection system design scheme.

[0193] Step 5, which describes solving the optimization problem of the pipe gallery inspection system using the successive iterative particle swarm optimization algorithm, includes the following steps:

[0194] Step 5.1: Initialize the position, velocity, and fitness value of each particle; calculate the initial iterative optimal fitness value and the initial iterative optimal particle position; calculate the global optimal fitness value and the global optimal particle position; and generate the mapping matrix.

[0195] Step 5.2: Calculate the particle position array and fitness value for the next iteration, update the iterative optimal fitness value and iterative optimal particle position, update the global optimal fitness value and global optimal particle position, and update the velocity of each particle for the next iteration;

[0196] Step 5.3: Determine the stopping condition for iteration. If the iteration stops, output the final value; otherwise, return to step 5.2.

[0197] The specific implementation of the iterative particle swarm optimization algorithm described in step 5 is as follows:

[0198] The position of each initialized particle described in step 5.1 is defined as:

[0199] A p (0) =[ΔK (0) ,ΔL (0) ],p=1,2,..,P

[0200] Among them, A p (0) Let p be the initial position of the p-th particle, where p is the particle number, P = 100 is the number of particles, and ΔK is the initial position of the p-th particle. (0) To initialize the inspection robot, set a sequence array, ΔL (0) An array representing the number of initial monitoring points each inspection robot is responsible for, [ΔK] (0) ,ΔL (0) ] is used to form an array;

[0201] The initial particle velocity described in step 5.1 is defined as:

[0202] V p (0) =[Rand p (1) K! , Rand p (K-1),M-∑Rand p [(K-1)], p = 1, 2, ..., P

[0203] in, Let P be the initial velocity of the p-th particle, where p is the particle number, P = 100 is the number of particles, M = 10000 is the number of monitoring points, K = 5 is the number of inspection robots, and Rand... p (K-1) is an array of length K-1 generated in the p-th random generation, ∑Rand p (K-1) represents the array Rand p (K-1) Summation operation, [Rand p (1) K! , Rand p (K-1),M-∑Rand p [(K-1)] represents the combined array operation, and K! represents the order multiplication of K.

[0204] The initial fitness value of each particle in step 5.1 is defined as follows:

[0205]

[0206] in, For located The initial fitness value of the particles. Let p be the initial position of the p-th particle, where p is the particle number, P = 100 is the number of particles, k is the inspection robot number, and K = 5 is the number of inspection robots. To initialize the single inspection cycle of the utility tunnel inspection system, Let Ω = 10 be the initial single inspection cycle for the k-th robot. 6 For constant maximum parameters, For array [L1,L2,..,L] k Summation operation;

[0207] Step 5.1, which initializes the iterative optimal fitness value and the iterative optimal particle position, is defined as follows:

[0208]

[0209] in, To initialize the iterative optimal fitness value, For located The initial fitness value of the particles. Let p be the initial position of the p-th particle. To initialize the optimal fitness for iteration, The initial particle position corresponding to the time, which is also the first The initial position of each particle, where p is the particle number and P = 100 is the number of particles. This is for calculating the maximum value.

[0210] The global optimal fitness value and the global optimal particle position mentioned in step 5.1 are defined as follows:

[0211]

[0212] in, This represents the globally optimal fitness value. The global optimal particle position. To initialize the iterative optimal fitness value, To initialize the optimal fitness for iteration, The initial particle position corresponding to the time, which is also the first The initial position of each particle, where p is the particle number;

[0213] The mapping matrix generated in step 5.1 is defined as follows:

[0214]

[0215] Where Q is a mapping matrix of size K! rows and K columns, i is the row count of mapping matrix Q, k is the column count of mapping matrix Q, K=5 is the number of inspection robots, Q(i,k) is the element in the i-th row and k-th column of mapping matrix Q, Q(i,q) is the element in the i-th row and q-th column of mapping matrix Q, and (k-1)! is the multiplication calculation.

[0216] The first element of the particle position array for the next iteration, as described in step 5.2, is defined as follows:

[0217]

[0218] Where, ΔA p (r+1) Let ΔA be the position of the p-th particle in the (r+1)-th iteration. p (r) Let ΔA be the position of the p-th particle in the r-th iteration, where r is the iteration number, p is the particle number, P = 100 is the number of particles, and ΔA is the position of the p-th particle in the r-th iteration. p (r+1) (1) represents the first element of the position of the p-th particle in the (r+1)-th iteration, Q(j,:) represents the j-th row of the array that makes up the mapping matrix Q, and Δt is a constant time parameter. Let p be the velocity of the p-th particle in the r-th iteration. This indicates the rounding up operation;

[0219] The 2nd to K=5th elements of the particle position array for the next iteration, as described in step 5.2, are defined as follows:

[0220]

[0221] Where, ΔA p (r+1) Let ΔA be the position of the p-th particle in the (r+1)-th iteration, where r is the iteration number, p is the particle number, P = 100 is the number of particles, and ΔA is the position of the p-th particle in the (r+1)-th iteration. p (r+1) (2:K) represents the array consisting of the 2nd to the K=5th elements of the p-th particle position in the (r+1)-th iteration, ΔA p (r) (2:K) represents an array consisting of the 2nd to the K=5th elements at the position of the p-th particle in the r-th iteration, where Δt is a constant time parameter. Let p be the velocity of the p-th particle in the r-th iteration. This indicates the rounding up operation.

[0222] The K+1=6th element of the particle position array for the next iteration, as described in step 5.2, is defined as follows:

[0223] ΔA p (r+1)(K+1)=M-∑ΔA p (r+1) (2:K), p=1,2,..,P

[0224] Where, ΔA p (r+1) Let P be the position of the p-th particle in the (r+1)-th iteration, where r is the iteration number, p is the particle number, P = 100 is the number of particles, M = 10000 is the number of monitoring points, and ΔA is the position of the p-th particle in the (r+1)-th iteration. p (r+1) (K+1) represents the (K+1)th element at the position of the p-th particle in the (r+1)-th iteration, ΔA p (r+1) (2:K) represents the array consisting of the 2nd to the K=5th elements of the p-th particle position in the (r+1)-th iteration, ∑ΔA p (r+1) (2:K) represents the array ΔA p (r+1) The summation operation of all elements in (2:K).

[0225] The particle position array for the next iteration, as described in step 5.2, is defined as follows:

[0226] ΔA p (r+1) =[ΔA p (r+1) (1),ΔA p (r+1) (2:K),ΔA p (r+1) [(K+1)], p = 1, 2, ..., P

[0227] Where, ΔA p (r+1) Let P be the position of the p-th particle in the (r+1)-th iteration, where P = 100 is the number of particles, and ΔA p (r+1) (1) represents the first element of the position of the p-th particle in the (r+1)-th iteration, ΔA p (r+1) (2:K) represents the array consisting of the 2nd to the K=5th elements of the p-th particle position in the (r+1)-th iteration, ΔA p (r+1) (K+1) represents the K+1=6th element at the position of the p-th particle in the (r+1)-th iteration, [ΔA p (r+1) (1),ΔA p (r+1 (2:K),ΔA p (r+1) [(K+1)] represents the combined array operation.

[0228] Step 5.2 defines the fitness value of each particle for the next iteration as follows:

[0229]

[0230] in, For located The fitness value of the particles, Let p be the position of the p-th particle in the r-th iteration, where p is the particle number, P = 100 is the number of particles, r is the iteration number, k is the inspection robot number, and K = 5 is the number of inspection robots. Let r be the single inspection cycle of the utility tunnel inspection system in the r-th iteration. Let Ω = 10 be the single inspection cycle of the k-th robot in the r-th iteration. 6 For constant maximum parameters, For from L1 to L k The summation operation.

[0231] The optimal fitness value and optimal particle position in step 5.2 are defined as follows:

[0232]

[0233] in, Let the optimal fitness value be the value obtained in the r-th iteration. For located The particle fitness value, Let be the position of the p-th particle in the r-th iteration. For located The particle fitness value, The optimal fitness for the r-th iteration is The particle position corresponding to time r, i.e., the position in the r-th iteration. The initial position of each particle, where p is the particle number, P is the number of particles, and r is the iteration number. This is for calculating the maximum value.

[0234] Step 5.2, which updates the global optimal fitness value and the global optimal particle position, is defined as follows:

[0235]

[0236] in, This represents the globally optimal fitness value. Let p be the optimal fitness value in the r-th iteration, where r is the iteration number and p is the particle number. For located The particle fitness value, The global optimal fitness is The particle position corresponding to time, i.e., the first The particle in the first The position of the next iteration. This is for calculating the maximum value.

[0237] Step 5.2 defines the update of the velocity of each particle in the next iteration as follows:

[0238]

[0239] in, Let p be the velocity of the p-th particle in the r-th iteration. Let be the velocity of the p-th particle in the (r+1)-th iteration, where r is the iteration number, p is the particle number, P = 100 is the number of particles, ω = 0.8 is the constant inertia coefficient, C1 = 0.5 is the constant individual learning factor, and C2 = 0.5 is the constant sociological coefficient. The first random number is within the interval [0,1]. The second interval is a random number within [0,1]. Let the optimal fitness value be the value obtained in the r-th iteration. This represents the globally optimal fitness value. For located The particle fitness value.

[0240] The convergence condition described in step 5.3 is defined as follows: [The condition is defined as follows] Where r is the number of iterations. The maximum number of iterations is given; if so, output the final value. The global optimal fitness is The corresponding particle position at time, which is also the first The particle in the first At the position of the next iteration, according to step 5.2, restore Δk to the sequence array for setting the inspection robot and ΔL to the number of monitoring points for each inspection robot, and the iteration ends; otherwise, update r = r + 1 and return to step 5.2.

[0241] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0242] Although this invention uses terms such as inspection robot and wireless sensor node frequently, the possibility of using other terms is not excluded. These terms are used merely for the convenience of describing and explaining the essence of this invention; interpreting them as any additional limitation would contradict the spirit of this invention.

[0243] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for planning periodic pipe gallery inspections using multiple inspection robots, characterized in that: Includes the following steps: Step 1: Lay an inspection track along the inside of the utility tunnel, and place the multiple inspection robots on the inspection track in sequence to move; fix multiple sensor nodes evenly at different positions inside the utility tunnel; set the start and end points of the inspection track, and evenly divide the area between the start and end points of the inspection track into multiple monitoring points, and have each inspection robot wirelessly interact with multiple sensor nodes at the monitoring points in sequence. Using the midline between two adjacent monitoring points as the boundary between them, the number and type of sensor nodes for each monitoring point are set, the distance between two adjacent monitoring points is obtained, the amount of interactive data for each type of sensor node is obtained, the number of inspection robots is obtained, and the maximum speed and data interaction speed of each inspection robot are obtained. Step 2: Construct an array of inspection robot numbers, construct the initialization setting order of the inspection robots based on the inspection robot numbers, construct an array of maximum speed for each inspection robot, an array of the number of initial monitoring points each inspection robot is responsible for, an array of data interaction capabilities for each inspection robot based on the initialization setting order of the inspection robots, construct an array of monitoring points, construct an array of the interaction data volume of all types of sensor nodes based on the interaction data volume of each type of sensor node, and construct an array of the number of each type of sensor node at each monitoring point; Step 3: Calculate the total amount of interactive data at each monitoring point, the inspection time of the inspection robot at each monitoring point, and the single inspection cycle of each inspection robot in sequence, using the setting order of the inspection robots and the number of monitoring points that each inspection robot is responsible for as variables. Step 4: Sequentially construct the constraints on the number of monitoring points, the inspection cycle of the utility tunnel inspection system, and the optimization objective of the utility tunnel inspection system to establish the optimization problem of the utility tunnel inspection system; Step 5: Solve the optimization problem of the pipe gallery inspection system using the successive iteration particle swarm optimization algorithm to obtain the optimal setting order of the inspection robots and the optimal number of monitoring points each inspection robot is responsible for. Each inspection robot moves on the pipe gallery inspection track according to the optimal setting order and the optimal number of monitoring points it is responsible for, and performs periodic inspections for the sensor nodes. Step 5, which describes solving the optimization problem of the pipe gallery inspection system using the successive iterative particle swarm optimization algorithm, includes the following steps: Step 5.1: Initialize the position, velocity, and fitness value of each particle; calculate the initial iterative optimal fitness value and the initial iterative optimal particle position; calculate the global optimal fitness value and the global optimal particle position; and generate the mapping matrix. Step 5.2: Calculate the particle position array and fitness value for the next iteration, update the iterative optimal fitness value and iterative optimal particle position, update the global optimal fitness value and global optimal particle position, and update the velocity of each particle for the next iteration; Step 5.3: Determine the stopping condition for iteration. If the iteration stops, output the final value; otherwise, return to step 5.

2.

2. The multi-inspection robot joint periodic pipe gallery inspection planning method according to claim 1, characterized in that: The inspection robot number array mentioned in step 2 is defined as follows: in, For the inspection robot's number array, Number the inspection robot. Indicates the number of inspection robots; The initialization setup sequence for the inspection robot described in step 2 is defined as follows: in, Set the sequence array for the initialization of the inspection robot. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots; The maximum speed array for each inspection robot described in step 2 is defined as follows: in, An array of maximum speeds for each inspection robot. For the first The maximum speed of the inspection robots in the set sequence. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots; The array of initial monitoring points for each inspection robot described in step 2 is defined as follows: in, An array representing the number of initial monitoring points each inspection robot is responsible for. For the first The number of monitoring points that the inspection robots are responsible for in the set sequence. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots; The data interaction capability array for each inspection robot described in step 2 is defined as follows: in, This is an array of data interaction capabilities for the inspection robot. For the first The data interaction capabilities of inspection robots with a set order. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots; The monitoring point array mentioned in step 2 is defined as follows: in, For the monitoring point array, Number the monitoring points. Number of monitoring points; The array of interactive data volume for all types of sensor nodes mentioned in step 2 is defined as follows: in, An array representing the amount of interaction data from all sensor nodes. Assign a type number to the sensor node. For the number of sensor node types, For the first The amount of interactive data from various sensors; The array of the number of each type of sensor node near each monitoring point in step 2 is defined as follows: in, An array of the number of each type of sensor node near each monitoring point. Assign a type number to the sensor node. For the number of sensor node types, Number the monitoring points. For the number of monitoring points, For the first Near the first monitoring point The number of sensor nodes.

3. The multi-inspection robot joint periodic pipe gallery inspection planning method according to claim 2, characterized in that: The total amount of interactive data for each monitoring point in step 3 is defined as follows: in, For the first The total amount of interactive data at each monitoring point For the first The amount of interactive data from these sensors For the first Near the first monitoring point The number of sensor nodes, Assign a type number to the sensor node. For the number of sensor node types, Number the monitoring points. Number of monitoring points; The inspection time of the inspection robot at each monitoring point in step 3 is defined as follows: in, For the inspection robot in the first Inspection time for each monitoring point For the first The total amount of interactive data at each monitoring point For the first The data interaction capabilities of inspection robots with a set order. For the first The number of monitoring points that the inspection robots are responsible for in the set sequence. To monitor the spacing between points, For the first The maximum speed of the inspection robots in the set sequence. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots. Number the monitoring points. For the number of monitoring points, From arrive The summation operation is used, and max is used to find the maximum value. The single inspection cycle of each inspection robot mentioned in step 3 is defined as follows: in, For the first The single inspection cycle of an inspection robot For the inspection robot in the first Inspection time for each monitoring point For the first The number of monitoring points that the inspection robots are responsible for in the set sequence. Number the monitoring points. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots. For array The summation operation.

4. The multi-inspection robot combined periodic pipe gallery inspection planning method according to claim 3, characterized in that: The constraint on the number of monitoring points mentioned in step 4 is defined as follows: in, For the first The number of monitoring points that the inspection robots are responsible for in the set sequence. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots. For the number of monitoring points, From arrive Summation operation; The inspection cycle constraint of the utility tunnel inspection system described in step 4 is defined as follows: in, For the first The single inspection cycle of an inspection robot This refers to the single inspection cycle of the utility tunnel inspection system. Assign sequential numbers to the inspection robots. Indicates the number of inspection robots; The optimization objective of the utility tunnel inspection system described in step 4 is defined as minimizing the single inspection cycle of the utility tunnel inspection system by optimizing the setting sequence of inspection robots and the number of monitoring points each inspection robot is responsible for: in, This refers to the single inspection cycle of the utility tunnel inspection system. Set a sequential array for the inspection robot. An array representing the number of monitoring points each inspection robot is responsible for. To optimize maximize Operations; The optimization problem of the utility tunnel inspection system described in step 4 is established as follows: By optimizing the order in which the inspection robots are set up and the number of monitoring points each robot is responsible for, the single inspection cycle of the utility tunnel inspection system is minimized to: in, Number the inspection robot. The number of inspection robots, For the number of monitoring points, For the first The number of monitoring points handled by each inspection robot This refers to the single inspection cycle of the utility tunnel inspection system. For the first The single inspection cycle of a robot, From arrive Summation operation, Set a sequential array for the inspection robot. An array representing the number of monitoring points each inspection robot is responsible for. To optimize maximize Calculation.

5. The multi-inspection robot joint periodic pipe gallery inspection planning method according to claim 4, characterized in that: The position of each initialized particle described in step 5.1 is defined as: in, For the first Initial position of each particle Number the particles. For the number of particles, Set a sequence array to initialize the inspection robot. An array representing the number of initial monitoring points each inspection robot is responsible for, with [] representing the constituent arrays; The initial particle velocity described in step 5.1 is defined as: in, For the first The initial velocity of each particle. Number the particles. For the number of particles, For the number of monitoring points, The number of inspection robots, For the first The length of each random generation is... An array of -1 For array Summation operation, For combined array operations, For about The calculation of the order of multiplication; The initial fitness value of each particle in step 5.1 is defined as follows: in, For located The initial fitness value of the particles. For the first Initial position of each particle Number the particles. For the number of particles, Number the inspection robot. The number of inspection robots, To initialize the single inspection cycle of the utility tunnel inspection system, For the first The initial single inspection cycle of a robot. For constant maximum parameters, For array Summation operation; Step 5.1, which initializes the iterative optimal fitness value and the iterative optimal particle position, is defined as follows: in, To initialize the iterative optimal fitness value, For located The initial fitness value of the particles. For the first Initial position of each particle To initialize the optimal fitness for iteration, The initial particle position corresponding to the time, which is also the first Initial position of each particle Number the particles. For the number of particles, This is for calculating the maximum value. The global optimal fitness value and the global optimal particle position mentioned in step 5.1 are defined as follows: in, This represents the globally optimal fitness value. The global optimal particle position. To initialize the iterative optimal fitness value, To initialize the optimal fitness for iteration, The initial particle position corresponding to the time, which is also the first Initial position of each particle Number the particles; The mapping matrix generated in step 5.1 is defined as follows: in, For size OK The mapping matrix of columns, For mapping matrix The row count, For mapping matrix Column count, The number of inspection robots, For mapping matrix The Middle Line number Column elements, For mapping matrix The Middle Line number Column elements, This is calculated using a multiplier.

6. The multi-inspection robot combined periodic pipe gallery inspection planning method according to claim 5, characterized in that: The first element of the particle position array for the next iteration, as described in step 5.2, is defined as follows: in, For the first The iteration of the ... Particle positions, For the first The iteration of the ... Particle positions, For the number of iterations, Number the particles. For the number of particles, Indicates the first The iteration of the ... The first element of the particle position, Represents the composition of the mapping matrix The row array, For constant time parameters, For the first In the nth iteration The speed of each particle This indicates the rounding up operation; Step 5.2 describes the second to third elements of the particle position array for the next iteration. Each element is defined as: in, For the first The iteration of the ... Particle positions, For the number of iterations, Number the particles. For the number of particles, Indicates the first The iteration of the ... The second to the third particle position An array consisting of 10 elements Indicates the first The iteration of the ... The second to the third particle position An array consisting of 10 elements For constant time parameters, For the first In the nth iteration The speed of each particle This indicates the rounding up operation; Step 5.2 describes the next iteration of the particle position array. Each element is defined as: in, For the first The iteration of the ... Particle positions, For the number of iterations, Number the particles. For the number of particles, For the number of monitoring points, Indicates the first The iteration of the ... The position of the first particle One element, Indicates the first The iteration of the ... The second to the third particle position An array consisting of 10 elements Represents array Summation of all elements; The particle position array for the next iteration, as described in step 5.2, is defined as follows: in, For the first The iteration of the ... Particle positions, Indicates the first The iteration of the ... The first element of the particle position, Indicates the first The iteration of the ... The second to the third particle position An array consisting of 10 elements Indicates the first The iteration of the ... The position of the first particle One element, For combined array operations; Step 5.2 defines the fitness value of each particle for the next iteration as follows: in, For located The fitness value of the particles, For the first In the nth iteration The position of each particle. Number the particles. For the number of particles, For the number of iterations, Number the inspection robot. The number of inspection robots, For the first The single inspection cycle of the utility tunnel inspection system in the next iteration For the first In the nth iteration The single inspection cycle of a robot, For constant maximum parameters, From arrive Summation operation; The optimal fitness value and optimal particle position in step 5.2 are defined as follows: in, For the first The optimal fitness value in the next iteration. For located The particle fitness value, For the first The first particle The position of the next iteration. For located The particle fitness value, For the first The optimal fitness in the next iteration is The particle position corresponding to time, i.e., the first In the nth iteration Initial position of each particle Number the particles. For the number of particles, For the number of iterations, This is for calculating the maximum value. Step 5.2, which updates the global optimal fitness value and the global optimal particle position, is defined as follows: in, This represents the globally optimal fitness value. For the first The optimal fitness value in the next iteration. For the number of iterations, Number the particles. For located The particle fitness value, The global optimal fitness is The particle position corresponding to time, i.e., the first The particle in the first The position of the next iteration. This is for calculating the maximum value. Step 5.2 defines the update of the velocity of each particle in the next iteration as follows: in, For the first In the nth iteration The speed of each particle For the first In the nth iteration The speed of each particle For the number of iterations, Number the particles. For the number of particles, For a constant coefficient of inertia, For constant individual learning factors, As a constant sociology department factor, For the first interval Random numbers within, For the second interval Random numbers within, For the first The optimal fitness value in the next iteration. This represents the globally optimal fitness value. For located The particle fitness value.

7. The multi-inspection robot combined periodic pipe gallery inspection planning method according to claim 6, characterized in that: The convergence condition described in step 5.3 is defined as follows: [The condition is defined as follows] ,in For the number of iterations, The maximum number of iterations is given; if so, output the final value. The global optimal fitness is The corresponding particle position at time, which is also the first The particle in the first The position of the next iteration is restored according to step 5.

2. Set the sequence array and The iteration ends when the array containing the number of monitoring points each inspection robot is responsible for is generated; otherwise, the iteration ends. Return to step 5.2.

Images