A Cooperative Path Planning Method and System for 5G Power Inspection UAVs

Abstract

The present invention proposes a collaborative path planning method for 5G power inspection drones, and also proposes a corresponding system, storage medium and electronic device. In view of the problem of the decline in communication signal quality encountered in multi-drone collaborative power inspection, the present invention studies a collaborative path planning method for 5G power inspection drones. First, based on the 5G signal transmission characteristics, a propagation loss model in the large-scale space of power inspection is established; then, based on the genetic algorithm framework, a multi-drone path planning method considering the joint constraints of 5G signal quality, flight mileage and inspection targets is proposed; finally, a simulation of the multi-drone collaborative power inspection task based on 5G signals is carried out. The results show that the path planned based on the proposed algorithm can effectively guarantee the 5G signal quality and detection effect during the drone inspection process.

Description

Technical Field

[0001] This application belongs to the field of intelligent operation and maintenance of power grid unmanned aerial vehicles (UAVs), and specifically relates to a collaborative path planning method and system for 5G power inspection UAVs. Background Art

[0002] As the pillars of the power transmission network, transmission towers are exposed to the outdoor environment for a long time and need to be regularly maintained. Traditional power inspection uses manual inspection methods, which have disadvantages such as high work intensity, low efficiency, and poor results, and also require a large number of maintenance personnel, and the problems faced are becoming increasingly prominent. In recent years, due to the rapid development of UAV technology, UAV inspection has gradually replaced manual inspection due to its advantages such as strong flexibility, low operation difficulty, and high inspection efficiency, and has played an important role in power inspection tasks.

[0003] As the complexity of inspection tasks gradually increases, when dealing with large-scale power system inspection tasks, due to the large number of power towers and the long distance between towers, a single UAV cannot complete the inspection tasks of the entire system well. Therefore, the method of using multiple UAVs for inspection simultaneously can effectively improve the inspection efficiency. Since UAVs need to transmit image data and analyze fault points in real time during the inspection process, there are high requirements for the quality of communication signals during the inspection process. The development of 5G technology provides an important communication guarantee for power inspection UAVs.

[0004] In the process of realizing the autonomous inspection task of unmanned aerial vehicles (UAVs), multi-UAV path planning is a very important research content, and many scholars have proposed solutions. Luo Longfu et al. proposed an improved algorithm based on the RRT (Rapidly Exploring Random Trees) algorithm. By inheriting the randomness of the RRT search in the XY plane and constraining the search process in the Z-axis height, the path gradually approaches the target point in the end (see cited document 1). Li Xiaohui et al. added VND (Variable Neighborhood Descent) as a subordinate strategy within the framework of the ALNS (Adaptive Large Neighborhood Search) algorithm, and proposed a new hybrid meta-heuristic algorithm (see cited document 2). By adding an adaptive mechanism to the large neighborhood search algorithm, according to the effects of the destruction operator and the repair operator, the weights are dynamically adjusted and different search operators are selected to find a better solution. Miao Dongdong et al. proposed a path planning method based on artificial potential field, which realizes obstacle avoidance and path planning of UAVs by setting a gravitational field at the projection point of the ideal path of the UAV (see cited document 3). Ma Rui et al. combined the theory of multi-agent reinforcement learning and proposed a multi-UAV large-scale power inspection task allocation algorithm based on QMIX. Each UAV selects actions according to its own environmental observations, realizing multi-UAV collaborative rapid inspection (see cited document 4).

[0005] It can be seen that most of the current research focuses on the UAV path planning method, and the communication signal quality of the planned path has not been considered. In view of this, in order to ensure the communication quality during the UAV power inspection process, the present invention introduces the 5G signal quality in the multi-UAV path planning process, and uses it together with the total flight distance and the tower position as constraint conditions to plan the UAV path. This method has good reference significance for the implementation of multi-UAV power inspection.

[0006] List of cited documents:

[0007] 1. Luo Longfu, Li Dong, Zhong Hang. UAV Power Tower Inspection Path Planning Based on Improved RRT [J]. Journal of Hunan University (Natural Sciences Edition), 2018, 45(10): 80-86.

[0008] 2. Li Xiaohui, Zhang Lu, Liu Chuanshui, Zhao Yi, Dong Yuan. UAV Swarm Path Planning Algorithm in Power Inspection [J]. Computer Systems & Applications, 2022, 31(03): 241-247.

[0009] 3. Miao Dongdong, Lü Pin, Wang Qing, Xu Haiming. Improved Artificial Potential Field Method for UAV Trajectory Planning in Power Inspection [J]. Computer & Digital Engineering, 2021, 49(11): 2260-2265.

[0010] 4. Ma Rui, Ouyang Quan, Wu Zhaoxiang, Cong Yuhua, Wang Zhisheng. Multi-UAV Power Inspection Task Planning Based on Deep Reinforcement Learning[J]. Computer and Modernization, 2022(01): 98-102. Summary of the Invention

[0011] In view of the deficiencies in the prior art, the present invention provides a collaborative path planning method for 5G power inspection UAVs, and proposes a corresponding system, storage medium and electronic device.

[0012] To achieve the above object, the present invention adopts the following technical solutions:

[0013] A collaborative path planning method for 5G power inspection UAVs, characterized by comprising:

[0014] Establish a 5G signal quality distribution model in the power inspection scenario, and calculate the path loss value at each tower based on the 5G signal quality distribution model;

[0015] According to the path loss value at each tower, combined with the maximum path loss value of the signal, calculate the signal quality at the current tower;

[0016] Add 5G signal constraints to the UAV path planning algorithm, and the UAV path planning algorithm adopts a genetic algorithm with signal quality and the total flight distance of the UAV as the fitness function;

[0017] Use the UAV path planning algorithm to perform optimal solution to obtain the optimal path for power inspection.

[0018] To optimize the above technical solutions, the specific measures taken also include:

[0019] Further, the 5G signal quality distribution model is specifically as follows:

[0020] Consider the distribution of power towers and characterize the coverage performance of the 5G signal base station channel propagation model:

[0021] The line-of-sight probability calculation formula is as follows:

[0022]

[0023] Wherein,

[0024]

[0025] In the formula, Pr LOS represents the line-of-sight probability, d 2D-out represents the planar distance between the tower and the user end outdoors, and h UT represents the height of the user;

[0026] The loss calculation formula during line-of-sight propagation is as follows:

[0027]

[0028] Among them,

[0029] PL1 = 28.0 + 22log 10 (d 3D ) + 20log 10 (f c )

[0030] PL2 = 28.0 + 40log 10 (d 3D ) + 20log 10 (f c ) - 9log 10 ((d' BP ) 2 + (h BS - h UT ) 2 )

[0031]

[0032] In the formula, PL UMa-LOS represents the large-scale spatial loss value during line-of-sight propagation, h BS is the height of the pole tower, h UT is the flight height of the UAV, d 2D is the planar distance between the pole tower and the UAV, d 3D is the three-dimensional distance between the pole tower and the UAV, d' BP is the breakpoint distance, f c is the signal frequency;

[0033] The loss calculation formula during non-line-of-sight propagation is as follows:

[0034] PL UMa-NLOS = max(PL UMa-LOS , PL' UMa-NLOS )

[0035] PL' UMa-NLOS = 13.54 + 39.08log 10 (d 3D ) + 20log 10 (f c ) - 0.6(h UT - 1.5)

[0036] In the formula, PL UMa-NLOS represents the large-scale spatial loss value during non-line-of-sight propagation;

[0037] The maximum allowable path loss of the signal is calculated according to the link budget, and its calculation formula is as follows:

[0038] PL max = P TX - L f + G TX - M f - M I + G RX - SR

[0039] In the formula, PL max represents the maximum allowable path loss value of the signal, P TX is the base station transmission power, L f is the feeder loss, G TX is the base station antenna gain, M f is the shadow fading and fading margin, M I is the interference margin, G RX is the receiving end antenna gain, and SR is the receiving end sensitivity.

[0040] Furthermore, the path loss value at each pole tower includes the large-scale spatial loss value PL UMa-LOS during line-of-sight propagation and the large-scale spatial loss value PL UMa-NLOS .

[0041] Furthermore, the signal quality at the current pole tower is as follows:

[0042] pathloss = [(Pr LOS ·PL UMa-LOS ) + (Pr NLOS ·PL UMa-NLOS )] / PL max

[0043] In the formula, pathloss represents the signal quality, Pr LOS represents the line-of-sight probability, PL UMa-LOS represents the large-scale spatial loss value during line-of-sight propagation, Pr NLOS represents the non-line-of-sight probability, PL UMa-NLOS represents the large-scale spatial loss value during non-line-of-sight propagation, and PL max represents the maximum allowable path loss value of the signal.

[0044] Furthermore, the fitness function of the UAV path planning algorithm is as follows:

[0045]

[0046] In the formula, S total represents the total fitness function, nsalesmen represents the number of UAVs, n represents the total number of pole towers to be inspected, Sijk represents the distance that the \(i\)-th UAV flies from the \(j\)-th tower to the \(k\)-th tower, pathloss ijk represents the signal quality that the \(i\)-th UAV flies from the \(j\)-th tower to the \(k\)-th tower.

[0047] Furthermore, the optimal solution is obtained by using the UAV path planning algorithm to obtain the optimal path for power inspection, which specifically includes the following steps:

[0048] 1) Given \(n\) target points, number the target points, randomly shuffle the numbers from 1 to \(n\) \(m - 1\) times, and store the \(m\) groups of numbered data together with the unshuffled order in a matrix; there are \(n_{Salesmen}\) UAVs performing inspection tasks, and use \(n_{Salesmen}-1\) breakpoints to divide each group of numbered data in the matrix into \(n_{Salesmen}\) segments, and randomly generate \(m\) groups of breakpoints and store them in the matrix; each group of numbered data is segmented at the breakpoint, each segment is assigned to a UAV, and the inspection is carried out in order, and a total of \(m\) paths are obtained as the initial population;

[0049] 2) Calculate the signal attenuation value at a certain distance from the base station, and divide it by the maximum allowable path loss value of the signal to obtain the signal attenuation degree as the weight of the fitness function; at the same time, calculate the distance traveled by each UAV in this generation to judge whether the UAV meets the endurance time constraint;

[0050] 3) Under the condition of meeting the UAV endurance requirements, select and store the information of the minimum product of the total distance and signal quality in each generation of the population, including the total flight distance of all UAVs on this path, the order of the tower numbers inspected by each corresponding UAV, and the breakpoint information;

[0051] 4) Simulate the crossover and mutation operations in genetics by randomly swapping the order of the target points of the parent generation and randomly changing the breakpoint information, so as to generate \(m\) paths of the next generation population;

[0052] 5) Loop through steps 2) to 4) until the iteration times are met, and output the shortest path as the optimal path.

[0053] Furthermore, in step 2), the endurance time constraint is calculated by the following formula:

[0054] endurance=(2·height + dUAV max -(num + 1)·(acce x + slowdown x )) / Vsalesmen+(num + 1)·2·slowdown t +π / Wsalesmen·(num - 1)

[0055] Where endurance is the endurance time of the UAV, height is the flight altitude, dUAV max is the longest flight distance of a single UAV, num represents the number of inspection towers, acce x represents the acceleration distance, slowdown x represents the deceleration distance, Vsalesmen represents the flight speed, slowdown t represents the deceleration time, and Wsalesmen represents the angular velocity during inspection.

[0056] The present invention also proposes a cooperative path planning system for 5G power inspection UAVs, including:

[0057] A path loss calculation module, which is used to establish a 5G signal quality distribution model in the power inspection scenario, and calculate the path loss value at each tower based on the 5G signal quality distribution model;

[0058] A signal quality calculation module, which is used to calculate the signal quality at the current tower according to the path loss value at each tower and in combination with the maximum path loss value of the signal;

[0059] A path planning module, which is used to add 5G signal constraints to the UAV path planning algorithm, and the UAV path planning algorithm uses a genetic algorithm with signal quality and the total flight distance of the UAV as the fitness function;

[0060] An output module, which is used to perform optimal solution using the UAV path planning algorithm to obtain the optimal path for power inspection.

[0061] The present invention also proposes a computer-readable storage medium storing a computer program, characterized in that the computer program causes a computer to execute the cooperative path planning method for 5G power inspection UAVs as described above.

[0062] The present invention also proposes an electronic device, characterized in that it includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, it implements the cooperative path planning method for 5G power inspection UAVs as described above.

[0063] The beneficial effects of the present invention are as follows: In view of the scenario of multi-UAV inspection of power towers, based on the genetic algorithm, 5G signal constraints are introduced in the multi-UAV trajectory planning process, and together with the total flight distance and tower positions, they are used as constraint conditions to plan the UAV paths. The simulation results show that there are obvious differences in the lengths of the paths with poor signal quality before and after the constraints. In the paths after adding the constraints, the UAVs will preferentially inspect the towers with stronger signals when the distances are relatively close, optimizing the overall inspection effect and further improving the inspection quality. Brief Description of the Drawings

[0064] Figure 1 is a flowchart of the UAV path planning algorithm under signal constraints adopted by the present invention.

[0065] Figure 2 is a schematic diagram of the path planning result without adding 5G signal constraints.

[0066] Figure 3 is a schematic diagram of the path planning result after adding constraints. Detailed Implementation Manner

[0067] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0068] In one embodiment, the present invention proposes a collaborative path planning method for 5G power inspection UAVs, which specifically includes the following contents:

[0069] 1. 5G Signal Quality Analysis and Modeling

[0070] Establish a 5G signal quality distribution model in the power inspection scenario. Based on the 5G signal quality distribution model, calculate the path loss value at each tower.

[0071] The present invention focuses on the research of the coverage performance of the 5G signal base station channel propagation model, including two parts: line-of-sight and non-line-of-sight (LOS, NLOS) probability propagation loss and large-scale spatial loss. Since the coverage range of power towers is all outdoors, the penetration loss effect of buildings is not considered.

[0072] If there is no obstruction during the propagation of 5G signals, it can be understood as straight-line propagation, that is, line-of-sight propagation. In the process of calculating the total path loss of the signal, both line-of-sight propagation and non-line-of-sight propagation should be comprehensively considered.

[0073] 1) Line-of-sight and non-line-of-sight (LOS, NLOS) probability propagation loss

[0074] The line-of-sight probability calculation formula is as follows:

[0075]

[0076] Among them,

[0077]

[0078] In the formula, PrLOS represents the line-of-sight probability, d 2D-out represents the horizontal distance between the pole tower and the user terminal outdoors, h UT represents the height of the user;

[0079] It can be seen from Equations (1) and (2) that when the horizontal distance between the pole tower and the user terminal is less than 18 m, it can be considered that the signal only propagates in a straight line and the line-of-sight probability is 1; when the horizontal distance is greater than 18 m, the signal may either propagate in a straight line or be blocked and then propagate non-linearly. Its line-of-sight probability is a function of its horizontal distance and the user's height, and as the distance increases, its line-of-sight probability gradually decreases.

[0080] 2) Large-scale spatial loss

[0081] The calculation formula for the large-scale spatial loss in the UMa scenario is as follows:

[0082] Calculation formula for the loss during line-of-sight propagation:

[0083]

[0084] where

[0085] PL1 = 28.0 + 22log 10 (d 3D ) + 20log 10 (f c )

[0086] PL2 = 28.0 + 40log 10 (d 3D ) + 20log 10 (f c ) - 9log 10 (((d' BP )) 2 +(h BS - h UT )) 2 ))

[0087]

[0088] In the formula, PL UMa-LOS represents the value of the large-scale spatial loss during line-of-sight propagation, h BS is the height of the pole tower, h UT is the flight height of the UAV, d 2D is the horizontal distance between the pole tower and the UAV, d 3D is the three-dimensional distance between the pole tower and the UAV, d' BP is the break point distance, f c is the signal frequency;

[0089] Loss calculation formula during non-line-of-sight propagation:

[0090] PL UMa-NLOS = max(PL UMa-LOS , PL' UMa-NLOS ) (4)

[0091] PL' UMa-NLOS = 13.54 + 39.08 log 10 (d 3D ) + 20 log 10 (f c ) - 0.6(h UT ) (5)

[0092] In the formula, PL UMa-NLOS represents the large-scale spatial loss value during non-line-of-sight propagation;

[0093] The maximum allowable path loss of the signal is measured according to the link budget, and its calculation formula is as follows:

[0094] PL max = P TX - L f + G TX - M f - M I + G RX - SR (6)

[0095] In the formula, PL max represents the maximum allowable path loss value of the signal, P TX is the base station transmission power, L f is the feeder loss, G TX is the base station antenna gain, M f is the shadow fading and fading margin, M I is the interference margin, G RX is the receiving end antenna gain, and SR is the receiving end sensitivity.

[0096] 2. Multi-UAV power inspection path planning under signal constraints

[0097] The present invention incorporates 5G signal constraints into the multi-UAV path planning algorithm, as Figure 1 shown. Since the multi-UAV trajectory planning problem can also be understood as a variant of the multi-traveling salesman problem, the present invention conducts research based on the genetic algorithm.

[0098] The Genetic Algorithm (GA for short) is a bionic algorithm based on Darwin's theory of evolution. It sets three basic operators: the crossover operator, the mutation operator, and the selection operator, and then reflects the relationship between the obtained solution and the optimal solution through the fitness function. The traditional path planning algorithm takes the total flight distance as the only fitness function, as shown in Equation (7). In the path planning algorithm under signal constraints, considering the influence of signal quality, the 5G signal quality and the total distance are taken as the fitness function at the same time to select the population in order to find the optimal solution, as shown in Equation (8).

[0099]

[0100] In the formula, S total represents the total fitness function, nsalesmen represents the number of UAVs, n represents the total number of inspection towers, and S ijk represents the distance from the i-th UAV flying from the j-th tower to the k-th tower, and pathloss ijk represents the signal quality of the i-th UAV flying from the j-th tower to the k-th tower.

[0101] The path loss values at each tower can be calculated through the 5G signal model obtained above, including the line-of-sight, non-line-of-sight propagation loss values, and large-scale spatial loss values. Then, according to the maximum allowable path loss value of the signal, the signal quality pathloss at the current tower can be obtained. Its calculation method is as shown in Equation (9). Substituting it into Equation (8) can obtain the improved fitness function.

[0102] pathloss = [(Pr LOS ·PL UMa-LOS )+(Pr NLOS ·PL UMa-NLOS )] / PL max (9)

[0103] Compared with the traditional path planning algorithm, the total distance does not account for the most important factor in the algorithm after adding signal constraints. When the total flight distances of two groups of breakpoints in the population are similar, the signal quality during flight will be considered. If the weighted distance obtained by multiplying the 5G signal quality and the total flight distance is shorter, then this group of breakpoints will be selected as the optimal path in this generation of the population and compared with the optimal paths in other populations, and the final optimal solution is obtained through iteration.

[0104] The path planning algorithm under signal constraints consists of the following five parts:

[0105] 1) Initialize variables and generate an initial population.

[0106] There are n target points known. The target points are numbered, and the numbers from 1 to n are randomly shuffled m - 1 times (m can take 80), and together with the order before shuffling, m groups of numbered data are stored in a matrix. There are nSalesmen drones performing inspection tasks. Dividing a set of data into nSalesmen segments requires nSalesmen - 1 breakpoints. Then, according to the requirement of the minimum number of target points to be inspected by each drone, m groups of breakpoints are randomly generated and stored in the matrix. One set of numbers corresponds to one set of breakpoints. The set of numbers is segmented at the breakpoints, and each segment is assigned to a drone and inspected in the shuffled order. Thus, m paths are obtained as the initial population.

[0107] 2) Calculate the fitness function.

[0108] By writing a subfunction pathloss, calculate the signal attenuation value at a certain distance from the base station, and divide it by the maximum allowable path loss to obtain the signal attenuation degree (i.e., signal quality), which is used as one of the weights of the fitness function.

[0109] At the same time, calculate the distance traveled by each drone in this generation, which is used to judge whether the drone meets the endurance time constraint. The calculation formula is as follows:

[0110]

[0111] In the formula, endurance is the endurance time of the drone, height is the flight altitude, dUAV max is the longest flight distance of a single drone, num represents the number of inspection poles, acce x represents the acceleration distance, slowdown x represents the deceleration distance, Vsalesmen represents the flight speed, slowdown t represents the deceleration time, and Wsalesmen represents the angular velocity during inspection.

[0112] 3) Select excellent parent generations.

[0113] Under the condition of meeting the endurance requirements of the drone, select and store the information of the minimum product of the total distance and signal quality in each generation of the population, including the total flight distance of all drones on this path, the order of the pole numbers inspected by each corresponding drone, and the breakpoint information. The judgment formula is as follows:

[0114]

[0115] In the formula, minDist represents the shortest path in this generation of the population, globalMin represents the historical shortest path, and Endurance represents the maximum endurance time of a single drone.

[0116] 4) Generate offspring using excellent parents to produce the next generation of the population.

[0117] By operating such as randomly swapping the order of target points of parents and randomly changing breakpoint information, simulate the crossover and mutation operations in genetics, so as to generate m paths of the next generation of the population.

[0118] 5) Loop from 2) to 4) until the number of iterations is satisfied, and then output the shortest path.

[0119] 3. Simulation Verification and Analysis

[0120] To verify the feasibility of the algorithm, conduct simulation analysis on the path planning algorithm without added constraints and the path planning algorithm after constraints respectively, and obtain the final path planning result diagrams, as Figure 2 and Figure 3 shown. The radius of the dashed circle in the figure represents the quality of the signal of the pole tower.

[0121] Figure 2 It is the path planning result without adding 5G signal constraints. It can be seen from the figure that the only selection criterion for the UAV path planning is the length of the distance. The criterion for the UAV to select a point each time is to select the pole tower that is the closest to itself and can ultimately lead to a certain landing point for inspection, in order to minimize the distance. This will result in poor signal quality at some pole towers that are far from the signal tower, and the UAV cannot perform maintenance quickly, affecting the inspection quality.

[0122] Figure 3 It is the UAV path planning result after adding 5G signal constraints. It can be seen from the figure that after adding signal constraints, the selection criteria for the UAV path planning are not only to be close in distance, but also to have a strong signal. Since the UAV communicates with only one base station each time, the strength of the signal here is equivalent to being as close as possible to the nearest takeoff and landing point (base station).

[0123] The path planning at the pole towers that are far from the takeoff and landing point (base station) is similar to the result without adding 5G signal constraints. The reason is that the farther the distance, the smaller the slope of the path loss and the smaller the difference, so the criterion becomes distance priority again; while at the pole towers that are closer to the takeoff and landing point (base station), due to the influence of the weight of signal strength, the pole tower closer to the base station will be selected for priority inspection under the condition of similar distances, in order to minimize the final weighted distance.

[0124] To quantitatively analyze the differences in path planning before and after constraints, count the inspection order and total flight distance of each UAV of the two algorithms. The results are shown in Table 1 and Table 2.

[0125] Table 1 Comparison of UAV Inspection Orders before and after Constraints

[0126]

[0127] As can be seen from Table 1, compared with the original inspection order, the overall inspection order after constraint is further optimized, reducing the long flight routes with poor signal quality in the original planned route, such as the No. 1 UAV from Tower 18 to Tower 11 and the No. 5 UAV from the starting point to Tower 51. This change is consistent with the improvement algorithm that takes both the 5G signal quality function and the distance function as fitness functions, further improving the inspection speed and efficiency.

[0128] Comparison of UAV flight distances before and after constraint in Table 2

[0129] Total distance (m) Distance with poor signal quality (m) Path planned by traditional method 9033.1 1012.9 Path planned by improved method 9139.7 555.1

[0130] As can be seen from Table 3, due to the existence of signal quality constraints, the total distance of the improved method is 106.6 meters longer than that of the traditional method, but the total flight distance of the towers with poor signal quality is significantly shorter than that of the traditional method, reducing the distance with poor signal by 45.2%, improving the 5G signal quality during UAV power inspection.

[0131] In another embodiment, the present invention proposes a collaborative path planning system for 5G power inspection UAVs, which corresponds to the collaborative path planning method for 5G power inspection UAVs in the first embodiment, including:

[0132] A path loss calculation module for establishing a 5G signal quality distribution model in the power inspection scenario and calculating the path loss value at each tower based on the 5G signal quality distribution model;

[0133] A signal quality calculation module for calculating the signal quality at the current tower according to the path loss value at each tower and in combination with the maximum path loss value of the signal;

[0134] A path planning module for adding 5G signal constraints to the UAV path planning algorithm, and the UAV path planning algorithm uses a genetic algorithm that takes signal quality and the total flight distance of the UAV as fitness functions;

[0135] An output module for performing optimal solution using the UAV path planning algorithm to obtain the optimal path for power inspection.

[0136] In another embodiment, the present invention proposes a computer-readable storage medium storing a computer program, and the computer program causes a computer to execute the collaborative path planning method for 5G power inspection UAVs as described in the first embodiment.

[0137] In another embodiment, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the collaborative path planning method for the 5G power inspection drone as described in the first embodiment is implemented.

[0138] In the embodiments disclosed in the present application, the computer storage medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the computer storage medium would include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0139] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present application can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0140] The above are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art in the technical field, several improvements and refinements made without departing from the principle of the present invention should be regarded as within the protection scope of the present invention.

Claims

1. A collaborative path planning method for 5G power inspection drones, characterized in that, Including: Establish a 5G signal quality distribution model in the power inspection scenario. Based on the 5G signal quality distribution model, calculate the path loss value at each tower; the specific 5G signal quality distribution model is as follows: Considering the distribution of power towers, characterize the coverage performance of the 5G signal base station channel propagation model: The formula for the line-of-sight probability is as follows: Where, Where Pr LOS represents the line-of-sight probability, d 2D-out represents the planar distance between the pole tower and the user end outdoors, h UT represents the height of the user; The formula for the loss during line-of-sight propagation is as follows: Where, PL1 = 28.0 + 22 log 10 (d 3D ) + 20 log 10 (f c ) PL2 = 28.0 + 40 log 10 (d 3D ) + 20 log 10 (f c ) - 9 log 10 ((d' BP ) 2 +(h BS -h UT ) 2 ) Wherein, PL UMa-LOS represents the large-scale spatial loss value during line-of-sight propagation, h BS is the height of the pole tower, h UT is the flight height of the UAV, d 2D is the planar distance between the pole tower and the UAV, d 3D is the three-dimensional distance between the pole tower and the UAV, d' BP is the break point distance, f c is the signal frequency; The formula for the loss during non-line-of-sight propagation is as follows: PL UMa-NLOS = max(PL UMa-LOS , PL' UMa-NLOS ) PL' UMa-NLOS = 13.54 + 39.08 log 10 (d 3D ) + 20 log 10 (f c ) - 0.6(h UT - 1.5) where, PL UMa-NLOS represents the large-scale spatial loss value in the non-line-of-sight propagation process; The maximum allowable path loss of the signal is measured according to the link budget, and its formula is as follows: PL max = P TX - L f + G TX - M f - M I + G RX - SR Wherein, PL max represents the maximum allowable path loss value of the signal, P TX is the base station transmission power, L f is the feeder loss, G TX is the base station antenna gain, M f is the shadow fading and fading margin, M I is the interference margin, G RX is the receiving end antenna gain, and SR is the receiving end sensitivity; According to the path loss value at each pole tower and in combination with the maximum path loss value of the signal, the signal quality at the current pole tower is calculated; the path loss value at each pole tower includes the large-scale spatial loss value PL during line-of-sight propagation UMa-LOS and the large-scale spatial loss value PL during non-line-of-sight propagation UMa-NLOS ; the signal quality at the current pole tower is as follows: pathloss = [(Pr LOS ·PL UMa-LOS ) + (Pr NLOS ·PL UMa-NLOS )] / PL max where pathloss represents signal quality, Pr LOS represents the line-of-sight probability, PL UMa-LOS represents the large-scale spatial loss value during line-of-sight propagation, Pr NLOS represents the non-line-of-sight probability, PL UMa-NLOS represents the large-scale spatial loss value during non-line-of-sight propagation, PL max represents the maximum allowable path loss value of the signal; Add 5G signal constraints to the UAV path planning algorithm. The UAV path planning algorithm uses a genetic algorithm with signal quality and the total flight distance of the UAV as the fitness function; the fitness function of the UAV path planning algorithm is as follows: Where S total represents the total fitness function, nsalesmen represents the number of UAVs, n represents the total number of inspection poles and towers, and S ijk represents the distance that the i-th UAV flies from the j-th pole and tower to the k-th pole and tower, and pathloss ijk represents the signal quality that the i-th UAV flies from the j-th pole and tower to the k-th pole and tower; Use the UAV path planning algorithm to perform optimal solution to obtain the optimal path for power inspection.

2. The collaborative path planning method for a 5G power inspection unmanned aerial vehicle according to claim 1, characterized in that: The step of using the UAV path planning algorithm to perform optimal solution to obtain the optimal path for power inspection specifically includes the following steps: 1) Given n target points, number the target points, randomly shuffle the numbers from 1 to n m - 1 times, and store the m groups of numbered data together with the order before shuffling into a matrix; there are nSalesmen UAVs performing inspection tasks, and use nSalesmen - 1 breakpoints to divide each group of numbered data in the matrix into nSalesmen segments, and randomly generate m groups of breakpoints and store them in the matrix; each group of numbered data is segmented at the breakpoint, and each segment is assigned to a UAV and inspected in order, and a total of m paths are obtained as the initial population; 2) Calculate the signal attenuation value at a certain distance from the base station, and divide it by the maximum allowable path loss value of the signal to obtain the signal attenuation degree as the weight of the fitness function; at the same time, calculate the distance of each UAV to judge whether the UAV meets the endurance time constraint; 3) Under the condition of meeting the UAV endurance requirement, select and store the information of the minimum product of the total distance and signal quality in each generation of population, including the total flight distance of all UAVs on this path, the order of the tower numbers inspected by each corresponding UAV, and the breakpoint information; 4) Simulate the crossover and mutation operations in genetics by randomly swapping the order of the target points of the parent generation and randomly changing the breakpoint information, so as to generate m paths of the next generation population; 5) Loop steps 2) to 4) until the iteration times are met, and output the shortest path as the optimal path.

3. The collaborative path planning method for a 5G power inspection drone according to claim 2, wherein: In step 2), the endurance time constraint is calculated by the following formula: endurance= (2·height + dUAV max -(num + 1)·(acce x + slowdown x )) / Vsalesmen+(num+1)·2·slowdown t +π / Wsalesmen·(num-1) Where endurance is the endurance time of the UAV, height is the flight altitude, dUAV max is the longest flight distance of a single UAV, num represents the number of inspection towers, acce x represents the acceleration distance, slowdown x represents the deceleration distance, Vsalesmen represents the flight speed, slowdown t represents the deceleration time, and Wsalesmen represents the angular velocity during inspection.

4. A collaborative path planning system for 5G power inspection drones, characterized in that, Including: A path loss calculation module for establishing a 5G signal quality distribution model in the power inspection scenario. Based on the 5G signal quality distribution model, calculate the path loss value at each tower; the specific 5G signal quality distribution model is as follows: Considering the distribution of power towers, characterize the coverage performance of the 5G signal base station channel propagation model: Where Pr LOS represents the line-of-sight probability, d 2D-out represents the horizontal distance between the pole tower and the user end outdoors, h UT represents the height of the user; The formula for the line-of-sight probability is as follows: PL1 = 28.0 + 22 log 10 (d 3D ) + 20 log 10 (f c ) PL2 = 28.0 + 40 log 10 (d 3D ) + 20 log 10 (f c ) - 9 log 10 ((d' BP ) 2 +(h BS -h UT ) 2 ) Wherein, PL UMa-LOS represents the large-scale spatial loss value during line-of-sight propagation, h BS is the height of the pole tower, h UT is the flight height of the UAV, d 2D is the planar distance between the pole tower and the UAV, d 3D is the three-dimensional distance between the pole tower and the UAV, d' BP is the break point distance, f c is the signal frequency; Where, The formula for the loss during line-of-sight propagation is as follows: Where, The formula for the loss during non-line-of-sight propagation is as follows: PL UMa-NLOS = max(PL UMa-LOS , PL' UMa-NLOS ) PL' UMa-NLOS = 13.54 + 39.08 log 10 (d 3D ) + 20 log 10 (f c ) - 0.6(h UT - 1.5) wherein, PL UMa-NLOS represents the large-scale spatial loss value in the non-line-of-sight propagation process; The maximum allowable path loss of the signal is measured according to the link budget, and its calculation formula is as follows: PL max = P TX - L f + G TX - M f - M I + G RX - SR Wherein, PL max represents the maximum allowable path loss value of the signal, P TX is the base station transmit power, L f is the feeder loss, G TX is the base station antenna gain, M f is the shadow fading and fading margin, M I is the interference margin, G RX is the receiving end antenna gain, and SR is the receiving end sensitivity; A signal quality calculation module, which is used to calculate the signal quality at the current pole tower according to the path loss value at each pole tower and in combination with the maximum path loss value of the signal; the path loss value at each pole tower includes the large-scale spatial loss value PL during line-of-sight propagation UMa-LOS and the large-scale spatial loss value PL during non-line-of-sight propagation UMa-NLOS ; the signal quality at the current pole tower is as follows: pathloss = [(Pr LOS ·PL UMa-LOS )+(Pr NLOS ·PL UMa-NLOS )] / PL max where pathloss represents signal quality, Pr LOS represents the line-of-sight probability, PL UMa-LOS represents the large-scale spatial loss value during line-of-sight propagation, Pr NLOS represents the non-line-of-sight probability, PL UMa-NLOS represents the large-scale spatial loss value during non-line-of-sight propagation, PL max represents the maximum allowable path loss value of the signal; A path planning module, configured to add 5G signal constraints to the UAV path planning algorithm, where the UAV path planning algorithm adopts a genetic algorithm that uses signal quality and the total flight distance of the UAV as a fitness function; the fitness function of the UAV path planning algorithm is as follows: Where S total represents the total fitness function, nsalesmen represents the number of UAVs, n represents the total number of inspection towers, and S ijk represents the distance from the j-th tower to the k-th tower by the i-th UAV, and pathloss ijk represents the signal quality from the j-th tower to the k-th tower by the i-th UAV; An output module, configured to perform optimal solution using the UAV path planning algorithm to obtain the optimal path for power inspection.

5. A computer-readable storage medium storing a computer program, characterized in that, The computer program causes the computer to execute the collaborative path planning method for a 5G power inspection UAV according to any one of claims 1-3.

6. An electronic device, characterized in that, Including: A memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the collaborative path planning method for a 5G power inspection UAV according to any one of claims 1-3 is implemented.

Images