Urban road inspection path planning method based on GIS

Abstract

The invention discloses a GIS-based urban road inspection path planning method, and belongs to the technical field of path planning. Comprising the following steps: acquiring information of each road in a city by using a GIS database, and performing gridding processing; the standard height value of each road is determined based on the coincidence condition of the road imaging edge and the actual road edge during unmanned aerial vehicle inspection; obtaining an inspection position error of each grid center point based on an evaluation result of the inspection accuracy when the inspection is carried out at each grid center point; the total cost of the unmanned aerial vehicle from the starting point to the grid center points is obtained based on the inspection position errors of all the grid center points between the starting point and each grid center point; and obtaining an optimal inspection path during inspection based on the total cost by using an A star algorithm. According to the invention, the flight height and the imaging quality are used as influence factors of path planning, so that the planned inspection path is more accurate, the probability of missed inspection is reduced, and the accuracy of the inspection result is improved.

Description

Technical Field

[0001] The present application relates to the technical field of path planning, and in particular to a method for urban road inspection path planning based on GIS. Background Art

[0002] At present, drones are usually used to inspect urban roads, and the inspection routes of drones are planned based on the road information in the GIS (Geographic Information System) database. At the same time, they are used to collect road information and assist urban road inspections, making it easier to observe and grasp urban road information.

[0003] When drones are used for urban road inspections, it is necessary to limit the flight altitude and no-fly zones of drones according to relevant regulations. On this basis, the coordinate information from the starting point to the target point is set, and the urban road inspection path is obtained using a path planning algorithm. Among them, the GIS database can provide a large amount of spatial information and coordinate information. When using the A-star algorithm to determine the urban road inspection path, usually only the distance cost between the starting point and the target point is considered. However, since the drone needs to dynamically adjust its flight altitude according to the height of surrounding buildings, obstacles and other objects during flight, the resolution of road imaging during inspection changes, resulting in blurred inspection images, which can easily lead to false detection problems during urban road inspections. Summary of the invention

[0004] In order to address the deficiencies in the prior art, the present application provides a GIS-based urban road inspection path planning method. The method is based on A-star algorithm calculation. When calculating the actual cost between the starting point and the current grid, the position quality of the road image collected by the drone above the center point of each grid is introduced. The height of the drone flying above the center point of each grid is used as a factor in planning the inspection path, so that the planned inspection path is more accurate, and the road image information collected by the drone along the planned path can be clearer, thereby improving the inspection quality.

[0005] The purpose of this application is to provide a GIS-based urban road inspection path planning method, including the following steps: Use GIS database to obtain information about each road in the city and perform grid processing; The standard altitude value for the drone to fly above each road is determined based on the overlap between the edge of the road imaged during the drone inspection and the actual edge of the road; Based on the evaluation results of the inspection accuracy when the road inspection is performed at each grid center point on each road, the inspection position error of each grid center point on each road is obtained; The total cost of the drone from the starting point to the center point of the grid is obtained based on the inspection position error of all grid center points between the starting point and each grid center point; The A-star algorithm is used to obtain the optimal inspection path for UAVs when inspecting urban roads based on the total cost of each grid center point.

[0006] In one embodiment, the method for obtaining the standard height value is: When the UAV inspects each road, the flight height of the UAV when the edge line in the road image coincides with the actual edge of the road is taken as the standard height value of each road.

[0007] In one embodiment, the method for obtaining the inspection position error of each grid on each road is: Determine the distance between each grid and the central axis of the road where each grid is located; Calculate the absolute value of the difference between the actual altitude value of the drone flying on each grid on each road and the standard altitude value of the drone on each road; The product of the absolute value and the distance is used as the inspection position error of each grid center point on each road.

[0008] In one embodiment, the total cost of the drone from the starting point to the center point of the grid is determined by: Determine the error characteristic value of each grid center point based on the inspection position error of all grid center points between the starting point and each grid; determine the actual cost of each grid center point based on the error characteristic value, combined with the inspection position error of each grid center point and the inspection position error of the starting point; The total cost from the starting point to each grid is determined based on the estimated cost from the starting point to each grid and the actual cost of the center point of each grid.

[0009] In one embodiment, the error characteristic value of each grid center point is determined as follows: For each grid on any road, the inspection position errors of all grid center points between the starting point and each grid are arranged in the order of the Euclidean distance from the starting point from small to large to obtain the characteristic sequence of each grid center point; The mean value of the fitting slopes of all elements in the characteristic sequence is obtained by data fitting as the error characteristic value of each grid center point.

[0010] In one embodiment, the actual cost of each grid center point is determined as follows: Calculate the difference between the inspection position error of each grid center point and the inspection position error of the starting point; and take the product of the mapping result of the difference in the exponential function and the mapping result of the error characteristic value of each grid center point in the exponential function as the actual cost of each grid center point.

[0011] In one embodiment, the total cost from the starting point to each grid is determined as follows: The product of the Euclidean distance between the starting point and each grid center point and the actual cost of each grid center point is calculated, and the sum of the product and the estimated cost of each grid center point is taken as the total cost from the starting point to each grid.

[0012] In one embodiment, the method for obtaining the actual altitude value of the drone flying on each grid on each road is: If the elevation of the road is not taken into account, the actual altitude value of the drone when flying above each grid is based on the altitude value obtained by the GPS module on the drone; If the altitude of the road is taken into account, the difference between the altitude value obtained by the GPS module and the altitude of the road is taken as the actual altitude value of the drone.

[0013] In one embodiment, the method for obtaining the optimal inspection path is: Determine the starting point of the path to be planned; Define two queues, openlist and closelist. Openlist stores grids to be explored, and closelist stores grids that have been explored. The A-star algorithm is used to cyclically update the grids in the openlist and closelist of the two queues based on the calculation method of the total cost V until there is an end point in the openlist or the openlist is empty, then the search stops; Based on the parent node determined during the search process, the optimal inspection path can be obtained by tracing back from the end point along the parent node pointer until it reaches the starting point.

[0014] In one embodiment, the inspection position error of the starting point of the path to be planned is initialized to 0.

[0015] The beneficial effects of the present application are as follows: the present application provides a GIS-based urban road inspection path planning method, which firstly performs grid processing on the urban inspection roads so that the accuracy of each flight position in the subsequently planned UAV flight path is high enough to avoid positioning deviation; secondly, the standard flight altitude of each road is determined based on the analysis of the road imaging error when the UAV inspects each road, and the inspection error caused by the change in flight altitude is used as an influencing factor for the subsequent calculation of the total cost, thereby successively improving the accuracy of selecting grids; thereafter, based on the change characteristics of the inspection error when the UAV between the starting point and each grid center point passes through the remaining grids, the overall level of the inspection error in the process from the starting point to the jth grid center point of the ith road is evaluated, and the actual cost of each grid center point is obtained; by optimizing the calculation method of the total cost in the A-star algorithm, the position quality of the road image collected by the UAV above each grid center point and the flight altitude are introduced as influencing factors for planning the inspection path, so that the planned inspection path is more accurate, and the road image information collected by the UAV along the planned path can be clearer, thereby improving the inspection quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] Figure 1 This is a schematic diagram of the implementation process of a GIS-based urban road inspection route planning method of this application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0020] This application provides a GIS-based urban road inspection path planning method, see Figure 1 As shown, the following steps are included: S001. Use GIS database to obtain information of various roads in the city and perform grid processing.

[0021] This application aims to obtain the inspection path during the inspection of urban roads, so first of all, it is necessary to obtain the relevant data of each road in the city. In this application, the information of each road in the city is obtained according to the urban road GIS database, and the roads are gridded to obtain grids evenly distributed on the urban roads.

[0022] Specifically, the GIS database mainly includes the geographic data of various geographic targets within the urban space, including but not limited to buildings, roads, subway stations, and squares within the urban area; for this reason, in an embodiment of the present application, by obtaining the GIS database of a certain city road, the information of each road in the city is obtained.

[0023] Furthermore, the existing GIS database is used to obtain the width of each road in the current city in the world coordinate system. It should be noted that if the GIS database has the width level of each road, then according to the width range corresponding to each width level, the middle value of the corresponding width range is taken as the width of the current city road. If the GIS database has the length and area of ​​each city road segment, since the width of each section of the city road in the same city is basically the same, the area can be divided by the length to obtain the corresponding width of each city road.

[0024] Secondly, the road surface is gridded to obtain evenly distributed grids on the road surface. It should be noted that the gridding of the road can determine the distance from each point on the road to the central axis of the road, and it is also convenient for the subsequent determination of the specific position of the drone above the center point of the grid.

[0025] S002. Obtain the distance between the center point of each grid and the central axis of the road and the actual height value of the drone when flying above each grid, and determine the standard height value of the drone flying above each road.

[0026] In the process of using drones to inspect urban roads, if the width of the road increases and the height remains unchanged during the inspection, the drone will not be able to obtain complete information about the road, resulting in missed inspections during the inspection of urban roads; on the other hand, if the path planning of the drone is too biased towards the side of the road during the inspection, it may also lead to missed inspections during the inspection of urban roads due to the inability to capture the road information on the other side of the road; and for narrow urban roads, if the height of the drone remains unchanged during the inspection, the narrow route information obtained by the drone will be blurred, and false detections will also occur. Therefore, the flight altitude is usually lowered to ensure that the road is not missed while ensuring sufficient clarity of the road imaging. Based on the above analysis, the height of the drone flying above each grid center point is used as a factor in planning the inspection path, making the planned inspection path more accurate, and enabling the drone to collect clearer road image information along the planned path, thereby improving the accuracy of the inspection results.

[0027] Firstly, according to the road information in the GIS database, the distance between the center point of each grid and the central axis of the road after the gridding process is determined.

[0028] Preferably, as an embodiment of the present application, the distance between the center point of each grid and the central axis of the road is obtained by obtaining the position information of the center point of each grid on the corresponding road according to the GIS database, and calculating the Euclidean distance between the center point of each grid and the central axis of the road surface where it is located in the world coordinate system.

[0029] In another embodiment, the width of each road in the current city in the world coordinate system can be obtained by counting the number of grids n between each grid and the central axis, and the ratio of the width of each road to the total number of grids is used as the width L of each grid. Then, the distance from the center point of each grid to the central axis of the road surface is For example, if the width of a road is 9 meters and there are 9 grids along the width of the road, then the size of each grid is 1 meter, and the distance between the center point of the grid at the edge of the road and the central axis of the road is rice.

[0030] Secondly, when the drone is inspecting urban roads, the inspection flight altitude of the drone is adjusted to the standard height value of each road according to whether the edge line in the road image coincides with the actual edge of the road. When the edge lines on both sides of the road displayed in the road image coincide with the actual edge, it is considered that the flight altitude of the drone on the road has reached the standard height value; otherwise, it is considered that the height of the drone during the road inspection has not reached the standard height value. The purpose of this is to reduce the probability of missed detection and false detection during the inspection process by dynamically adjusting the flight altitude of the drone to the standard height value of each road according to the actual width of each road during the inspection process.

[0031] It should be noted that the actual altitude value of the drone when flying above each grid is based on the altitude value obtained by the GPS module on the drone if the altitude of the road is not taken into account. If the altitude of the road is taken into account, the altitude value obtained by the GPS module minus the altitude of the road is required, and the difference is used as the actual flight altitude of the drone.

[0032] S003. Analyze the evaluation results of the inspection accuracy when performing road inspections at the center point of each grid on each road, and obtain the inspection position error of each grid on each road.

[0033] When planning the route for drones to conduct road inspections, adjusting the drone's flight altitude to the standard altitude value for each road can avoid the impact of missed inspections or unclear road imaging on the inspection results. However, when the drone switches from the current road to the next road, or is affected by the unstable positions of buildings, road signs, roadblocks, etc. on both sides of the road, the altitude of the drone during road inspections is not always maintained at the standard altitude value of each road, which will cause a certain difference between the road imaging results on the drone and the imaging results at the standard altitude value. When using the A-star algorithm to plan the training path of the drone, the difference between the road imaging results and the imaging results at the standard altitude value should be made as small as possible to ensure the accuracy of the road inspection results.

[0034] Furthermore, the position quality of the drone collecting road images above each grid is obtained based on the distance between the center point of each grid and the central axis of the road surface, the standard altitude value of the drone flying above each road, and the actual altitude value of the drone when flying above each grid.

[0035] Specifically, by comparing the actual flight altitude of each grid in each road when the UAV inspects each road with the standard altitude value of the UAV when inspecting each road, the inspection error caused by the altitude difference when the UAV inspects each grid is evaluated. Then, combined with the distance between the center of the grid and the central axis of the road, the overall error of the UAV when flying in each grid is comprehensively evaluated, and then the cost of the UAV using each grid as an inspection path is determined.

[0036] Here, the inspection position error of each grid on each road is calculated to reflect the error when the drone performs road imaging inspection above the center point of each grid. The calculation formula for the inspection position error of the jth grid on the i-th road is: In the formula, represents the inspection position error of the jth grid in the i-th road; Indicates the standard height value of the drone's inspection above the i-th road; represents the actual height value of the UAV when it flies above the jth grid in the i-th road; It represents the distance between the center point of the jth grid in the i-th road and the center axis of the road surface of the i-th road.

[0037] It should be noted that the distance between the center point of the jth grid and the central axis of the ith road is The larger the value, the more likely it is that the UAV will be missed when the jth grid is used as a flight position on the planned inspection path. The larger the value, the closer the UAV is to the boundary of the i-th road. When switching from the i-th road to the i+1-th road, the UAV needs to make a larger adjustment in the flight angle to reduce the impact of buildings and obstacles on both sides of the road, while ensuring that there is no obvious deviation in the inspection position on the i+1-th road. The higher the planning cost of the A-star algorithm is when the j-th grid is used as a flight position on the planned inspection path. and The larger the difference is, the greater the difference between the current UAV's flight altitude and the standard altitude on the i-th road, the worse the resolution of the road imaging quality, and the higher the probability of missed inspections; that is, The larger the value of , the greater the error when the UAV inspects the road through the jth grid in the i-th road.

[0038] S004. Obtain the total cost of the drone from the starting point to the grid center point based on the inspection position errors of all grid center points between the starting point and each grid center point.

[0039] When using the A-star algorithm to plan a route, it is necessary to first determine the location information of the starting point and the end point. In this application, according to the coordinate information of the starting point and the end point of the urban road inspection, the grids where the starting point and the end point are located are found in the gridded road respectively.

[0040] It should be noted that at present, the A-star algorithm only estimates the expected cost from the grid to the end point through Manhattan distance or Euclidean distance as the heuristic function. This type of calculation method ignores the dynamic adjustment of flight altitude on different roads and the impact of different standard altitude values ​​on the total estimate when planning the UAV urban road inspection path, which will lead to certain inspection errors in the planned inspection path.

[0041] Based on this, the present application calculates the actual cost G of moving the starting point along the planned path to the center point of the grid according to the distance between the starting point and the center point of the grid where it is located. Secondly, the estimated cost H is determined based on the inspection position error of the drone in each grid of each road and the distance between the center point of the grid to which the starting point belongs and the end point, and the sum of the actual cost G and the estimated cost H is used as the total estimate of the starting point. In this way, the calculation method of the total estimate is optimized by comparing the actual flight altitude with the standard altitude value and the distance of each grid from the central axis of the road, which not only considers the inspection path distance of the drone but also takes into account the road imaging quality of the drone on each road, thereby reducing the probability of missed detection.

[0042] Furthermore, the inspection position errors of the starting point and each grid center point on each road are calculated respectively. For any grid center point, taking the jth grid center point of the ith road as an example, the inspection position errors of all center points between the starting point and the jth grid center point of the ith road are arranged in order of the Euclidean distance from the starting point from small to large as the feature sequence of the jth grid center point of the ith road, and the mean of the slope of the elements in the feature sequence is obtained by using a data fitting algorithm as the error feature value of the jth grid center point of the ith road, so as to characterize the overall level of the inspection error in the process from the starting point to the jth grid center point of the ith road.

[0043] Among them, data fitting is a commonly used technology in the field of sequence analysis, and commonly used data fitting algorithms include but are not limited to linear fitting, polynomial fitting, and least squares fitting. Preferably, as an embodiment of the present application, the slope of the elements in the characteristic sequence is determined by least squares fitting. Least squares fitting is a well-known technology, and the specific process is not repeated here.

[0044] Specifically, in this application, the total cost V of the drone from the starting point to each grid center point on each road is calculated as follows: In the formula, is the actual cost of training from the starting point to the jth grid center point of the i-th road, is the error characteristic value of the jth grid center point on the i-th road, is an exponential function with a natural constant as base, , are the inspection position errors of the starting point and the jth grid in the ith road respectively; is the total cost of the drone from the starting point to the center point of the jth grid on the i-th road, is the Euclidean distance between the starting point and the center point of the jth grid in the i-th road, is the estimated cost of the center point of the jth grid on the i-th road, The size is equal to the Euclidean distance between the center point of the jth grid and the end point in the i-th road.

[0045] in, It reflects the change in the inspection position error of the drone from the starting point to the jth grid center point of the i-th road, and then uses the error characteristic value of the jth grid center point of the i-th road The change is weighted to represent the actual cost of the UAV from the starting point to the jth grid center point of the i-th road. ; It reflects that the farther the distance to the center point of the jth grid of the i-th road is, the greater the inspection distance loss cost from the starting point to the center point of the jth grid of the i-th road is; Indicates the distance of the current grid when it moves to the end point. The longer the distance, the greater the loss of the distance traveled. The larger the value of The larger the value of , the greater the total cost of the drone from the starting point to the center point of the jth grid on the ith road during the inspection. The reliability of the inspection result of the current grid will be worse than that of the inspection result at the starting point, and this grid should not be selected during path planning.

[0046] S005. Use the A-star algorithm to obtain the optimal inspection path for the drone when inspecting urban roads based on the total cost of each grid center point.

[0047] First, based on the A-star algorithm, at each step of the drone's flight during the road inspection process, the grid center point with the smallest total cost is selected as the drone's flight position until it reaches the end of the road inspection. The determined flight path is the optimal inspection path for the drone when conducting urban road inspections.

[0048] Specifically, in the embodiment of the present application, based on the A-star algorithm, first determine the starting point of the path to be planned, take any grid center point at the starting end of the path to be planned as the starting point, define two queues, openlist and closelist, the openlist stores the grids to be explored, and the closelist stores the grids that have been explored; and initialize the inspection position error of the starting point to 0, and use the A-star algorithm based on the calculation method of the total cost V to cyclically update the grids in the two queues, openlist and closelist, until there is an end point in the openlist, or the openlist is empty, and the search is stopped. Based on the parent node determined during the search process, reverse backtracking from the end point according to the parent node pointer until backtracking to the starting point can obtain the optimal inspection path, ensuring that when the optimal inspection path is selected for inspection, the current drone inspection quality of urban roads can also be guaranteed. Among them, the A-star algorithm is a commonly used technology in the field of path planning, and the specific cyclic iteration process will not be repeated.

[0049] In summary, the present application provides a GIS-based urban road inspection path planning method. This method introduces the quality evaluation of road images collected by drones above each grid center point when calculating the actual cost between the starting point and the current grid through the A-star algorithm. The adjustment of the flying height of the drone above each grid center point is used as a factor in planning the inspection path, so that the planned inspection path is more accurate, and the road image information collected by the drone along the planned inspection path can be clearer, thereby improving the inspection quality.

[0050] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0051] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A GIS-based urban road inspection path planning method, characterized in that: The following steps are involved: Use GIS database to obtain information about each road in the city and perform grid processing; The standard altitude value for the drone to fly above each road is determined based on the overlap between the edge of the road imaged during the drone inspection and the actual edge of the road; Based on the evaluation results of the inspection accuracy when the road inspection is performed at each grid center point on each road, the inspection position error of each grid center point on each road is obtained; The total cost of the drone from the starting point to the center point of the grid is obtained based on the inspection position error of all grid center points between the starting point and each grid center point; The A-star algorithm is used to obtain the optimal inspection path for UAVs to inspect urban roads based on the total cost of each grid center point. The method for determining the total cost of the drone from the starting point to the center point of the grid is: Determine the error characteristic value of each grid center point based on the inspection position error of all grid center points between the starting point and each grid; Based on the error characteristic value, combined with the inspection position error of each grid center point and the inspection position error of the starting point, the actual cost of each grid center point is determined; The total cost from the starting point to each grid is determined based on the estimated cost from the starting point to each grid and the actual cost of the center point of each grid.

2. The GIS-based urban road inspection route planning method according to claim 1 is characterized in that: The method for obtaining the standard height value is: The flight height of the drone when the edge line in the road image coincides with the actual edge of the road when the drone inspects each road is taken as the standard height value of each road.

3. The GIS-based urban road inspection route planning method according to claim 1 is characterized in that: The method for obtaining the inspection position error of each grid on each road is as follows: Determine the distance between each grid and the central axis of the road where each grid is located; Calculate the absolute value of the difference between the actual altitude value of the drone flying on each grid on each road and the standard altitude value of the drone on each road; The product of the absolute value and the distance is used as the inspection position error of each grid center point on each road.

4. The GIS-based urban road inspection route planning method according to claim 1 is characterized in that: The error characteristic value of each grid center point is determined as follows: For each grid on any road, the inspection position errors of all grid center points between the starting point and each grid are arranged in the order of the Euclidean distance from the starting point from small to large to obtain the characteristic sequence of each grid center point; The mean value of the fitting slopes of all elements in the characteristic sequence is obtained by data fitting as the error characteristic value of each grid center point.

5. The GIS-based urban road inspection route planning method according to claim 1 is characterized in that: The actual cost of each grid center point is determined as follows: Calculate the difference between the inspection position error of each grid center point and the inspection position error of the starting point; and take the product of the mapping result of the difference in the exponential function and the mapping result of the error characteristic value of each grid center point in the exponential function as the actual cost of each grid center point.

6. The GIS-based urban road inspection route planning method according to claim 1 is characterized in that: The total cost from the starting point to each grid is determined as follows: The product of the Euclidean distance between the starting point and each grid center point and the actual cost of each grid center point is calculated, and the sum of the product and the estimated cost of each grid center point is taken as the total cost from the starting point to each grid.

7. The method for urban road inspection route planning based on GIS according to claim 3 is characterized in that: The method for obtaining the actual altitude value of the UAV flying on each grid on each road is as follows: If the elevation of the road is not taken into account, the actual altitude value of the drone when flying above each grid is based on the altitude value obtained by the GPS module on the drone; If the altitude of the road is taken into account, the difference between the altitude value obtained by the GPS module and the altitude of the road is taken as the actual altitude value of the drone.

8. The GIS-based urban road inspection route planning method according to claim 1 is characterized in that: The method for obtaining the optimal inspection path is: Determine the starting point of the path to be planned; Define two queues, openlist and closelist. Openlist stores grids to be explored, and closelist stores grids that have been explored. The A-star algorithm is used to cyclically update the grids in the two queues openlist and closelist based on the calculation method of the total cost V until there is an end point in the openlist or the openlist is empty, then the search stops; Based on the parent node determined during the search process, the optimal inspection path can be obtained by tracing back from the end point along the parent node pointer until it reaches the starting point.

9. The GIS-based urban road inspection route planning method according to claim 8 is characterized in that: The inspection position error of the starting point of the path to be planned is initialized to 0.

Images