Path planning method of inspection robot and storage medium

Abstract

The invention provides a path planning method of an inspection robot and a storage medium, and belongs to the technical field of path planning, and the method comprises the steps: taking the inspection robot only turning at a turning point in the inspection process as a target, and converting an inspection undirected graph into an inspection directed graph; determining a mapping point corresponding to each inspection point on the inspection directed graph, and determining an inspection point sequence of all the inspection points of the inspection robot by analyzing a topological relation between the mapping points; and in response to the issued current inspection task, determining the inspection sequence of the to-be-inspected points contained in the current inspection task according to the inspection point sequence, and realizing path planning of the current inspection task by using the inspection directed graph. According to the route planning method of the inspection robot, route planning of multiple target points of the inspection robot can be rapidly achieved according to the inspection requirement of a user, and the safety of the inspection robot and equipment in an inspection area is ensured by introducing a turning point turning method and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a path planning method and storage medium for an inspection robot. Background Art

[0002] With the advancement of technology, the market demand for equipment inspection has increased significantly, the cost of manual inspection has gradually increased, and manual inspection is difficult to complete under certain conditions. These factors have prompted the application of inspection robots (such as quadruped inspection robots).

[0003] At present, the mainstream path planning algorithm used by most inspection robots on the market is the A* algorithm. As a heuristic search algorithm, the A* algorithm can find the shortest path between two points in a known environment and has the advantages of high efficiency and accuracy.

[0004] However, due to the complexity of inspection tasks and scenes, the inspection robot is limited in mobility and has difficulty adapting to complex environments. The traditional A* algorithm can only plan paths between two points, which has great limitations and is difficult to adapt to changing tasks. In addition, since some inspection scenes have many internal devices, narrow roads, and inspection tasks have requirements for direction and require multiple U-turns, the path planned by the traditional A* algorithm may cause the inspection robot to collide when turning at the corner of a narrow road, damaging the equipment in the inspection area and the inspection robot itself. Summary of the invention

[0005] The present invention provides a path planning method for an inspection robot, so as to solve at least one defect existing in the above-mentioned prior art.

[0006] In a first aspect, the present invention provides a path planning method for a patrol robot, comprising: obtaining a skeleton map of an inspection area, determining a turning point of the patrol robot on the skeleton map, and dividing the skeleton map into multiple intervals according to the turning points; when the length of any interval is greater than a preset length, setting path points at preset intervals with one end point of the any interval as a starting point, completing the setting of the path points of the any interval, traversing all intervals, and constructing a patrol undirected graph according to all path points; with the goal that the patrol robot only turns at turning points during the patrol process, converting the patrol undirected graph into a patrol directed graph; on the patrol directed graph, using the directed path point with the shortest distance to each patrol point and the direction in the same preset interval as a mapping point, and determining the patrol point ranking of all patrol points of the patrol robot by analyzing the topological relationship between the mapping points; wherein the patrol points are pre-planned and set, and include a patrol starting point; in response to the current patrol task issued, determining the patrol order of the points to be patrolled contained in the current patrol task according to the patrol point ranking, and using the patrol directed graph to implement path planning for the current patrol task.

[0007] According to the path planning method of the inspection robot provided by the present invention, a skeleton map of the inspection area is obtained, including: converting the SLAM map of the inspection area into a binary black-and-white image, determining the maximum connected area of ​​the white part in the black-and-white image, and generating a safe area map; extracting a skeleton map from the safe area map; wherein the maximum connected area of ​​the white part is the feasible area of ​​the inspection robot.

[0008] According to the path planning method for the inspection robot provided by the present invention, the turning point of the inspection robot is determined on the skeleton map, including: drawing a circle with any pixel point on the skeleton map as the center and a first preset number of pixels as the radius, and counting the number of white boundary pixels of the circle; when the number of white boundary pixels is greater than a second preset number, taking any pixel point as a turning point; traversing all pixel points on the skeleton map to determine all turning points of the inspection robot on the skeleton map.

[0009] According to the path planning method of the inspection robot provided by the present invention, the inspection robot is targeted to turn only at turning points during the inspection process, and the inspection undirected graph is converted into an inspection directed graph, including: Starting from the path point at one end of any interval in the inspection undirected graph, all the path points are traversed in sequence to obtain the direction angle of each path point. There are two specific cases for obtaining the direction angle of the path point: The first case: find the direction angle of each path point from the path point at one end as the starting point to the second to last path point, specifically: determine the connecting line between the currently traversed path point and the next path point to be traversed, and use the angle between the connecting line and the positive direction of the horizontal coordinate axis as the direction angle of the currently traversed path point; wherein the range of the angle is [-90°, 90°); The second case: find the direction angle of the last traversed path point, specifically: take the direction angle of the second to last traversed path point as the direction angle of the last traversed path point; According to the position and direction angle of any path point in the inspection undirected graph, two directed path points corresponding to any path point in the inspection directed graph are generated; wherein the positions of the two corresponding directed path points are the same as the position of any path point, the direction of one of the two directed path points is the direction angle of any path point, and the direction of the other directed path point is the direction angle of any path point plus 180 degrees; Traverse all two adjacent directed path points in any interval. When the directions of the two adjacent directed path points are in the range of [-90°, 90°), add an edge between the two adjacent directed path points, and the direction of the edge points to the positive semi-axis side of the horizontal coordinate axis; when the directions of the two adjacent directed path points are in the range of [90°, 270°), add an edge between the two adjacent directed path points, and the direction of the edge points to the negative semi-axis side of the horizontal coordinate axis; Among them, by analyzing the topological relationship between the mapping points, the inspection point order of all inspection points of the inspection robot is determined, including: The mapping point corresponding to the inspection starting point is used as the starting point, and the path planning is performed with the mapping points corresponding to other inspection points whose order has not been determined, so as to determine the next inspection point with the shortest path; Taking the mapping point corresponding to the next inspection point with the shortest path as the starting point, path planning is performed with the mapping points corresponding to the inspection points whose sorting is not determined, and the next inspection point with the shortest path is determined, and so on, until all the inspection points are sorted.

[0010] According to the path planning method of the inspection robot provided by the present invention, the method for determining the mapping point of any inspection point on the inspection directed graph includes: calculating the distance between any inspection point and each directed path point on the inspection directed graph, and determining the directed path point that is closest to any inspection point; determining, from the directed path points with the closest distance, a directed path point that is in the same preset interval as the direction of any inspection point as the mapping point of any inspection point; the direction of any inspection point is pre-planned and set.

[0011] According to the path planning method for the inspection robot provided by the present invention, the inspection point order of all the inspection points of the inspection robot is determined by analyzing the topological relationship between the mapping points, including: taking the mapping point corresponding to the inspection starting point as the starting point, and performing path planning with the mapping points corresponding to other inspection points whose order has not been determined, so as to determine the next inspection point with the shortest path; taking the mapping point corresponding to the next inspection point with the shortest path as the starting point, and performing path planning with the mapping points corresponding to the inspection points whose order has not been determined, so as to determine the next inspection point with the shortest path, and so on, until the order of all the inspection points is completed.

[0012] According to the path planning method for the inspection robot provided by the present invention, the inspection order of the points to be inspected included in the current inspection task is determined according to the inspection point sorting, and the path planning of the current inspection task is realized by using the inspection directed graph, including: determining the inspection order of the points to be inspected included in the current inspection task according to the relative positions of the points to be inspected in the inspection point sorting, and determining the inspection order of the mapping points corresponding to the points to be inspected based on the inspection order of the points to be inspected; taking the path point closest to the current position of the inspection robot as the starting point, and determining the shortest path between each pair of adjacent mapping points according to the inspection order of the mapping points, so as to realize the path planning of the current inspection task.

[0013] According to the path planning method for the inspection robot provided by the present invention, when there is an entry and exit door in the path planning, it also includes: an entry process: controlling the inspection robot to stop moving before entering a dangerous area, and after determining that the door is completely opened, controlling the inspection robot to pass through the dangerous area and enter the door; an exit process: sending a door closing command when it is determined that the inspection robot has left the dangerous area; wherein the dangerous area is determined based on the location information of the door and the rotation radius of the door, and no path point is set in the dangerous area.

[0014] According to the path planning method of the inspection robot provided by the present invention, the shortest path between each pair of adjacent mapping points is determined, including: when the shortest path between adjacent mapping points has been predetermined, directly adopting the predetermined shortest path; when the shortest path between adjacent mapping points has not been predetermined, using a preset path optimization algorithm to generate the shortest path.

[0015] On the other hand, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the path planning method for the inspection robot as described in any one of the above.

[0016] The path planning method for the inspection robot provided by the present invention can quickly implement the path planning of multiple target points of the inspection robot according to the inspection requirements of the user, and ensure the safety of the inspection robot and the equipment in the inspection area by introducing methods such as turning at turning points.

[0017] The present invention successfully reduces the time complexity significantly (time complexity from It has dropped to ), which not only greatly shortens the planning time, but also realizes efficient path planning for multiple target points.

[0018] The present invention introduces a turning point design. This innovative measure can effectively prevent the inspection robot from colliding at the corners of narrow roads, thereby ensuring that the factory equipment and the inspection robot itself are not damaged during the inspection process. The present invention converts an undirected graph into a directed graph by applying a specific conversion criterion. This improvement greatly reduces the need for the inspection robot to frequently adjust its direction while moving, thereby improving the inspection efficiency and operation stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a flow chart of the path planning method of the inspection robot provided by the present invention; Figure 2 It is a schematic diagram of the process of constructing an inspection undirected graph provided by the present invention; Figure 3 It is a schematic diagram of the SLAM map provided by the present invention; Figure 4 It is a schematic diagram of the maximum connected area of ​​the white part provided by the present invention; Figure 5 It is a schematic diagram of a skeleton map of an inspection area provided by the present invention; Figure 6 It is a schematic diagram for judging a turning point provided by the present invention; Figure 7 It is a waypoint map provided by the present invention; Figure 8 This is a comparison diagram of the effects of path planning using directed graphs and undirected graphs provided by the present invention; Fig. 9 It is a comparison diagram of the undirected graph path diagram and the directed graph path diagram provided by the present invention; Fig.10 It is a path planning schematic diagram with directions provided by the present invention; Fig.11 It is a logic block diagram of the path planning of the current inspection task provided by the present invention; Fig.12 is a schematic diagram of a dangerous area formed by door rotation provided by the present invention; Fig.13 It is a schematic diagram of opening and closing a door during an inspection provided by the present invention; Fig.14It is a schematic diagram of applying the reuse idea in the path planning provided by the present invention; Fig.15 It is the early stage work flow chart provided by the present invention; Fig.16 It is the overall operation flow chart provided by the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] It should be noted that, in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0023] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more.

[0024] Combine the following Figure 1-Figure 16 The path planning method of the inspection robot provided in an embodiment of the present invention is described.

[0025] Figure 1 is a flow chart of the path planning method of the inspection robot provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps: Step 101: Obtain a skeleton map of the inspection area, determine the turning points (also called intersections) of the inspection robot on the skeleton map, and divide the skeleton map into multiple sections according to the turning points.

[0026] Among them, the inspection robot can be a quadruped, wheeled or tracked inspection robot.

[0027] Step 102: When the length of any interval is greater than a preset length, path points are set at preset intervals with one end point of the interval as a starting point, the path points of the interval are set, and all intervals are traversed to construct an inspection undirected graph according to all path points; Step 103: With the inspection robot turning only at turning points during the inspection process as the goal, the inspection undirected graph is converted into an inspection directed graph; Step 104: on the inspection directed graph, the directed path point corresponding to each inspection point with the shortest distance and the same direction in the same preset interval is used as a mapping point, and the inspection point order of all inspection points of the inspection robot is determined by analyzing the topological relationship between the mapping points; wherein the inspection points are pre-planned and set, and include the inspection starting point; Step 105: In response to the issued current inspection task, the inspection order of the inspection points to be inspected included in the current inspection task is determined according to the inspection point ranking, and the path planning of the current inspection task is implemented using the inspection directed graph.

[0028] The specific instructions are given below.

[0029] (1) Generate inspection path points and construct an undirected inspection graph Figure 2 It is a schematic diagram of the process of constructing an inspection undirected graph provided by the present invention, such as Figure 2 The detailed steps are as follows: Step a: Get the SLAM map of the inspection area.

[0030] Specifically, the inspection robot is manually controlled to move to various locations in the inspection area using a remote control. During the movement, the robot will conduct a comprehensive scan of its surrounding environment. Through the onboard radar system, electromagnetic waves are emitted and reflected signals are received, thereby accurately capturing the distance data of the surrounding environment. These data are integrated with the robot's own movement trajectory information to accurately locate the robot's position. In addition, the robot will also carefully compare the newly acquired environmental information with the previous scanning results, continuously update and optimize the SLAM map, and generate the final SLAM map, such as Figure 3 As shown, Figure 3 It is a schematic diagram of the SLAM map provided by the present invention.

[0031] Step b: Convert the SLAM map of the inspection area into a binary black-and-white image, determine the maximum connected area of ​​the white part in the black-and-white image, and generate a safe area map.

[0032] Specifically, for the SLAM map obtained in step a, each pixel is judged. If the pixel value is less than 128, it is assigned a value of 0. If it is greater than 128, it is assigned a value of 255. In this way, the SLAM map is converted into a binary black-and-white image. Then, the largest connected area of ​​the white part is found in the black-and-white image. This area is the feasible area of ​​the inspection robot and is extracted, as shown in Figure 4 As shown, Figure 4 It is a schematic diagram of the maximum connected area of ​​the white part provided by the present invention.

[0033] Step c: extracting a skeleton map from the safe area map.

[0034] Specifically, for each foreground pixel, the maximum connected area obtained in step b is determined based on the pixel status of its surrounding neighborhood to determine whether it can be removed. Only when removing the pixel will not destroy the connectivity of the image or change its topological structure will it be marked as removable. The skeleton can be extracted by repeatedly iteratively refining the binary image through this morphological skeleton extraction method. The extracted skeleton is deburred to obtain a smoother skeleton. The generated skeleton map effect is as follows: Figure 5 As shown, Figure 5 It is a schematic diagram of the skeleton map of the inspection area provided by the present invention.

[0035] Step d: Determine the turning point (intersection point) of the inspection robot on the skeleton map.

[0036] Optionally, a circle is drawn with any pixel point on the skeleton map as the center and a first preset number of pixels as the radius, and the number of white boundary pixels of the circle is counted; when the number of white boundary pixels is greater than a second preset number, any pixel point is taken as a turning point; all pixel points on the skeleton map are traversed to determine all turning points of the inspection robot on the skeleton map.

[0037] Specifically, for the skeleton map generated in step c, draw a circle with one (i.e., the first preset number) pixel as the radius, and check whether the boundary pixels of the circle are white one by one. If the number of white boundary pixels is 3 (i.e., the second preset number) or more, the pixel point is considered to be a turning point. Traverse the entire skeleton map line to find all the turning points to form a turning point set. ,in m is the number of turning points, Figure 6 The points circled in red are all turning points (intersection points). Figure 6 This is a schematic diagram for judging the turning point provided by the present invention: ; in, Represents the first m A turning point.

[0038] Step e: Divide the skeleton map into multiple intervals according to the turning points. When the length of any interval is greater than the preset length, set the path points at preset intervals with one endpoint of the interval as the starting point, complete the setting of the path points of the interval, traverse all the intervals, and construct an inspection undirected graph based on all the path points.

[0039] Specifically, the turning points obtained in step d are used to divide the skeleton map into multiple intervals. For each interval, if the distance from one end to the other end exceeds 1.5u pixels, a path point is set every u pixels (in practice, u=20, which can be adjusted according to actual conditions), and the distance to the other end is continuously determined. If the distance is less than 1.5u pixels, no additional path point is set. Taking the segmented line segment 14 as an example, the generated effect diagram is as follows Figure 7 As shown, Figure 7 is the path point map provided by the present invention. The generated path points and turning points are added to the path point set V, where n is the number of path points: ; in, Represents the first n waypoints.

[0040] Through the above steps, the present invention can obtain all the path points needed in the robot inspection process. Combining these path points and the skeleton image can generate the inspection undirected graph G 无 =( V , E ),in, is the set of path points in the undirected graph, that is, the set of vertices in the undirected graph, is an undirected graph edge set. For any path point, if there are other path points directly connected to it, an undirected edge is added between the path point and the other path points.

[0041] (2) With the goal of ensuring that the inspection robot only turns at turning points during the inspection process, the inspection undirected graph (referred to as the undirected graph) is transformed into the inspection directed graph (referred to as the directed graph) G 有 =( , ),in, is the set of directed path points in the directed graph, that is, the set of directed graph vertices, is the set of edges in a directed graph.

[0042] If the path planning is done based on the undirected graph generated above, there will be a risk that the inspection robot will turn around in a narrow place and hit the equipment. To solve this problem, the present invention converts the inspection undirected graph obtained above into an inspection directed graph through a specific conversion logic. Figure 8 As shown, Figure 8 This is a comparison diagram of the effects of path planning using directed graphs and undirected graphs provided by the present invention. The black lines are the equipment boundaries, the green lines in the middle are the patrol lines on the skeleton map, and the inspection order is ①→②→③. If the previous undirected graph is used for planning, it will turn at point ② and easily touch the equipment on the black boundary. After the conversion of the present invention, the inspection robot will only turn at wide turning points, thereby avoiding the risk of collision and damage to equipment.

[0043] To achieve the above-described effect, the method of converting the inspection undirected graph into the inspection directed graph is as follows: (a) Find the direction angle of the path point in the inspection undirected graph: The direction angle of the path point in the inspection undirected graph is obtained by interval. The method for obtaining the direction angle of the path point in any interval of the inspection undirected graph is: starting from the path point at one end of any interval, all the path points are traversed in sequence to obtain the direction angle of each path point. Specifically, the direction angle of the path point is obtained in two cases: The first case: find the direction angle of each path point from the path point at one end of the starting point to the second to last path point, specifically: determine the connecting line between the currently traversed path point and the next path point to be traversed, and use the angle between the connecting line and the positive direction of the horizontal coordinate axis as the direction angle of the currently traversed path point; the range of this angle is [-90°, 90°); where the horizontal coordinate axis is the X-axis of the pre-established standard horizontal rectangular coordinate system, see Fig. 9 in the settings.

[0044] The second case: finding the direction angle of the last traversed path point, specifically: taking the direction angle of the second to last traversed path point as the direction angle of the last traversed path point.

[0045] (b) First determine the directed path points in the inspection directed graph: Specifically, according to the position and direction angle of any path point in the inspection undirected graph, two directed path points corresponding to any path point in the inspection directed graph are generated; Among them, the positions of the corresponding two directed path points are the same as the position of any path point, the direction of one of the two directed path points is the direction angle of any path point, ranging from [-90°, 90°), and the direction of the other directed path point is the direction angle of any path point plus 180 degrees, ranging from [90°, 270°).

[0046] (c) In the above step (b), only the directed path points in the inspection directed graph are determined. Next, the edges in the inspection directed graph are set according to the directed path points in the inspection directed graph to form the final inspection directed graph: The edges in the inspection directed graph are determined by intervals. The method for determining the edges in any interval of the inspection directed graph is as follows: Traverse all two adjacent directed path points in any interval. When the directions of the two adjacent directed path points are in the range of [-90°, 90°), add an edge between the two adjacent directed path points, and the direction of the edge points to the positive semi-axis side of the horizontal coordinate axis, that is, the direction of the edge is also in the range of [-90°, 90°); when the directions of the two adjacent directed path points are in the range of [90°, 270°), add an edge between the two adjacent directed path points, and the direction of the edge points to the negative semi-axis side of the horizontal coordinate axis, that is, the direction of the edge is also in the range of [90°, 270°).

[0047] According to the above transformation, the direction of the inspection robot can be changed only at the turning point and the boundary through the transformation of the directed path point, such as Fig. 9 As shown, Fig. 9 It is a comparison diagram of the undirected graph path diagram and the directed graph path diagram provided by the present invention. Fig. 9 The present invention uses red dotted lines to present two paths with different directions, but in fact the positions of these two red dotted lines are the same, both at the position of the middle black solid line.

[0048] The number of directed path points obtained by conversion is calculated as follows, where n is the number of path points in the undirected graph, and m is the number of turning points. is the set of turning points, is the number of directed graph vertices (i.e. directed path points): ; Through the processing of this part, the present invention generates a patrol directed graph, which avoids the situation where the patrol robot turns in a narrow place and damages the equipment. Subsequent path planning and other operations can be achieved based on this directed graph.

[0049] (3) On the inspection directed graph, the directed path point with the shortest distance to each inspection point and the direction in the same preset interval is used as a mapping point. The inspection points in the present invention are pre-planned and set, and include the inspection starting point. The specific setting method is as follows: The positions of the inspection points are fixed and unchanged. In the early stage, the staff manually remotely controls the inspection robot to reach the appropriate position and adjusts the angle according to the requirements of the equipment identification task. The coordinates of the inspection robot and the angle value of its direction at this time are recorded (the angle range is [-90°, 270°) as the parameters of the inspection point.

[0050] The present invention ultimately aims to achieve path planning between inspection points, but the inspection point positions are not consistent with the positions of the directed path points generated above, so the present invention needs to store the inspection starting point of the inspection robot and the position information of all other inspection points into a set P. The structure of the set P is as follows, where: is the location of the inspection starting point (here it can be a charging pile), and r is the number of other inspection points:

[0051] Due to the limitation of the A* algorithm, only directed path points in the directed graph can be planned. To realize the planning of inspection points, the present invention needs to find the corresponding mapping point of each inspection point in the directed graph, and then the inspection point planning can be realized by planning the mapping point. The method of determining the mapping point of any inspection point on the inspection directed graph includes but is not limited to the following steps: Step a: Calculate the distance between any inspection point and each directed path point on the inspection directed graph, and determine the directed path point closest to any inspection point.

[0052] The distance may be a Euclidean distance.

[0053] Step b: Determine, from the nearest directed path points, a directed path point that is in the same preset interval as the direction of any of the inspection points as a mapping point of any of the inspection points.

[0054] The preset interval is [-90°, 90°) or [90°, 270°), [-90°, 90°) represents the positive interval of the X-axis, and [90°, 270°) represents the negative interval of the X-axis.

[0055] For example, the direction of the inspection point is 110°, and the directions of the two closest directed path points determined in step a are 30° and 210° respectively; since the directions 110° and 210° of the inspection point are both in the negative range of the X-axis, the directed path point with a direction of 210° is determined as the mapping point of the inspection point.

[0056] Among them, an optional determination method is as follows: Subtract 90° from the direction of the inspection point. If the result is greater than 0, a directed path point with a direction between [90, 270) is selected as the mapping point. If the result is less than 0, a directed path point with a direction between [-90, 90) is selected as the mapping point.

[0057] Store the mapping points obtained above in the corresponding point set The structure is as follows, where It can be the mapping point corresponding to the charging pile location (inspection starting point), rThe number of other inspection points:

[0058] Using the generated corresponding point set The overall planning idea of ​​the invention is as follows: and There are two inspection points to be planned. and The directed path points (i.e. mapping points) corresponding to the two inspection points are:

[0059] (4) By analyzing the topological relationship between the mapping points, the inspection point order of all inspection points of the inspection robot is determined.

[0060] Optionally, the mapping point corresponding to the inspection starting point is taken as the starting point, and path planning is performed with the mapping points corresponding to other inspection points whose sorting has not been determined, so as to determine the next inspection point with the shortest path; the mapping point corresponding to the next inspection point with the shortest path is taken as the starting point, and path planning is performed with the mapping points corresponding to the inspection points whose sorting has not been determined, so as to determine the next inspection point with the shortest path, and so on, until the sorting of all inspection points is completed.

[0061] Because the inspection points are determined, the present invention can determine the inspection order of all inspection points in advance by using the greedy optimization algorithm, so that the trouble of re-arranging the inspection points every time for all subsequent inspection tasks can be avoided, which greatly saves the time of each inspection. The specific sorting method is as follows: The charging pile First add it to the sorting queue Q, then select the charging pile First, use the planning ideas mentioned in (3) to plan the path with all other inspection points. The specific steps of path planning are as follows: For two inspection points and , first in the corresponding set Find the corresponding directed path points in and Then, in the directed path point set of the directed graph Find these two path points and use the A* algorithm to plan the shortest path. Compare all planned paths, select the shortest path and store it in the two-dimensional array Array, and add the inspection points corresponding to the shortest path to the sorting queue Q. Repeat the operation until all inspection points are in Q, and the sorting queue Q is the final inspection point sorting.

[0062] It is particularly noted that, because the inspection points in the present invention are directional, the direction requirements must be taken into consideration when planning the route. Fig.10is a schematic diagram of path planning with directions provided by the present invention, such as Fig.10 As shown in the figure, if there is no direction requirement, the inspection order is ①→②→③→④→⑤, and when there is a direction requirement, the inspection order becomes ①→④→⑤→③→②. The red arrow is the path. At this time, the inspection point closest to inspection point 1 becomes inspection point 4.

[0063] (5) In response to the issued current inspection task, the inspection order of the inspection points to be inspected included in the current inspection task is determined according to the inspection point ranking, and the path planning of the current inspection task is realized by using the inspection directed graph.

[0064] Optionally, the inspection order of the points to be inspected included in the current inspection task is determined according to the relative positions of the points to be inspected in the inspection point sorting, and the inspection order of the mapping points corresponding to the points to be inspected is determined based on the inspection order of the points to be inspected; the path point closest to the current position of the inspection robot is taken as the starting point, and the shortest path between each pair of adjacent mapping points is determined according to the inspection order of the mapping points to realize the path planning of the current inspection task.

[0065] Fig.11 It is a logical block diagram of the path planning of the current inspection task provided by the present invention, such as Fig.11 The specific processing ideas are as follows: According to the user's instructions, obtain the current position and direction (as the starting point of the inspection) of all inspection points and inspection robots (which can be quadruped inspection robots) of the current inspection task. Find the corresponding mapping point in the vertex set of the directed graph based on the current position and direction; sort all the inspection points of the current task according to the sorting queue Q obtained in (4) to obtain an ordered queue of inspection points .exist Find each corresponding mapping point in , then take the mapping point corresponding to the current position as the starting point, and follow The order in the queue uses the A* algorithm to regularize the path between two points one by one, and finally obtains the planned path.

[0066] Of course, the present invention can also directly add the nearest path point to Queue, and directly carry out subsequent path planning, which will not be described here.

[0067] (6) Implement automatic door opening and closing in planning There may be doors in the inspection path. In order to ensure the safety of the inspection robot (which may be a four-legged inspection robot) and the smooth completion of the inspection task, the present invention enables the inspection robot to automatically identify the door and complete the action of opening and closing the door. Fig.12 Schematic diagram of the danger zone formed by the door rotation provided by the present invention, such as Fig.12As shown, the present invention pre-stores the door location information and the door radius, and draws a circle with the door axis as the vertex and the door radius. This area is a dangerous area, and no path points can be set in the dangerous area. Fig.13 The present invention provides a schematic diagram of the door opening and closing during the inspection process, such as Fig.13 As shown, the inspection robot must send the command to open or close the door outside this dangerous area.

[0068] Optionally, during the door entry process: the inspection robot is controlled to stop moving before entering the dangerous area, and after determining that the door is completely opened, the inspection robot is controlled to pass through the dangerous area and enter the door; during the door exit process: when it is determined that the inspection robot has left the dangerous area, a door closing command is sent.

[0069] In addition, in the scenario where the inspection area is a substation, there is also the case of a rat guard, which can be set on the door and can move up and down along the door frame. Specifically, in path planning, if there is a door, it will stop moving forward before entering the dangerous area and send a command to move the rat guard up to a fixed position, and at the same time send a command to open the door. When the inspection robot recognizes that the door is fully open, it will pass through the door. When leaving the dangerous area, it will stop moving forward and send a command to move the rat guard down to the bottom, and at the same time set to send a command to close the door. After the door is closed, it will continue to move forward. This can ensure that the inspection robot passes through the door smoothly and safely.

[0070] (7) Fast planning strategy: When the shortest path between adjacent mapping points has been predetermined, the predetermined shortest path is directly adopted; when the shortest path between adjacent mapping points has not been predetermined, the preset path optimization algorithm is used to generate the shortest path.

[0071] In the actual planning process, the present invention introduces the idea of ​​reuse, because in the waiting inspection queue There may be adjacent inspection points in the two-dimensional array that are also adjacent in the two-dimensional array, that is, the path between the two inspection points has been generated before. In view of this situation, for the planning of these two points, the present invention can directly copy the corresponding path in the two-dimensional array, and for other inspection points, the path planning can be carried out normally, such as Fig.14 As shown, Fig.14 It is a schematic diagram of the idea of ​​applying reuse in the path planning provided by the present invention.

[0072] The following is a brief description of the operation process of the inspection robot.

[0073] Fig.15 It is the early stage work flow diagram provided by the present invention, such as Fig.15As shown, go to the inspection scene, scan with radar and generate a SLAM map, and upload it to the cloud server database; perform a series of processing on the map to obtain an undirected graph; use the inspection robot to mark points, and upload the obtained inspection point information to the cloud server database; upload the generated turning points and path points (i.e., patrol points) to the cloud server database for storage.

[0074] Fig.16 It is the overall operation flow chart provided by the present invention, such as Fig.16 As shown, the user sends an inspection request to the server through a specific IP and port. After receiving the command, the server sends a request to the robot dog (i.e., a four-legged inspection robot) to obtain its current location. The server combines the user's request and the current location of the four-legged inspection robot, reads the information in the cloud service database, and performs path planning. The planned path is returned to the client, and the planned path instructions are also sent to the four-legged inspection robot. After receiving the instructions, the four-legged inspection robot executes the task step by step.

[0075] In summary, the path planning method for the inspection robot provided by the present invention has the following advantages: (1) The present invention proposes the concept of turning points (intersections), and allows the inspection robot to turn only at intersections during planning, thus avoiding damage to equipment and the inspection robot itself caused by turning in narrow places.

[0076] (2) The present invention provides a method for automatically generating path points based on turning points, which solves the problem of low efficiency and low accuracy of traditional manual marking, and can generate path points of different densities by changing variables.

[0077] (3) In order to meet the inspection requirements, the present invention proposes a new criterion for converting an undirected graph into a directed graph.

[0078] (4) To solve the problem that A* cannot be used directly for planning between two inspection points, the present invention proposes a new planning idea. The planning of the inspection points can be indirectly achieved by finding the directed path points (mapping points) corresponding to the inspection points and planning between the path points.

[0079] (5) In the traditional path planning, the planning of each inspection point needs to be calculated with all the remaining points to find the shortest path, and the time complexity is In order to improve planning efficiency, the present invention proposes to sort all inspection points in advance, so that the subsequent inspection tasks can be sorted first and then path planning can be performed, thus achieving a time complexity of It has dropped to .

[0080] (6) In order to achieve smooth inspection, the present invention sets instructions to enable the inspection robot to automatically open and close the door.

[0081] (7) To further improve planning efficiency, the present invention utilizes the idea of ​​reuse and can directly copy the previously stored shortest path in certain circumstances.

[0082] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the path planning method of the inspection robot provided in the above-mentioned embodiments. On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the path planning method for the inspection robot provided in the above-mentioned embodiments.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A path planning method for an inspection robot, characterized in that: include: Obtain a skeleton map of the inspection area, determine the turning point of the inspection robot on the skeleton map, and divide the skeleton map into multiple sections according to the turning points; When the length of any interval is greater than a preset length, path points are set at preset intervals with one end point of the interval as a starting point, the setting of the path points of the interval is completed, and all intervals are traversed to construct an inspection undirected graph according to all path points; With the goal of making the inspection robot turn only at turning points during the inspection process, the inspection undirected graph is transformed into an inspection directed graph. On the inspection directed graph, the directed path point corresponding to each inspection point with the shortest distance and the same direction in the same preset interval is used as a mapping point, and the inspection point order of all inspection points of the inspection robot is determined by analyzing the topological relationship between the mapping points; wherein the inspection points are pre-planned and set, and include the inspection starting point; In response to the issued current inspection task, the inspection order of the inspection points to be inspected included in the current inspection task is determined according to the inspection point ranking, and the path planning of the current inspection task is realized by using the inspection directed graph; Among them, the inspection robot is aimed at turning only at the turning point during the inspection process, and the inspection undirected graph is transformed into the inspection directed graph, which specifically includes: Starting from the path point at one end of any interval in the inspection undirected graph, all the path points are traversed in sequence to obtain the direction angle of each path point. There are two specific cases for obtaining the direction angle of the path point: The first case: find the direction angle of each path point from the path point at one end as the starting point to the second to last path point, specifically: determine the connecting line between the currently traversed path point and the next path point to be traversed, and use the angle between the connecting line and the positive direction of the horizontal coordinate axis as the direction angle of the currently traversed path point; wherein the range of the angle is [-90°, 90°); The second case: find the direction angle of the last traversed path point, specifically: take the direction angle of the second to last traversed path point as the direction angle of the last traversed path point; According to the position and direction angle of any path point in the inspection undirected graph, two directed path points corresponding to any path point in the inspection directed graph are generated; wherein the positions of the two corresponding directed path points are the same as the position of any path point, the direction of one of the two directed path points is the direction angle of any path point, and the direction of the other directed path point is the direction angle of any path point plus 180 degrees; Traverse all two adjacent directed path points in any interval. When the directions of the two adjacent directed path points are in the range of [-90°, 90°), add an edge between the two adjacent directed path points, and the direction of the edge points to the positive semi-axis side of the horizontal coordinate axis; when the directions of the two adjacent directed path points are in the range of [90°, 270°), add an edge between the two adjacent directed path points, and the direction of the edge points to the negative semi-axis side of the horizontal coordinate axis; Among them, by analyzing the topological relationship between the mapping points, the inspection point order of all inspection points of the inspection robot is determined, including: The mapping point corresponding to the inspection starting point is used as the starting point, and the path planning is performed with the mapping points corresponding to other inspection points whose order has not been determined, so as to determine the next inspection point with the shortest path; Taking the mapping point corresponding to the next inspection point with the shortest path as the starting point, path planning is performed with the mapping points corresponding to the inspection points whose sorting is not determined, and the next inspection point with the shortest path is determined, and so on, until all the inspection points are sorted.

2. The inspection robot path planning method according to claim 1, characterized in that: Get a skeleton map of the inspection area, including: Convert the SLAM map of the inspection area into a binary black-and-white image, determine the maximum connected area of ​​the white part in the black-and-white image, and generate a safe area map; wherein the maximum connected area of ​​the white part is the feasible area of ​​the inspection robot; A skeleton map is extracted from the safe area map.

3. The path planning method of the inspection robot according to claim 1, characterized in that: Determine the turning point of the inspection robot on the skeleton map, including: Taking any pixel point on the skeleton map as the center and a first preset number of pixels as the radius, draw a circle and count the number of white boundary pixels of the circle; When the number of white boundary pixels is greater than a second preset number, any pixel point is used as a turning point; Traverse all the pixel points on the skeleton map to determine all the turning points of the inspection robot on the skeleton map.

4. The path planning method for the inspection robot according to claim 1, characterized in that: The method for determining the mapping point of any inspection point on the inspection directed graph includes: Calculate the distance between any inspection point and each directed path point on the inspection directed graph, and determine the directed path point closest to any inspection point; Determine, from the nearest directed path points, a directed path point in the same preset interval as the direction of any inspection point as the mapping point of any inspection point; wherein the preset interval is [-90°, 90°) or [90°, 270°); The direction of any inspection point is pre-planned and set.

5. The path planning method for the inspection robot according to claim 4, characterized in that: According to the inspection point ranking, the inspection order of the inspection points to be inspected included in the current inspection task is determined, and the inspection directed graph is used to implement the path planning of the current inspection task, including: Determine the inspection order of the points to be inspected included in the current inspection task according to the relative position of the points to be inspected in the inspection point sequence, and determine the inspection order of the mapping points corresponding to the points to be inspected based on the inspection order of the points to be inspected; The path point closest to the current position of the inspection robot is taken as the starting point, and the shortest path between each pair of adjacent mapping points is determined according to the inspection order of the mapping points to realize the path planning of the current inspection task.

6. The path planning method for the inspection robot according to claim 1, characterized in that: When there are in-and-out doors in the path planning, it also includes: Door entry process: Control the inspection robot to stop before entering the dangerous area. After confirming that the door is open, control the inspection robot to pass through the dangerous area and enter the door. Exit process: After the inspection robot is determined to have left the dangerous area, a door closing command is sent; The danger zone is determined based on the location information of the door and the rotation radius of the door, and no path points are set in the danger zone.

7. The path planning method for the inspection robot according to claim 5, characterized in that: Determine the shortest path between each pair of adjacent mapped points, including: When the shortest path between adjacent mapping points has been predetermined, the predetermined shortest path is directly adopted; When the shortest path between adjacent mapping points is not predetermined, a preset path optimization algorithm is used to generate the shortest path.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the path planning method of the inspection robot as claimed in any one of claims 1 to 7 are implemented.

Images