Power inspection robot path planning method and system

Abstract

The invention provides an electric power inspection robot path planning method and system, and relates to the technical field of path planning. The method comprises the following steps: carrying out grid processing on an inspection environment; determining a search neighborhood range of the current point according to the coordinates of the current point and the terminal point; according to collision detection and a path cost penalty function, calculating the total cost of the to-be-searched nodes in the current point search neighborhood, taking the node with the minimum total cost as the next node, and taking the next node as the current point to continue path finding; the path cost penalty function is used for representing a position relation between a connecting line of a current point and a next node and an obstacle area; and obtaining an initial path based on all the child nodes, and smoothing the initial path by using an improved third-order Bezier curve to obtain an optimal path. According to the improved algorithm, the path length can be effectively reduced, the turning frequency is reduced, and the safety of the path is improved, so that the inspection path with higher quality is obtained.

Description

Technical Field

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

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] Power inspection robots can monitor power equipment in all directions, quickly locate anomalies, improve efficiency and avoid safety risks, which is crucial to power work. Power inspection robots can only complete tasks safely and efficiently if they move along reasonable routes, which requires path planning to achieve.

[0004] A * As a classic heuristic algorithm, A is widely used in path planning problems. It determines the path nodes by calculating the cost function, and searches continuously from the starting point to finally reach the task point. It is an effective method to obtain the shortest path in a static environment. However, due to the limitations of the algorithm principle, A * The paths planned by the algorithm have problems such as poor safety, low accuracy, many turning points and insufficient smoothness, which will reduce inspection efficiency and increase inspection costs. Therefore, it is necessary to improve the algorithm, optimize the path length and smoothness, and improve the path safety to meet the needs of inspection tasks. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a path planning method and system for a power inspection robot, constructs a path cost penalty function based on collision detection, and introduces the path cost penalty function and an improved third-order Bezier curve into A * The algorithm is improved * The algorithm effectively reduces the number of turns, shortens the path length, and improves path safety and smoothness, making the path safer and more reliable.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a path planning method for a power inspection robot, comprising:

[0008] Perform grid processing on the inspection environment;

[0009] According to the coordinates of the current point and the end point, determine the search neighborhood range of the current point;

[0010] The total cost of the nodes to be searched in the search neighborhood of the current point is calculated according to the collision detection and path cost penalty function, the node with the smallest total cost is used as the next node, and the next node is used as the current point to continue the path search; the path cost penalty function is used to represent the positional relationship between the connection line between the current point and the next node and the obstacle area;

[0011] The initial path is obtained based on all child nodes, and the initial path is smoothed using an improved third-order Bezier curve to obtain the optimal path.

[0012] Preferably, the grid processing comprises:

[0013] The grid size is set according to the inspection environment size and inspection accuracy requirements to obtain a number of grids;

[0014] Assign different status values ​​to the obstacle areas and non-obstacle areas in the inspection environment and construct a grid map;

[0015] Create an open table and a closed table, store the obstacle points of the obstacle area in the closed table, and store the starting points in the open table.

[0016] Preferably, determining the search neighborhood range of the current point according to the coordinates of the current point and the end point specifically includes: taking the current point as the center, the number of variable layers of the search neighborhood is m;

[0017] m=max{|x n -x t |,|y n -y t |}

[0018] Among them, x n ,y n is the horizontal and vertical coordinates of the current point, x t and t are the horizontal and vertical coordinates of the end point; as the pathfinding proceeds, the coordinates of the end point are fixed, and the coordinates of the current point change. The search neighborhood range of the current point is dynamically adjusted according to the coordinates of the current point and the end point.

[0019] Preferably, the calculation of the total cost of the nodes to be searched in the search neighborhood of the current point based on the collision detection and path cost penalty function also includes:

[0020] If the node to be searched already exists in the closed table, skip this node and continue searching for other nodes; if there is no information about this node in both the open table and the closed table, calculate its total cost and add it to the open table; if this node already exists in the open table, recalculate the total cost of the node and compare its newly calculated total cost with the original value in the open table. If the newly calculated total cost is smaller, update the total cost value in the table.

[0021] Preferably, the collision detection is specifically:

[0022] The current point is taken as the parent node, the nodes to be searched in the search neighborhood of the current point are taken as child nodes, and the parent node and the child node are connected to obtain multiple sub-paths to be selected;

[0023] The two diagonal lines of the obstacle area are used as virtual collision detection lines;

[0024] The selected sub-path and the virtual collision detection line are subjected to collinearity and intersection detection; the collinearity detection is as follows:

[0025]

[0026] in, Represents the vector from the parent node to the child node, expressed as the difference between the coordinate values ​​of the two points, and × represents the cross product of the vector; represents any virtual collision detection line; when the above formula is true, it means that the sub-path to be selected is colinear with any virtual collision detection line;

[0027] If the sub-path to be selected is collinear with any virtual collision detection line, then only the other virtual collision detection line needs to be intersected; if the sub-path to be selected is not collinear with either virtual collision detection line, then both virtual collision detection lines need to be intersected:

[0028]

[0029] Among them, x A 、x B 、x C 、x D ,y A ,y B ,y C ,y D Represent the horizontal and vertical coordinate values ​​of the parent node, child node and the two endpoints of the virtual collision detection line respectively; solve p and q by the above formula. If both are in the range of [0,1], the selected sub-path intersects with the virtual collision detection line;

[0030] The sub-nodes of the candidate sub-paths that do not produce intersections are taken as candidate nodes, and the total cost of the candidate nodes is calculated according to the cost penalty function.

[0031] Preferably, the path cost penalty function is specifically:

[0032]

[0033] Among them, λ is the penalty coefficient, d1 is the distance from the obstacle vertex to the path segment, and d0 is the safety threat radius of the obstacle vertex. The improved path total cost function is:

[0034] F(n)=G(n)+H(n)+γ(n)

[0035] Among them, F(n) is the estimated total path cost; G(n) is the actual cost from the starting point to node n; H(n) is the estimated cost from node n to the end point, and γ(n) is the path cost penalty value.

[0036] Preferably, the initial path is obtained based on all child nodes, and the initial path is smoothed using an improved third-order Bezier curve to obtain the optimal path, specifically:

[0037] Get three consecutive points W1W2W3 on the initial path and calculate eight control points B0-B3 and E0-E3. The calculation formula is as follows:

[0038]

[0039] Where d is the distance between B0W2 and E0W2, u1u2u3 are the unit vectors in the directions of W2B0, W2E0 and B2E2 respectively; c1 = 7.2364, c2 = 0.5798, c3 = (c2 + 4) / (c1 + 6), β = γ / 2, k max is the maximum curvature; h = c3d, g = c2c3d, k = (6c3cosβ) / (c2+4)d. Finally, the eight control points are connected to form a smooth path consisting of two groups of trajectories, W0-B1-B2-B3 and W3-E0-E1-E3. The curvatures are equal at the connection points B3 or E3, so that the inspection path can smoothly transition at the connection points and obtain the optimal path.

[0040] In a second aspect, the present invention provides a path planning system for a power inspection robot, comprising:

[0041] Initialization module, used to perform grid processing on the inspection environment;

[0042] The neighborhood dynamic adjustment module is used to determine the search neighborhood range of the current point according to the coordinates of the current point and the end point;

[0043] A collision detection module is used to calculate the total cost of the nodes to be searched in the search neighborhood of the current point according to the collision detection and path cost penalty function, and take the node with the smallest total cost as the next node, and continue the path search with the next node as the current point; the path cost penalty function is used to represent the positional relationship between the connection line between the current point and the next node and the obstacle area;

[0044] The optimal path acquisition module is used to obtain an initial path based on all child nodes, and use an improved third-order Bezier curve to smooth the initial path to obtain the optimal path.

[0045] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the path planning method for a power inspection robot described in the first aspect.

[0046] In a fourth aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the path planning method for a power inspection robot described in the first aspect are implemented.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. The present invention adjusts the search neighborhood in real time according to the current node position, removes the restriction on the search direction of the traditional eight-neighborhood search method, and simultaneously selects the safe sub-node with the lowest cost in the neighborhood through collision detection and path cost penalty function, thereby solving the problem of many turns and easy collision accidents, and can effectively shorten the path length.

[0049] 2. The present invention proposes a method for dynamically adjusting the search neighborhood. By dynamically adjusting the size of the search neighborhood, that is, expanding from the traditional fixed eight-neighborhood to a dynamic range (4m*(m+1) neighborhood) based on the position relationship between the current search point and the end point, the number of potential path points that can be considered in each search is greatly increased. This expansion not only covers more possible path directions, but also reduces the frequent turns caused by the traditional eight-neighborhood limitation, thereby improving the overall efficiency of the search.

[0050] 3. The present invention uses a third-order Bezier curve with continuous curvature to smooth the path corners, which can ensure a smooth transition of the path at the connection point, better fit the kinematic constraints of the inspection robot, avoid jittering of the inspection robot during turning, improve the stability of turning, and enable path planning to obtain a smoother and more reliable inspection path in inspection scenarios of different complexities.

[0051] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their description are used to explain the present invention but do not constitute a limitation of the present invention.

[0053] Figure 1 A main flow chart of a path planning method for an electric power inspection robot provided by an embodiment of the present invention;

[0054] Figure 2A schematic diagram of dynamically adjusting the search neighborhood provided by an embodiment of the present invention;

[0055] Figure 3 A schematic diagram of a collision detection screening node provided by an embodiment of the present invention;

[0056] Figure 4 A schematic diagram of a third-order Bezier curve with continuous curvature provided by an embodiment of the present invention;

[0057] Figure 5 A schematic diagram of the optimal path provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0059] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0060] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0061] Embodiment 1

[0062] like Figure 1 As shown, this embodiment discloses a path planning method for a power inspection robot, comprising the following steps:

[0063] S1: Perform grid processing on the inspection environment;

[0064] S2: Determine the search neighborhood range of the current point based on the coordinates of the current point and the end point;

[0065] S3: Calculate the total cost of the nodes to be searched in the search neighborhood of the current point according to the collision detection and path cost penalty function, take the node with the smallest total cost as the next node, and continue path finding with the next node as the current point; the path cost penalty function is used to represent the positional relationship between the connection line between the current point and the next node and the obstacle area;

[0066] S4: The initial path is obtained based on all child nodes, and the initial path is smoothed using an improved third-order Bezier curve to obtain the optimal path.

[0067] Next, combine Figure 1 , a path planning method for an electric power inspection robot disclosed in this embodiment is described in detail.

[0068] S1: In the inspection environment, the grid method is used to divide the overall environment into small grids of the same size, and two different state information is given to the grids to distinguish the feasible area and the obstacle area in the inspection environment of the power inspection robot. The construction of the grid map includes the following steps:

[0069] S101: setting the grid size according to the inspection environment size and inspection accuracy requirements to obtain a number of grids;

[0070] S102: assigning different status values ​​to the obstacle areas and non-obstacle areas in the inspection environment to construct a grid map;

[0071] S103: Create an open table and a closed table, store the obstacle points of the obstacle area in the closed table, and store the starting points in the open table.

[0072] In this embodiment, the inspection environment is segmented by the grid method, and different state information is given to the grid, which can clearly distinguish the feasible area and the obstacle area, so as to construct an accurate grid map, thereby improving the accuracy of map construction and providing a reliable basis for subsequent path planning. Secondly, by creating an open table and a closed table, the problem of repeated search and falling into a local optimal solution is effectively avoided.

[0073] Start exploring from the starting point to the task point (end point), calculate the total cost of each child node in the real-time neighborhood of the current point, the total cost of each child node is the sum of the Euclidean distances from the node to the starting point and the target point, select the child node with the smallest total cost as the next node, and repeat the above process until you reach the task point and connect the points to form a path. Obtaining the path point includes the following steps:

[0074] (1) Create an Open table and a Close table, store the coordinate information of the obstacle area in the task environment in the Close table, and put the starting point coordinates into the Open table;

[0075] (2) Check whether the open table is empty. If it is empty, no feasible path is found and the search ends. Otherwise, determine whether the current node is consistent with the task point. If they are consistent, backtrack to find the path and the program ends. Otherwise, based on the dynamic search neighborhood, select the node with the smallest total cost value in the open table as the current point and continue searching the neighborhood.

[0076] (3) If the point to be searched already exists in the Close table, skip this point and continue searching for other nodes; if there is no information about this node in either table, calculate the total cost of the point to be searched and add it to the Open table; if this point already exists in the Open table, compare its newly calculated total cost F with the original value in the Open table. If F is smaller, update the F value in the table (and update G and H values ​​at the same time, G is the actual cost from the starting point to node n, and H is the estimated cost from node n to the task point); and mark the current node (the point with the lowest cost) as the parent node for the next search;

[0077] (4) Return to step (2) and continue searching for the path.

[0078] In the search process, this embodiment dynamically calculates the total cost of each child node and selects the child node with the smallest total cost as the next node, thereby ensuring the optimality of the path. In addition, the situation where the point to be searched already exists in the closed table or the open table is also considered, thereby avoiding unnecessary calculations and improving the search efficiency.

[0079] S2 dynamically adjusts the search neighborhood. According to the horizontal and vertical coordinates of the current search point (parent node) and the task point, the expandable number of layers for each search is changed from 1 to m, and the traditional eight-neighborhood is expanded to 4m*(m+1) to increase the selectable child nodes, break the restrictions on the search direction, and reduce unnecessary turns. The child node with the smallest total cost value is locked through one traversal. The formula for the expandable number of layers m is:

[0080] m=max{|x n -x t |,|y n -y t |}

[0081] Among them, x n ,y n is the horizontal and vertical coordinates of the current search point, x t and t are the horizontal and vertical coordinates of the task point;

[0082] This embodiment can significantly reduce unnecessary turns and lock the child node with the smallest total cost value through one traversal, thereby greatly reducing the search space while ensuring the search accuracy. Figure 2As shown in (a), when searching for a path from S to T, the traditional algorithm can only search the eight adjacent grids, that is, the grid marked with number 1. The improved dynamic neighborhood search method proposed in this embodiment substitutes the horizontal and vertical coordinate values ​​of S and T into the above formula to obtain m=3, and the searchable neighborhood is expanded to 48. The grids marked with numbers 1-3 can all be reached directly, so that the search is no longer limited to 0 degree, 180 degree, plus or minus 45 degrees, plus or minus 90 degrees, and plus or minus 135 degrees. The possible movement directions are increased, the number of turns is reduced, the number of searches is effectively reduced, and unnecessary losses are avoided.

[0083] After the current node finds the next node based on the current neighborhood search range, the next node updates the neighborhood search range based on its coordinates and the coordinates of the end point. That is, as the search process proceeds, the current node continuously updates the neighborhood search range based on its coordinates and the coordinates of the end point, realizing dynamic adjustment of the search neighborhood. This gradually focused search method helps to reduce the search space while maintaining the accuracy of the search, thereby finding the best path from the starting point to the end point more efficiently.

[0084] like Figure 2 (b) shows the path obtained by searching from S to T before and after the neighborhood adjustment, where the dotted line is the traditional A * The solid line is the path solved by the algorithm's eight-neighborhood search method, and the solid line is the path solved after the neighborhood is dynamically adjusted in this embodiment. It can be seen from the figure that the path solved after the neighborhood adjustment has two fewer turns than the path solved by the traditional eight-neighborhood search method, and the path length is shorter, thereby reducing the number of turning decelerations and travel losses of the inspection robot, and improving the efficiency of executing inspection tasks.

[0085] In S3, collision detection is performed on the child nodes in the neighborhood and a cost penalty function is established to filter out the safe child nodes with the lowest cost.

[0086] After the search neighborhood is dynamically adjusted, two virtual collision detection lines CD and C1D1 are established by connecting the diagonal vertices of the obstacle as endpoints, and the path segment AB connected by the parent node (current node) and the child node (next node) is tested for collinearity and intersection with the virtual collision detection line. At the same time, a corresponding cost penalty function is established to impose a cost penalty on the path within the safety threat range of the obstacle vertex, and the safe child node with the lowest cost is selected. The collinearity detection formula is:

[0087]

[0088] in, Represents the vector from A to B, represented by the difference in the coordinates of the two points, × represents the cross product of the vectors, represents any virtual collision detection line; substitute the virtual collision detection line CD into the above formula. When the above formula is true, it means that AB and CD are collinear. Similarly, perform collinear detection and judgment on C1D1. If AB is collinear with one of the collision detection lines, only the other one needs to be intersected. If it is not collinear with both collision detection lines, both need to be intersected. The intersection detection formula is:

[0089]

[0090] Among them, x A 、x B 、x C 、x D ,y A ,y B ,y C ,y D They represent the horizontal and vertical coordinate values ​​of the current point, child node and the two endpoints of any virtual collision detection line respectively; substitute the virtual collision detection line CD into the above formula, and solve p and q through the above formula. If both are in the range of [0,1], AB and CD intersect. Similarly, perform intersection detection on C1D1.

[0091] The collision detection of this embodiment determines the relationship between the path segment and the obstacle through colinearity and intersection detection, and performs accurate safety assessment on the sub-nodes in the neighborhood, effectively avoiding collision between the path and the obstacle, and ensuring that the path is unobstructed.

[0092] The child nodes of the path segment that does not produce intersection points are taken as candidate nodes and the path cost penalty value is calculated. The cost penalty function formula is:

[0093]

[0094] Among them, λ is the penalty coefficient, d1 is the distance from the obstacle vertex to the path segment, and d0 is the safety threat radius of the obstacle vertex. The improved path total cost function is:

[0095] F(n)=G(n)+H(n)+γ(n)

[0096] Among them, F(n) is the estimated total path cost; G(n) is the actual cost from the starting point to node n; H(n) is the estimated cost from node n to the task point, γ(n) is the path cost penalty value, and execute the other steps in S2 in sequence until the task point is searched.

[0097] This embodiment introduces a cost penalty, which can accurately measure and avoid potential high-risk or high-cost paths, ensuring the economy and safety of the solution. In a complex environment, the cost penalty function can effectively distinguish the pros and cons of different paths and guide the algorithm to choose a more reasonable solution. In addition, the function is also flexible and can adjust the penalty intensity according to actual needs to achieve optimization goals at different levels.

[0098] like Figure 3 As shown in the figure, A is the parent node, B0-B3 are optional child nodes in the neighborhood, CD and C1D1 are virtual collision detection lines, d1 is the distance from the vertex to the path segment, the dotted circle with d0 as the radius is the obstacle vertex security threat range, AB2 and AB3 correspond to the two cases where the path segment and the collision detection line are collinear or not collinear, respectively. Through collinear detection and intersection detection, it can be obtained that traveling along the two path segments AB2 and AB3 will collide with the obstacle. Combined with the path cost penalty function, it can be seen that γ(n) of the path segment AB0 is 0, and γ(n) of AB1 is After substituting into the total path cost function, it can be seen that B0 is the safe subnode with the lowest total cost in the neighborhood.

[0099] In S4, a curvature-continuous third-order Bezier curve is used to smoothly optimize the path corners. Eight control points are calculated through three consecutive path points to form a path with continuous curvature and tangent directions at two connection points at the corner. The path corner smoothing optimization includes the following steps:

[0100] Given points P0, P1, ..., P n Define an n-th order Bezier curve as:

[0101]

[0102] Among them, P i is the coordinate value of the control point, B i,n (t) is the basis function, and the expression is as follows:

[0103]

[0104] Where n is the order, i is the control point subscript, and when n=3, the third-order Bezier curve is obtained:

[0105] P(t)=P0(1-t) 3 +3P1(1-t) 2 t+3P2(1-t)t 2 +P3t 3

[0106] An n-order Bezier curve has n+1 control points, such as Figure 3 The curve shown is a third-order Bezier curve generated by four control points P0-P3.

[0107] The eight control points B0-B3 and E0-E3 are calculated using the three consecutive points W1W2W3 on the original path. The calculation formula is as follows:

[0108]

[0109] Where d is the distance between B0W2 and E0W2, u1u2u3 are the unit vectors in the directions of W2B0, W2E0 and B2E2 respectively; c1 = 7.2364, c2 = 0.5798, c3 = (c2 + 4) / (c1 + 6), β = γ / 2, k max is the maximum curvature, c1 and c2 are control points that satisfy the curvature continuous third-order Bezier curve property; h = c3d, g = c2c3d, k = (6c3cosβ) / (c2+4)d, and finally connect the control points to obtain the complete path.

[0110] This embodiment uses a third-order Bezier curve to perform smooth optimization of path corners, which can significantly improve the smoothness and continuity of the path. By accurately calculating eight control points, it is ensured that a path with continuous curvature and tangent direction is formed at the corner, thereby avoiding abrupt turns in the path. This method is not only rigorous and reliable in mathematical theory, but also effective in practical applications. Using three continuous points on the original path, combined with a specific calculation formula, the coordinates of the control points can be efficiently calculated, and then a smooth third-order Bezier curve can be constructed. The third-order Bezier curve can reduce friction and resistance on the path and improve overall efficiency.

[0111] like Figure 4 The curve shown is a smooth trajectory with continuous curvature generated by smoothing the original path W1-W2-W3. It consists of two groups of trajectories, W0-B1-B2-B3 and W3-E0-E1-E3. The curvature is equal at the connection point E3 / B3, so that the inspection path can have a smooth transition at the connection point, avoiding jitter of the inspection robot when turning at sharp corners of the inspection path, and providing a guarantee for the inspection robot to successfully complete the task.

[0112] To further demonstrate the positive effects of the above embodiment, the present invention conducts simulation experiments based on the above technical solution in two map environments with different complexities, 20×20 and 35×35. Figure 5 As shown, S is the starting point, T is the task point, and the blue solid line is the traditional A * The path planned by the algorithm, the red dotted line is the path planned by the technical solution provided by the present invention, and the following Tables 1 and 2 are comparison tables of two groups of path simulation data.

[0113] Table 1 Comparison of simulation data under 20×20 environment

[0114]

[0115] Table 2 Comparison of simulation data under 35×35 environment

[0116]

[0117] From the data in Tables 1 and 2 and Figure 5 As can be seen from the path diagram, the present invention improves A * The algorithm is superior to the traditional A in terms of path length and number of turns. * The algorithm can effectively avoid collisions at the vertices of obstacles and make the path transition smoothly at turns, thus improving the quality and feasibility of the path.

[0118] This embodiment discloses a path planning method for an electric power inspection robot. First, the inspection environment is modeled, and the environmental space information is converted into a mathematical model that can be recognized by the algorithm using a grid method, and the corresponding information is transmitted to A * The algorithm is used for path planning; then the improved strategies of adjusting the search neighborhood, conducting collision detection and establishing a cost penalty function are proposed to improve the algorithm and solve the traditional A * The path searched by the algorithm has problems such as suboptimal path length, many turns and prone to collisions. The path is then smoothly optimized using a third-order Bezier curve with continuous curvature. Simulation verification shows that the improved algorithm can effectively reduce the number of turns, reduce the path length, improve the safety and smoothness of the path, and make the path safer and more reliable.

[0119] Embodiment 2

[0120] This embodiment provides a path planning system for an electric power inspection robot, including:

[0121] Initialization module, used to perform grid processing on the inspection environment;

[0122] The neighborhood dynamic adjustment module is used to determine the search neighborhood range of the current point according to the coordinates of the current point and the end point;

[0123] A collision detection module is used to calculate the total cost of the nodes to be searched in the search neighborhood of the current point according to the collision detection and path cost penalty function, and take the node with the smallest total cost as the next node, and continue the path search with the next node as the current point; the path cost penalty function is used to represent the positional relationship between the connection line between the current point and the next node and the obstacle area;

[0124] The optimal path acquisition module is used to obtain an initial path based on all child nodes, and use an improved third-order Bezier curve to smooth the initial path to obtain the optimal path.

[0125] In this embodiment, the initialization module performs rasterization processing on the inspection environment, which greatly simplifies the complexity of the environment and facilitates subsequent path planning. The neighborhood dynamic adjustment module dynamically determines the search neighborhood range according to the coordinates of the current point and the end point, effectively narrowing the search space and improving the search efficiency. The collision detection module ensures the safety and feasibility of the path by calculating the total cost of the node to be searched and selecting the node with the smallest total cost as the next node. Finally, the optimal path acquisition module uses an improved third-order Bezier curve to smooth the initial path, which not only makes the path smoother and more natural, but also reduces the energy consumption and vibration of the robot during movement, and improves the accuracy and stability of the inspection. Therefore, the system provided in this embodiment can significantly improve the path planning efficiency and inspection quality of the power inspection robot.

[0126] Embodiment 3

[0127] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the path planning method for a power inspection robot as described in the first embodiment above are implemented.

[0128] Embodiment 4

[0129] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the path planning method for a power inspection robot as described in the first embodiment above are implemented.

[0130] The steps or modules involved in the above embodiments 2 to 4 correspond to those in embodiment 1. For the specific implementation, please refer to the relevant description of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0131] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0132] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A path planning method for a power inspection robot, characterized in that: include: Perform grid processing on the inspection environment; According to the coordinates of the current point and the end point, determine the search neighborhood range of the current point; Calculate the total cost of the nodes to be searched in the search neighborhood of the current point based on collision detection and path cost penalty function, take the node with the smallest total cost as the next node, and continue pathfinding with the next node as the current point; The path cost penalty function is used to represent the positional relationship between the line connecting the current point and the next node and the obstacle area; The initial path is obtained based on all child nodes, and the initial path is smoothed using an improved third-order Bezier curve to obtain the optimal path.

2. A method for path planning of a power inspection robot according to claim 1, characterized in that: The grid processing includes: The grid size is set according to the inspection environment size and inspection accuracy requirements to obtain a number of grids; Assign different status values ​​to the obstacle areas and non-obstacle areas in the inspection environment and construct a grid map; Create an open table and a closed table, store the obstacle points of the obstacle area in the closed table, and store the starting points in the open table.

3. A path planning method for a power inspection robot as claimed in claim 1, characterized in that: The step of determining the search neighborhood range of the current point according to the coordinates of the current point and the end point specifically includes: taking the current point as the center, the number of variable layers of the search neighborhood is m; m=max{x n -x t |,|y n -y t |} Among them, x n ,y n is the horizontal and vertical coordinates of the current point, x t and t are the horizontal and vertical coordinates of the end point; as the pathfinding proceeds, the coordinates of the end point are fixed, and the coordinates of the current point change. The search neighborhood range of the current point is dynamically adjusted according to the coordinates of the current point and the end point.

4. A method for path planning of a power inspection robot as claimed in claim 2, characterized in that: The calculation of the total cost of the nodes to be searched in the search neighborhood of the current point based on the collision detection and path cost penalty function also includes: If the node to be searched already exists in the closed table, skip this node and continue searching for other nodes; if there is no information about this node in both the open table and the closed table, calculate its total cost and add it to the open table; if this node already exists in the open table, recalculate the total cost of the node and compare its newly calculated total cost with the original value in the open table. If the newly calculated total cost is smaller, update the total cost value in the table.

5. A method for path planning of a power inspection robot according to claim 1, characterized in that: The collision detection is specifically as follows: The current point is taken as the parent node, the nodes to be searched in the search neighborhood of the current point are taken as child nodes, and the parent node and the child node are connected to obtain multiple sub-paths to be selected; The two diagonal lines of the obstacle area are used as virtual collision detection lines; The selected sub-path and the virtual collision detection line are subjected to collinearity and intersection detection; the collinearity detection is as follows: in, Represents the vector from the parent node to the child node, expressed as the difference between the coordinate values ​​of the two points, and × represents the cross product of the vector; represents any virtual collision detection line; when the above formula is true, it means that the sub-path to be selected is colinear with any virtual collision detection line; If the sub-path to be selected is collinear with any virtual collision detection line, then only the other virtual collision detection line needs to be intersected; if the sub-path to be selected is not collinear with either virtual collision detection line, then both virtual collision detection lines need to be intersected: Among them, x A 、x B 、x C 、x D ,y A ,y B ,y C ,y D Represents the parent node and child node respectively And the horizontal and vertical coordinate values ​​of the two endpoints of the virtual collision detection line; solve p and q through the above formula, if both are in the range of [0, 1], the sub-path to be selected intersects with the virtual collision detection line; The sub-nodes of the candidate sub-paths that do not produce intersections are taken as candidate nodes, and the total cost of the candidate nodes is calculated according to the cost penalty function.

6. A method for path planning of a power inspection robot as claimed in claim 4, characterized in that: The path cost penalty function is specifically: Among them, λ is the penalty coefficient, d1 is the distance from the obstacle vertex to the path segment, and d0 is the safety threat radius of the obstacle vertex. The improved path total cost function is: F(n)=G(n)+H(n)+γ(n) Among them, F(n) is the estimated total path cost; G(n) is the actual cost from the starting point to node n; H(n) is the estimated cost from node n to the end point, and γ(n) is the path cost penalty value.

7. A method for path planning of a power inspection robot according to claim 1, characterized in that: The initial path is obtained based on all child nodes, and the initial path is smoothed using an improved third-order Bezier curve to obtain the optimal path, specifically: Get three consecutive points W1W2W3 on the initial path and calculate eight control points B0-B3 and E0-E3. The calculation formula is as follows: Where d is the distance between B0W2 and E0W2, u1u2u3 are the unit vectors in the directions of W2B0, W2E0 and B2E2 respectively; c1=7.2364, c2=0.5798c3=(c2+4) / (c1+6)β=γ / 2, k max is the maximum curvature; h = c3d, g = c2c3d, k = (6c3cosβ) / (c2+4)d, and finally the eight control points are connected to form a smooth path consisting of two groups of trajectories W0-B1-B2-B3 and W3-E0-E1-E3. The curvatures are equal at the connection points B3 or E3, so that the inspection path can smoothly transition at the connection points and obtain the optimal path.

8. A path planning system for an electric power inspection robot, characterized in that: include: Initialization module, used to perform grid processing on the inspection environment; The neighborhood dynamic adjustment module is used to determine the search neighborhood range of the current point according to the coordinates of the current point and the end point; The collision detection module is used to calculate the total cost of the nodes to be searched in the search neighborhood of the current point based on the collision detection and path cost penalty function, and take the node with the smallest total cost as the next node, and continue the path search with the next node as the current point; The path cost penalty function is used to represent the positional relationship between the line connecting the current point and the next node and the obstacle area; The optimal path acquisition module is used to obtain an initial path based on all child nodes, and use an improved third-order Bezier curve to smooth the initial path to obtain the optimal path.

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

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the path planning method for a power inspection robot as described in any one of claims 1-7 are implemented.

Images