CN117308969B - Heuristic three-dimensional route planning method for electric power pole tower for quickly exploring random tree - Google Patents
Heuristic three-dimensional route planning method for electric power pole tower for quickly exploring random tree Download PDFInfo
- Publication number
- CN117308969B CN117308969B CN202311259360.9A CN202311259360A CN117308969B CN 117308969 B CN117308969 B CN 117308969B CN 202311259360 A CN202311259360 A CN 202311259360A CN 117308969 B CN117308969 B CN 117308969B
- Authority
- CN
- China
- Prior art keywords
- point
- dimensional
- target
- route
- heuristic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 96
- 238000007689 inspection Methods 0.000 claims abstract description 12
- 238000001816 cooling Methods 0.000 claims description 18
- 238000002922 simulated annealing Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 238000013461 design Methods 0.000 abstract description 3
- 238000002474 experimental method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000000137 annealing Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/24—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for cosmonautical navigation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Astronomy & Astrophysics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Automation & Control Theory (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a three-dimensional route planning method for a power tower of a heuristic rapid exploration random tree, which relates to the technical field of power tower inspection, and comprises the following steps: s1, acquiring a three-dimensional point cloud of an electric power tower through an unmanned aerial vehicle, and performing three-dimensional modeling; s2, acquiring the space position of a key component of the power tower in advance, and taking the space position as an intermediate target point of the unmanned aerial vehicle, which needs to hover and shoot in the course of the route, so as to form a route planning constraint point; s3, adopting a heuristic three-dimensional route planning algorithm based on RRT to plan a three-dimensional route of the power tower of the unmanned aerial vehicle, adopting the heuristic three-dimensional route planning method of the power tower for quickly exploring the random tree in the steps, integrating a heuristic search strategy to perform cost function design, accelerating algorithm convergence, ensuring optimal route, and compared with the improved RRT (remote-Bias) algorithm, aiming at the non-collision three-dimensional route planned by the power tower, the method has stable and reliable shortest path length and can meet the application requirement of intelligent monitoring.
Description
Technical Field
The invention relates to the technical field of power tower inspection, in particular to a three-dimensional route planning method for a power tower, which is used for heuristically and rapidly exploring a random tree.
Background
The electric power tower is used as a support of a power transmission line, is exposed in a field environment for a long time, and is easy to generate phenomena of tower body inclination, hardware wear and corrosion and the like, and normal transmission of electric energy is affected, so that monitoring analysis and fault diagnosis work of the operation state of the electric power tower are particularly necessary. However, the traditional manual inspection workload is large, the condition is hard and the efficiency is low, and a great burden is added to power operation and maintenance personnel, so that the improvement of the intelligent inspection level of a power system is urgent. In recent years, with the wide application of unmanned aerial vehicle technology and the rapid development of artificial intelligence technology, an electric power operation and maintenance department starts to attempt to collect and monitor information of electric power target objects such as an electric power tower through various sensing devices carried on the unmanned aerial vehicle. Unmanned aerial vehicle inspection is becoming an important means for assisting manual inspection and improving intelligent operation and maintenance level of the power grid gradually.
The unmanned aerial vehicle route planning method is a precondition and key of unmanned aerial vehicle inspection, and the barrier-free three-dimensional flight route is required to be reasonably planned according to inspection areas, surrounding environments, inspection task requirements and the like. The current online three-dimensional route planning method suitable for unmanned aerial vehicles mainly comprises the following 3 types: sampling-based algorithms such as fast explore random tree (rapidly exploring random trees, RRT) series, artificial potential field series, voronoi diagram series, probabilistic roadmap series, etc.; node-based optimization algorithms, such as Djistia algorithm, a algorithm, etc.; fusion algorithms, such as unmanned aerial vehicle path planning based on the A * algorithm and the dynamic window algorithm.
The search process of the Goal-Bias RRT algorithm is shown in FIG. 1. The Goal-Bias RRT algorithm is used for quickly searching the space by constructing a random tree, verifying the feasibility of the track by using collision detection, ensuring that the track passes through all intermediate target points, and being more widely applied compared with other planning algorithms. However, in the random sampling process, the node utilization rate is low, the randomness is high, and the optimality of the result cannot be ensured. The RRT algorithm is an improvement and optimization of the Goal-Bias RRT algorithm, and the main difference between the RRT algorithm and the Goal-Bias RRT is that the RRT algorithm introduces a search for new node adjacent nodes, so that the Goal is to select a low-cost father node, and in addition, the rerouting process further reduces the path cost, so that the method is a breakthrough method for solving the high-dimensional optimal path planning problem. The RRT algorithm is progressively optimized, and given sufficient run time, the RRT algorithm always converges to an optimal solution. The RRT algorithm solves the optimization problem of the gold-Bias RRT algorithm to some extent, but the searching for new parent nodes and the rewiring process also greatly reduces the efficiency of the algorithm.
Disclosure of Invention
The invention aims to provide a three-dimensional route planning method for a power tower, which is used for quickly exploring a random tree, integrates a heuristic search strategy to perform cost function design, accelerates algorithm convergence, ensures route optimization, and has stable and reliable shortest path length for a collision-free three-dimensional route planned by the power tower compared with an improved RRT algorithm by using a Goal-Bias RRT algorithm, so that the method can meet the application requirements of intelligent monitoring.
In order to achieve the above purpose, the invention provides a three-dimensional route planning method for a power tower for heuristic rapid exploration of a random tree, which comprises the following steps:
s1, acquiring a three-dimensional point cloud of an electric power tower through an unmanned aerial vehicle, and performing three-dimensional modeling;
S2, acquiring the space position of a key component of the power tower in advance, and taking the space position as an intermediate target point of the unmanned aerial vehicle, which needs to hover and shoot in the course of the route, so as to form a route planning constraint point;
S3, a heuristic three-dimensional route planning algorithm based on RRT, which is used for planning a three-dimensional route of the unmanned aerial vehicle inspection power tower and comprises the following specific steps:
(1) Inputting a starting point p and all target control points;
(2) Calculating the distance g (x, p) from the p point to a target control point x in a window with the height d taking the p point as the center, wherein x is one target point in the route target point set;
(3) Setting L_0 as an initial solution, calculating the shortest distance h (x, t ar) from each x point to the end point through all other target control points by using simulated annealing, and enabling L_0=L;
(4) Taking a point x of F (x, p) =min (g (x, p) +h (x, t ar)) as a next-step searching direction, marking as a p+1 point, and g (x, p) represents the actual cost of the current point p from the target point x; h (x, t ar) is a heuristic representing an estimated cost from the target point x to the endpoint t ar; f (x, p) is the total cost from the current point p through the target point x to the end point;
(5) Performing random tree expansion to generate a collision-free shortest path taking the current point p as a starting point and p+1 as an end point;
(6) Removing p points in the target control point set;
(7) If the target control point set is empty, ending the algorithm, and outputting an optimal path;
(8) If the target control point set is not empty, let p=p+1, and loop through steps (2) - (7).
Preferably, the specific step of acquiring the intermediate target point in S2 includes: based on the three-dimensional point cloud data of the power tower obtained in the step S1, manually assisting in identifying straight lines and corner points, adding interpolation points between the corner points, taking the corner points and all interpolation points as starting points, and extending outwards a certain safe distance along the normal direction to form a target point which needs hovering photographing in the course of the unmanned aerial vehicle route.
Preferably, when the direction is determined by first heuristic search, the sequence of target points is used as an initial path for simulated annealing, and the cooling rate is properly reduced in the process so as to effectively jump out a local optimal solution; after a relatively ordered target point sequence is obtained through one-time calculation, the target point sequence is used as an initial path for simulated annealing in subsequent heuristic search, and the cooling rate is properly increased to increase the solving rate.
Therefore, the invention adopts a heuristic three-dimensional route planning method for quickly exploring the electric power pole tower of the random tree in the steps, and aims at the jump problem of the generated path in height caused by the randomness of the Goal-Bias RRT algorithm searching in the three-dimensional space, and the cost function design is carried out by fusing a heuristic searching strategy, so that the random searching is constrained, and the searching direction is guided for path planning; aiming at the problems of random, suboptimal and error accumulation caused by simulated annealing, which lead to the cyclic jump of different sides of the generated path in the horizontal direction, the path searching direction is recalculated and optimized by adopting the simulated annealing guided by initialization, so that the path is further optimized. The invention not only can improve the convergence stability of the Goal-Bias RRT algorithm, but also can ensure that the route meets the better path length. In an actual scene, in order to solve the problem of algorithm efficiency reduction caused by the rapid increase of the number of target points, an optimal dynamic cooling rate group is adopted in the searching process of different stages, so that the algorithm efficiency is further improved.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a schematic diagram of the process of the Goal-Bias RRT algorithm;
FIG. 2 is a schematic flow chart of embodiment 1 of the present invention;
FIG. 3 is a three-dimensional point cloud of the power tower according to embodiment 1 of the present invention;
fig. 4 is a schematic diagram of straight lines, corner points, and interpolation points in the three-dimensional point cloud according to embodiment 1 of the present invention;
FIG. 5 is a flowchart of a heuristic three-dimensional route planning algorithm based on RRT according to embodiment 1 of the invention;
FIG. 6 shows the path planned in the experiment of 30 random points in the simulation environment of example 2 of the present invention, (a) based on the Goal-Bias RRT algorithm; (b) RRT-based algorithm; (c) is based on the algorithm herein.
FIG. 7 shows the path planned in an experiment of 70 random points in the simulation environment of example 2 of the present invention, (a) based on the Goal-Bias RRT algorithm; (b) RRT-based algorithm; (c) is based on the algorithm herein.
FIG. 8 is a graph showing the variation of the cooling rate parameter according to embodiment 3 of the present invention, (a) is a graph showing the variation of the operation time, and (b) is a graph showing the variation of the path length;
Fig. 9 is a planning path diagram of the tower 1 in the actual scenario of embodiment 3 of the present invention, (a) is based on the gol-Bias RRT algorithm; (b) RRT-based algorithm; (c) is based on the algorithm herein.
FIG. 10 is a planning path diagram of the tower 2 in the actual scenario of embodiment 3 of the present invention, (a) based on the Goal-Bias RRT algorithm; (b) RRT-based algorithm; (c) is based on the algorithm herein.
FIG. 11 is a planning path diagram of a tower 3 in the actual scenario of embodiment 3 of the present invention, (a) based on the Goal-Bias RRT algorithm; (b) RRT-based algorithm; (c) is based on the algorithm herein.
Detailed Description
The technical scheme of the invention is further described below through the attached drawings and the embodiments.
Example 1
As shown in fig. 2, a heuristic three-dimensional route planning method for a power tower of a rapid exploration random tree comprises the following steps:
S1, acquiring a three-dimensional point cloud of the power tower through an unmanned aerial vehicle, and performing three-dimensional modeling, as shown in FIG. 3.
S2, acquiring the space position of the key component of the power pole tower in advance, and taking the space position as an intermediate target point for hovering and photographing of the unmanned aerial vehicle on the way of the route to form a route planning constraint point.
As shown in fig. 4, the specific steps of acquiring the intermediate target point include: based on the three-dimensional point cloud data of the power tower obtained in the step S1, manually assisting in identifying straight lines and corner points, adding interpolation points between the corner points, taking the corner points and all interpolation points as starting points, and extending outwards a certain safe distance along the normal direction to form a target point which needs hovering photographing in the course of the unmanned aerial vehicle route.
S3, planning a three-dimensional route of the unmanned aerial vehicle inspection power tower by adopting a heuristic three-dimensional route planning algorithm based on RRT, wherein the heuristic search is based on depth-first search and breadth-first search, and is a search strategy with a better scheme obtained by being inspired by the prior knowledge experience, and the heuristic strategy guides the search by utilizing inspired information owned by the problem, so that the purposes of reducing the blindness of the search and improving the feasibility and the efficiency of the search are achieved. The cost function is constructed based on the heuristic function, so that the cost of the distance between the new state and the old state is considered, and the cost of the distance between the new state and the target state is also considered. Cost function:
F(x,p)＝min(g(x,p)+h(x,tar))
Wherein: x is one target point in the route target point set; g (x, p) represents the actual cost of the current point p from the target point x; h (x, t ar) is a heuristic representing an estimated cost from the target point x to the endpoint t ar; f (x, p) is the total cost from the current point p through the target point x to the end point, and the requirement for optimization is that the estimated cost h (x, t ar) from the x point to the end point t ar is not greater than the actual cost.
As shown in fig. 5, the specific steps of the algorithm are as follows:
(1) The starting point p and all target control points are input.
(2) The distance g (x, p) of the p-point to the target control point x in a window of height d centered at the p-point is calculated, x being one of the set of route target points.
(3) Setting L_0 as an initial solution, calculating the shortest distance h (x, t ar) from each x point to the end point through all other target control points by using simulated annealing, and enabling L_0=L; the simulated annealing algorithm is adopted to estimate h (x, tar), the core idea is to accept a solution worse than the current solution with a certain probability, and search is continued by using the worse solution, so that the algorithm is prevented from falling into local optimum.
(4) Taking the point x of F (x, p) =min (g (x, p) +h (x, t ar)) as the next searching direction, marking as p+1 point, calculating to obtain an ordered target point sequence according to the objective function minimization principle, and searching the next point x of the current point p in the sequence as the next searching direction. In addition, after reaching the point x, the ordered target point sequence calculated on the point p is used as an initial value calculated on the point x, so that the randomness brought by a simulated annealing algorithm is effectively solved, and the ordered target point sequence is further optimized.
(5) Performing random tree expansion to generate a collision-free shortest path taking the current point p as a starting point and p+1 as an end point;
(6) Removing p points in the target control point set;
(7) If the target control point set is empty, ending the algorithm, and outputting an optimal path;
(8) If the target control point set is not empty, let p=p+1, and loop through steps (2) - (7).
Example 2
And (3) establishing a three-dimensional simulation environment in Matlab software for experiments, loading and displaying a three-dimensional tower model with the origin as the center and the height of 70m, randomly generating 5 groups of target points around the tower, respectively comprising 30, 40, 50, 60 and 70 random points, and setting the starting point of the unmanned aerial vehicle route, namely the bottommost dot in fig. 6 and 7. As shown in fig. 6 (a) and fig. 7 (a), experimental results based on the gold-Bias RRT algorithm show that the random tree expansion process has higher randomness, the situation of far from the target point exists in the search process, the situation of continuous convolution of high-low jump occurs in the z-axis direction, and a great amount of redundancy exists in the finally generated route path. This is because the gold-Bias RRT algorithm itself does not consider the optimization strategy, nor is the resulting route path the optimal path, which is used primarily herein as a baseline reference path for verifying the validity and practicality of the research results of the present invention. The random tree expansion process based on the RRT algorithm is shown in (b) of fig. 6 and (b) of fig. 7, so that the problem of unnecessary turning redundancy of paths in the result is relieved, and the randomness of searching is obviously reduced.
Fig. 6 (c) and fig. 7 (c) show the three-dimensional route planning results of the method of the present invention. Compared with the Goal-Bias RRT algorithm and the RRT algorithm, the method reserves the randomness of searching among 2 target points, is favorable for avoiding obstacle searching of space, and reduces the randomness of space searching by restraining the target points which need to be reached in each step on the basis. In addition, when the search direction is determined on each target point by utilizing the heuristic strategy, the ordered target point sequence obtained by the calculation in the previous step is taken as an initial sequence, so that the randomness brought by the simulated annealing algorithm can be reduced, the stability of the algorithm is improved, and the search of a 'useless' space is further reduced, thereby finally improving the space search efficiency and the convergence speed of the route path and ensuring that a relatively better route path is obtained.
Table 1 shows the comparison results of the method of the present invention with the gold-Bias RRT algorithm and RRT algorithm under a simulation environment. In general, as the number of random target points increases, the extent of decrease in the planned route length and variance of the method of the present invention increases further, i.e., the advantages of the method become more apparent, relative to the RRT algorithm and RRT algorithm. It should be noted that the variance of the route length is intended to reflect the deviation degree of the route length from the expected value obtained in the case of executing the algorithm multiple times, and is used to reveal the stability of the route length. Compared with the route path of the Goal-Bias RRT algorithm, the route length obtained by the method is relatively stable, the variance of the route length is reduced by 99.94% on average, and the influence of randomness is reduced; the average course length was reduced by 62.02% and the calculation time increased from 0.12s to 9.99s in the 30 random point experiment, although there was an increase in course calculation time, it was still acceptable. Compared with the route path of the RRT algorithm, the route length planned by the method is reduced by 48.05 percent on average, the variance of the route length is reduced by 99.94 percent on average, and the calculation time is increased from 2.51s to 9.99s in an experiment of 30 random points.
Table 1 comparison of the results of the present method and the gold-Bias RRT algorithm, RRT algorithm in a simulation environment
Example 3
In an actual scene, 3 different power towers under the Guangdong tail market kV power transmission line are selected in the embodiment. And extracting angular points and interpolation points in the 3 electric towers to respectively obtain 292, 128 and 122 target points, and setting corresponding unmanned aerial vehicle route starting point coordinates. The calculation efficiency of the invention is reduced due to the rapid increase of the number of target points in the actual scene, and the algorithm parameters are required to be further optimized, namely the dynamic cooling rate is adopted in the simulated annealing, rather than the single static cooling rate. When the direction is determined by first heuristic search, a random target point sequence is used as an initial path for simulated annealing, the probability of accepting a suboptimal solution is gradually reduced in each cooling process, and the cooling rate needs to be properly reduced in order to effectively jump out of the locally optimal solution; after a relatively ordered target point sequence is obtained through one-time calculation, the target point sequence is used as an initial path for simulating annealing in subsequent heuristic search, and in order to improve the solving efficiency, the cooling rate needs to be properly improved. Taking alpha 1 as a first simulated annealing cooling rate, alpha 2 as a subsequent simulated annealing cooling rate, carrying out a plurality of groups of experiments, and drawing a cooling rate parameter change chart according to the operation effect and efficiency of an algorithm, wherein as shown in fig. 8, an optimal cooling rate parameter combination alpha 1 = 0.998, alpha 2 = 0.75, the operation time of the algorithm is 54.24s, and the optimal path length is 1020.51m; however, if the same cool down rate α=0.95 is used, the algorithm runs for an average time of 255.62s and the optimal path for an average length of 1107.88m. Therefore, the selection of the proper cooling rate parameter group not only improves the operation efficiency of the algorithm, but also maintains the better path search result.
Fig. 9 (a), fig. 10 (a) and fig. 11 (a) show the result of route planning for 3 different power towers gol-Bias RRT algorithms in a real scene, wherein a large number of high-low jump redundant paths exist in a three-dimensional space, and the proportion of the redundant paths increases with the increase of target points; fig. 9 (b), fig. 10 (b) and fig. 11 (b) show the result of the algorithm route planning for 3 different power towers RRT in the actual scenario, and it is seen that the redundant paths are relieved to some extent; fig. 9 (c), fig. 10 (c) and fig. 11 (c) show the planning result of the method of the present invention under the reasonable setting of dynamic parameters, and compared with the golal-Bias RRT algorithm and the RRT algorithm, three-dimensional routes of 3 different power towers all have advantages, since heuristic search is a path planning direction, the height jump of the route in the vertical direction is reduced, and meanwhile, since the initialized simulated annealing is performed for many times, the solution with error is optimized every time the searching direction is planned, and the cycle jump of different sides of the route in the horizontal direction is reduced.
Table 2 shows the quantitative comparison of the method of the present invention with the gold-Bias RRT algorithm and RRT algorithm under actual scenarios, showing the stability and superiority of the method of the present invention to different tower route planning results. Specifically, the planned route length of the RRT algorithm is reduced by 30.56% on average compared with the route length planned by the gold-Bias RRT algorithm, but the calculation time is increased from 1.99s to 75.10s; compared with the route length planned by the Goal-Bias RRT algorithm and the route length planned by the RRT algorithm, the route length planned by the method is respectively reduced by 71.51 percent and 58.98 percent on average, and the variance of the route length is respectively reduced by 99.69 percent and 99.83 percent on average. However, the method of the present invention requires an average 19.64s in terms of computation time, which is an increase compared to 1.99s for the GoalBias RRT algorithm, but a substantial decrease compared to 75.10s for the RRT algorithm. It should be noted that the calculation time of the method is basically controlled within 30s, so that the application requirements in the actual scene can be met.
Table 2 results of the method in actual scenario are compared with the Goal-Bias RRT algorithm and the RRT algorithm
In an actual scene, in order to solve the problem of algorithm efficiency reduction caused by the rapid increase of the number of target points, the invention adopts an optimal dynamic cooling rate group in the searching process of different stages, so that the algorithm efficiency is further improved. Experiments are carried out in a simulation environment and an actual scene, and experimental results show that compared with a gold-Bias RRT algorithm and an improved RRT algorithm, the path length drawn by the law is stable and reliable, the length is shorter, and auxiliary support can be provided for effective monitoring and fault diagnosis of the running state of the power tower.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention and not for limiting it, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that: the technical scheme of the invention can be modified or replaced by the same, and the modified technical scheme cannot deviate from the spirit and scope of the technical scheme of the invention.
Claims (2)
1. A heuristic three-dimensional route planning method for a power tower for quickly exploring a random tree is characterized by comprising the following steps of: the method comprises the following steps:
s1, acquiring a three-dimensional point cloud of an electric power tower through an unmanned aerial vehicle, and performing three-dimensional modeling;
S2, acquiring the space position of a key component of the power tower in advance, and taking the space position as an intermediate target point of the unmanned aerial vehicle, which needs to hover and shoot in the course of the route, so as to form a route planning constraint point; the specific step of acquiring the intermediate target point in S2 includes: based on the three-dimensional point cloud data of the power tower obtained in the step S1, manually assisting in identifying straight lines and corner points, adding interpolation points between the corner points, and taking the corner points and all interpolation points as starting points, and extending outwards a certain safe distance along the normal direction to form a target point needing hovering photographing in the course of the unmanned aerial vehicle route;
S3, adopting a heuristic three-dimensional route planning algorithm based on RRT to plan a three-dimensional route of the unmanned aerial vehicle inspection power tower, wherein the specific steps of the algorithm are as follows:
(1) Inputting a starting point p and all target control points;
(2) Calculating the distance g (x, p) from the p point to a target control point x in a window with the height d taking the p point as the center, wherein x is one target point in the route target point set;
(3) Setting L_0 as an initial solution, calculating the shortest distance h (x, t ar) from each x point to the end point through all other target control points by using simulated annealing, and enabling L_0=L;
(4) Taking a point x of F (x, p) =min (g (x, p) +h (x, t ar)) as a next-step searching direction, marking as a p+1 point, and g (x, p) represents the actual cost of the current point p from the target point x; h (x, t ar) is a heuristic representing an estimated cost from the target point x to the endpoint t ar; f (x, p) is the total cost from the current point p through the target point x to the end point;
(5) Performing random tree expansion to generate a collision-free shortest path taking the current point p as a starting point and p+1 as an end point;
(6) Removing p points in the target control point set;
(7) If the target control point set is empty, ending the algorithm, and outputting an optimal path;
(8) If the target control point set is not empty, let p=p+1, and loop through steps (2) - (7).
2. The method for three-dimensional planning of the electric power tower of the heuristic rapid exploration random tree according to claim 1, wherein the method comprises the following steps: in S3, when the number of target points is increased sharply, a dynamic cooling rate is adopted in the simulated annealing calculation, and the specific steps are as follows: when the direction is determined by first heuristic search, taking a sequence of random target points as an initial path of simulated annealing, and properly reducing the cooling rate in the process to effectively jump out of a local optimal solution; after a relatively ordered target point sequence is obtained through one-time calculation, the target point sequence is used as an initial path for simulated annealing in subsequent heuristic search, and the cooling rate is properly increased to increase the solving rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311259360.9A CN117308969B (en) | 2023-09-27 | 2023-09-27 | Heuristic three-dimensional route planning method for electric power pole tower for quickly exploring random tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311259360.9A CN117308969B (en) | 2023-09-27 | 2023-09-27 | Heuristic three-dimensional route planning method for electric power pole tower for quickly exploring random tree |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117308969A CN117308969A (en) | 2023-12-29 |
| CN117308969B true CN117308969B (en) | 2024-05-14 |
Family
ID=89261548
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311259360.9A Active CN117308969B (en) | 2023-09-27 | 2023-09-27 | Heuristic three-dimensional route planning method for electric power pole tower for quickly exploring random tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117308969B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108896052A (en) * | 2018-09-20 | 2018-11-27 | 鲁东大学 | A kind of mobile robot smooth paths planing method under the environment based on DYNAMIC COMPLEX |
| CN110083173A (en) * | 2019-04-08 | 2019-08-02 | 合肥工业大学 | The optimization method of unmanned plane formation patrol task distribution |
| CN110609569A (en) * | 2019-09-26 | 2019-12-24 | 温岭市非普电气有限公司 | Autonomous control unmanned aerial vehicle accurate inspection system and method applied to electric power tower |
| CN110727288A (en) * | 2019-11-13 | 2020-01-24 | 昆明能讯科技有限责任公司 | Point cloud-based accurate three-dimensional route planning method for power inspection |
| CN113110521A (en) * | 2021-05-26 | 2021-07-13 | 中国科学技术大学 | Mobile robot path planning control method, control device thereof and storage medium |
| CN114610026A (en) * | 2022-03-04 | 2022-06-10 | 广东电网有限责任公司广州供电局 | Line planning method, system and medium for accessing centralized automobile charging station to power distribution network |
| CN115629617A (en) * | 2022-10-25 | 2023-01-20 | 北京拓维思科技有限公司 | Automatic planning method for unmanned aerial vehicle tower pole inspection air route |
| CN116149355A (en) * | 2022-12-05 | 2023-05-23 | 国网福建省电力有限公司电力科学研究院 | Power transmission operation and maintenance unmanned aerial vehicle inspection route planning method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018095278A1 (en) * | 2016-11-24 | 2018-05-31 | 腾讯科技（深圳）有限公司 | Aircraft information acquisition method, apparatus and device |
| US11231715B2 (en) * | 2018-05-22 | 2022-01-25 | King Fahd University Of Petroleum And Minerals | Method and system for controlling a vehicle |
-
2023
- 2023-09-27 CN CN202311259360.9A patent/CN117308969B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108896052A (en) * | 2018-09-20 | 2018-11-27 | 鲁东大学 | A kind of mobile robot smooth paths planing method under the environment based on DYNAMIC COMPLEX |
| CN110083173A (en) * | 2019-04-08 | 2019-08-02 | 合肥工业大学 | The optimization method of unmanned plane formation patrol task distribution |
| CN110609569A (en) * | 2019-09-26 | 2019-12-24 | 温岭市非普电气有限公司 | Autonomous control unmanned aerial vehicle accurate inspection system and method applied to electric power tower |
| CN110727288A (en) * | 2019-11-13 | 2020-01-24 | 昆明能讯科技有限责任公司 | Point cloud-based accurate three-dimensional route planning method for power inspection |
| CN113110521A (en) * | 2021-05-26 | 2021-07-13 | 中国科学技术大学 | Mobile robot path planning control method, control device thereof and storage medium |
| CN114610026A (en) * | 2022-03-04 | 2022-06-10 | 广东电网有限责任公司广州供电局 | Line planning method, system and medium for accessing centralized automobile charging station to power distribution network |
| CN115629617A (en) * | 2022-10-25 | 2023-01-20 | 北京拓维思科技有限公司 | Automatic planning method for unmanned aerial vehicle tower pole inspection air route |
| CN116149355A (en) * | 2022-12-05 | 2023-05-23 | 国网福建省电力有限公司电力科学研究院 | Power transmission operation and maintenance unmanned aerial vehicle inspection route planning method |
Non-Patent Citations (3)
| Title |
|---|
| Peng Xiao ; Shiyou Mu ; Yan Wen ; 等.Design of a Route Planning Method for Inspecting Distribution Network.2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC).2018,全文. * |
| 基于RRT算法的某型无人机航路在线规划设计;路引;郭昱津;王道波;;兵器装备工程学报;20161231;37(12);全文 * |
| 无人机倾斜摄影技术在电力巡线树障检测中的实践应用研究;周小红, 李向欢, 石蕾, 龙焕;电力大数据;20190831;第22卷(第8期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117308969A (en) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106371445B (en) | A kind of unmanned vehicle planning control method based on topological map | |
| CN113219998B (en) | Improved bidirectional-RRT-based vehicle path planning method | |
| CN109343345B (en) | Mechanical arm polynomial interpolation track planning method based on QPSO algorithm | |
| CN109163722B (en) | Humanoid robot path planning method and device | |
| CN105841702A (en) | Method for planning routes of multi-unmanned aerial vehicles based on particle swarm optimization algorithm | |
| CN108268042A (en) | A kind of path planning algorithm based on improvement Visual Graph construction | |
| CN111426323B (en) | Routing planning method and device for inspection robot | |
| CN112990538B (en) | Method, device and equipment for determining collecting line of mountain photovoltaic power station | |
| CN112270111A (en) | Power system active scheduling method and system for coping with ice disaster | |
| CN114489052A (en) | Path planning method for improving RRT algorithm reconnection strategy | |
| CN112947480A (en) | Mobile robot path planning method, storage medium and system | |
| CN106981194B (en) | A kind of recognition methods of highway network key road segment | |
| CN114326726B (en) | Formation path planning control method based on A and improved artificial potential field method | |
| CN115562272A (en) | Planning method and device for robot traveling path | |
| Chen et al. | Global path planning method by fusion of A-star algorithm and sparrow search algorithm | |
| CN117308969B (en) | Heuristic three-dimensional route planning method for electric power pole tower for quickly exploring random tree | |
| CN111664851A (en) | Robot state planning method and device based on sequence optimization and storage medium | |
| Ewing et al. | Betweenness centrality in multi-agent path finding | |
| Yu et al. | PMPSO: a near-optimal graph planarization algorithm using probability model based particle swarm optimization | |
| CN117075632A (en) | Substation unmanned aerial vehicle inspection route planning method, device, equipment and medium | |
| CN114911233A (en) | Football robot path planning method based on multi-optimization rapid expansion random tree | |
| CN116307325A (en) | Line planning method and device for power distribution network, electronic equipment and storage medium | |
| CN113885567B (en) | Collaborative path planning method for multiple unmanned aerial vehicles based on conflict search | |
| CN116009552A (en) | Path planning method, device, equipment and storage medium | |
| Jiang | Branch Cable Path Planning Method Based on “Genetic-A*” Fusion Algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |