CN-121995951-A - Intelligent point selection method and system for unmanned aerial vehicle inspection take-off and landing points

CN121995951ACN 121995951 ACN121995951 ACN 121995951ACN-121995951-A

Abstract

The application relates to an intelligent point selection method and system for an unmanned aerial vehicle inspection take-off and landing point, which belong to the technical field of operation and maintenance of power transmission lines, and the method comprises the steps of constructing a high-precision power grid three-dimensional space reference model with attribute information by fusing multi-source data such as laser point cloud, oblique photography, power grid ledger and the like; and in the inspection process, the state and environment dynamic information of the unmanned aerial vehicle are acquired in real time, the dynamic calculation is carried out by utilizing an improved multi-target particle swarm optimization algorithm, and the alternative take-off and landing points with the lowest energy consumption, optimal task aging and maximum safety margin are rapidly output from the candidate point set. The application realizes jump from static presetting to dynamic real-time optimization of the landing point planning, and effectively solves the problem that the prior art depends on manual experience and cannot adapt to the dynamic change of the inspection task.

Inventors

LIANG MANSHU
LI XIAODONG
RUAN YING
WU WENBIN
Yao Shuning
WANG RENSHU
ZHANG WEIHAO
LIN CHENGHUA
CHEN ZHUOLEI
LI WENQI
WU XIAOJIE

Assignees

国网福建省电力有限公司电力科学研究院
国网福建省电力有限公司

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (10)

1. An intelligent point selecting method for an unmanned aerial vehicle inspection take-off and landing point is characterized by comprising the following steps: Acquiring multi-source data of an unmanned aerial vehicle inspection area, and constructing a three-dimensional space reference model containing a power grid equipment model and attribute information of the power grid equipment model based on the multi-source data; generating a plurality of candidate take-off and landing points in an unmanned aerial vehicle inspection area based on preset constraint conditions and a three-dimensional space reference model, and forming a candidate point set; And acquiring unmanned aerial vehicle state information and environment dynamic information in real time, dynamically calculating through a multi-objective optimization algorithm based on the unmanned aerial vehicle state information, the environment dynamic information and the candidate point set, and outputting at least one optimal take-off and landing point from the candidate point set.
2. The intelligent point selecting method for the unmanned aerial vehicle inspection take-off and landing points of claim 1, wherein the multi-source data comprise laser point cloud data, oblique photogrammetry data and grid equipment ledger data.
3. The intelligent point selecting method for the unmanned aerial vehicle inspection take-off and landing points of claim 2 is characterized in that the construction flow of the three-dimensional space reference model comprises the following steps: registering and fusing laser point cloud data and oblique photogrammetry data, and correcting geometric features of the power equipment to form a fused three-dimensional model; and carrying out association binding on the attribute information in the grid equipment ledger data and the corresponding power equipment model in the fusion three-dimensional model to generate a three-dimensional space reference model.
4. The method for intelligently selecting the unmanned aerial vehicle patrol take-off and landing points according to claim 1, wherein the step of generating a plurality of candidate take-off and landing points in the unmanned aerial vehicle patrol area based on the preset constraint condition and the three-dimensional space reference model comprises the following steps: constructing an fitness function according to the constraint conditions, wherein the fitness function is used for scoring any point, and the point which violates the constraint conditions is given a punishment value; and adopting an improved genetic algorithm, taking an fitness function as an evaluation standard, performing iterative search in a three-dimensional space in an unmanned aerial vehicle inspection area, and screening out points meeting the constraint condition as the candidate take-off and landing points.
5. The intelligent point selection method for the unmanned aerial vehicle inspection take-off and landing points of claim 4, wherein the constraint conditions comprise a safe distance constraint, a terrain constraint, a airspace constraint and a redundancy constraint.
6. The intelligent point selecting method for the unmanned aerial vehicle inspection take-off and landing points of claim 1, wherein the acquiring the unmanned aerial vehicle state information and the environment dynamic information in real time comprises the following steps: The unmanned aerial vehicle state information comprises real-time positions, residual electric quantity, duration and task quantity to be executed; the environmental dynamic information includes real-time weather data and/or sudden obstacle information.
7. The method for intelligently selecting the take-off and landing points for the unmanned aerial vehicle according to claim 1, wherein the step of dynamically calculating by a multi-objective optimization algorithm comprises the following steps: constructing a multi-objective optimization function comprising flight energy consumption, task timeliness and safety margin; solving the multi-objective optimization function by adopting an improved multi-objective particle swarm optimization algorithm, wherein the specific improvement of the multi-objective particle swarm optimization algorithm comprises the introduction of a dynamic inertia weight and an elite retention mechanism; and responding to the mutation of the unmanned aerial vehicle state information exceeding a preset threshold value, and triggering the particle mutation operation of the multi-target particle swarm optimization algorithm.
8. An intelligent point selection system for unmanned aerial vehicle inspection take-off and landing points, which is characterized by comprising: the reference model construction module is used for acquiring multi-source data of the unmanned aerial vehicle inspection area and constructing a three-dimensional space reference model containing a power grid equipment model and attribute information thereof based on the multi-source data; The candidate set generation module is used for generating a plurality of candidate take-off and landing points in the unmanned aerial vehicle inspection area based on a preset constraint condition and a three-dimensional space reference model, and forming a candidate point set; The intelligent point selection module acquires unmanned aerial vehicle state information and environment dynamic information in real time, dynamically calculates through a multi-target optimization algorithm based on the unmanned aerial vehicle state information, the environment dynamic information and the candidate point set, and outputs at least one optimal take-off and landing point from the candidate point set.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements an intelligent point selection method for unmanned aerial vehicle inspection take-off and landing points as claimed in claims 1-7 when executing the program.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements an intelligent point selection method for a drone patrol landing point according to claims 1-7.

Description

Intelligent point selection method and system for unmanned aerial vehicle inspection take-off and landing points Technical Field The application relates to the technical field of operation and maintenance of power transmission lines, in particular to an intelligent point selection method and system for unmanned aerial vehicle inspection take-off and landing points. Background Unmanned aerial vehicle technology is increasingly widely applied to transmission line inspection, and reasonable planning of take-off and landing points is a key link for guaranteeing inspection operation safety and efficiency. The prior art has the common pain points that 1) dynamic factors in the flying process are not fully considered in the taking-off and landing point planning stage, the flexibility is poor, 2) the optimization targets are single, only single factors such as distance or topography are often considered, a plurality of conflict targets such as flying energy consumption, task aging and safety margin are difficult to consider, 3) the precision and the reliability depend on manual work, the adaptability to the complex environment of a power grid is insufficient, and key links such as safety distance verification still need to be judged by experience of patrol personnel. The prior art discloses an automatic generation method of the take-off and landing points of an electric power inspection unmanned aerial vehicle, for example, according to the Chinese patent with the publication number of CN115268488A, by introducing electric tower positions, topographic data, airspace data and the like, topographic analysis and regional division are carried out, so that alternative take-off and landing points are automatically generated. Although the method reduces the workload of manual investigation to a certain extent, the method is still a static off-line planning method. The method cannot dynamically adjust and reprogram according to the real-time state and sudden environmental change of the unmanned aerial vehicle in the actual inspection process of the unmanned aerial vehicle. When the inspection task is prolonged or the environmental conditions are deteriorated, the pre-planned take-off and landing points may no longer be applicable, even with safety risks. Therefore, the field is urgent to need an intelligent take-off and landing point selection method capable of responding to dynamic changes in real time, comprehensively optimizing multiple targets and deeply fusing three-dimensional scenes of a power grid so as to truly realize the intellectualization and automation of unmanned aerial vehicle inspection. Disclosure of Invention In order to solve the technical problems, the invention provides an intelligent point selecting method and system for unmanned aerial vehicle inspection take-off and landing points. The technical scheme of the invention is as follows: the invention provides an intelligent point selecting method for unmanned aerial vehicle inspection taking-off and landing points, which comprises the following steps: Acquiring multi-source data of an unmanned aerial vehicle inspection area, and constructing a three-dimensional space reference model containing a power grid equipment model and attribute information of the power grid equipment model based on the multi-source data; generating a plurality of candidate take-off and landing points in an unmanned aerial vehicle inspection area based on preset constraint conditions and a three-dimensional space reference model, and forming a candidate point set; And acquiring unmanned aerial vehicle state information and environment dynamic information in real time, dynamically calculating through a multi-objective optimization algorithm based on the unmanned aerial vehicle state information, the environment dynamic information and the candidate point set, and outputting at least one optimal take-off and landing point from the candidate point set. Preferably, the multi-source data includes laser point cloud data, oblique photogrammetry data and grid equipment ledger data. Preferably, the construction process of the three-dimensional space reference model includes: registering and fusing laser point cloud data and oblique photogrammetry data, and correcting geometric features of the power equipment to form a fused three-dimensional model; and carrying out association binding on the attribute information in the grid equipment ledger data and the corresponding power equipment model in the fusion three-dimensional model to generate a three-dimensional space reference model. Preferably, the step of generating a plurality of candidate landing points in the unmanned aerial vehicle inspection area based on the preset constraint condition and the three-dimensional space reference model includes: constructing an fitness function according to the constraint conditions, wherein the fitness function is used for scoring any point, and the point which violates the constraint conditions is g