CN-121994239-A - Plant protection unmanned aerial vehicle path planning method and plant protection system based on end edge cooperation

CN121994239ACN 121994239 ACN121994239 ACN 121994239ACN-121994239-A

Abstract

A plant protection unmanned aerial vehicle path planning method based on end-edge cooperation comprises the steps of obtaining offline map data of an area to be worked, constructing a multi-layer planning map based on the offline map data, wherein the multi-layer planning map comprises a plurality of electricity changing points and a plurality of medicine adding points, establishing a residual capacity model and a residual quantity model based on characteristic parameters of a plant protection unmanned aerial vehicle, constructing a multi-objective optimization function based on the residual capacity model, the residual quantity model and navigation constraint conditions, generating a plurality of initial path planning schemes, screening the initial path planning schemes by utilizing an improved genetic particle algorithm and combining the multi-objective optimization function to obtain an optimal path planning scheme, and optimizing the optimal path planning scheme in real time based on dynamic early warning information in the navigation process of the plant protection unmanned aerial vehicle based on the optimal path planning scheme. The invention provides a method for solving the problem of multi-constraint matching, dynamic path adjustment and end edge data coordination in the operation of a plant protection unmanned aerial vehicle.

Inventors

WANG ZHIKAI
JI BAOFENG
HUANG JINGTAO
GAO SONG
LI NA
ZHANG XU
GUO QI
FU ZHUMU
TAO FAZHAN
WANG NAN
ZHU LONGLONG
LI LING
WANG JUN
YIN SHANSHAN

Assignees

河南科技大学

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (10)

1. A plant protection unmanned aerial vehicle path planning method based on end-edge cooperation is characterized by comprising the following steps: Acquiring offline map data of an area to be worked, and constructing a multi-layer planning map based on the offline map data, wherein the multi-layer planning map comprises a plurality of power change points and a plurality of medicine adding points; establishing a residual electric quantity model and a residual electric quantity model based on characteristic parameters of the plant protection unmanned aerial vehicle; constructing a multi-objective optimization function based on the residual electric quantity model, the residual electric quantity model and navigation constraint conditions; generating a plurality of initial path planning schemes, and screening the initial path planning schemes by utilizing an improved genetic particle algorithm in combination with a multi-objective optimization function to obtain an optimal path planning scheme; And in the sailing process of the plant protection unmanned aerial vehicle based on the optimal path planning scheme, the optimal path planning scheme is optimized in real time based on the dynamic early warning information.
2. The end-edge collaboration-based plant protection unmanned aerial vehicle path planning method of claim 1, wherein the method for constructing the multi-layer planning map based on the offline map data comprises the following steps: Noise reduction is carried out on offline map data of an area to be worked, and topographic data, obstacle data, land block boundary data and functional node data are analyzed from the offline map data, wherein the functional node data comprise a plurality of electricity changing points and a plurality of medicine adding points; Constructing a local rectangular coordinate system of the area to be worked, and converting the offline map data into the local rectangular coordinate system to obtain a basic planning map; And carrying out layer division on the basic planning map to obtain a multi-layer planning map, wherein the multi-layer planning map comprises a basic terrain layer, an obstacle layer, a functional node layer and an operation land parcel layer.
3. The method for planning a path of a plant protection unmanned aerial vehicle based on end-to-end cooperation according to claim 2, wherein after analyzing the data of the functional nodes, determining a first attribute of a power change point and a second attribute of a medicine adding point, wherein the first attribute comprises a power change coordinate, a power change time period, a single power change duration, a maximum simultaneous power change unmanned aerial vehicle number and a battery residual capacity, and the second attribute comprises a medicine adding coordinate, a pesticide type, a single medicine adding duration and a residual medicine amount; After the topographic data is analyzed, the area to be worked is divided into a plurality of plots based on the topographic data, and third attributes of the plots are determined, wherein the third attributes comprise plot identifications, boundary coordinate sets, plot areas, plot types and pesticide application parameters.
4. The end-to-edge synergy-based plant protection unmanned aerial vehicle path planning method of claim 1, wherein the characteristic parameters of the unmanned aerial vehicle comprise intrinsic parameters, weight parameters, energy consumption related parameters and drug consumption related parameters, the intrinsic parameters comprise maximum flight speed, drug application speed, battery capacity and drug tank volume, the weight parameters comprise self net weight and pesticide density, the energy consumption related parameters comprise unit flight energy consumption, basic hover power and load power coefficients, and the drug consumption related parameters comprise unit area drug application amount and drug application rate.
5. The method for planning a path of a plant protection unmanned aerial vehicle based on end-to-end coordination according to claim 4, wherein the residual capacity model is: ; Wherein, the For the number of job paths corresponding to the parcel, In order to round up the function, Is the width of the plot in a first direction, For the line spacing of the plant protection unmanned aerial vehicle, For the total length of the working path, Is the width of the land in the second direction, and the first direction is perpendicular to the second direction, In order to achieve a long period of time for the application of the medicament, In order to achieve a high rate of administration, As the residual quantity of electricity, the residual quantity, For the initial amount of power, In order to achieve the energy consumption of the flight, Is the energy consumption of the unit flight and is used for controlling the speed of the aircraft, The distance between the nearest dosing point and the nearest power change point is returned to the unmanned aerial vehicle in sequence, And reserving safe electric quantity for the user.
6. The end-edge collaboration-based plant protection unmanned aerial vehicle path planning method of claim 4, wherein the multi-objective optimization function is: ; ; ; ; ; ; Wherein, the The sum of flight distances among the functional nodes is that the functional nodes are power exchange points or dosing points, For the sum of the job path lengths for all plots, For the integration of the power consumption of the flight, In order to sum up the power consumption of the drug administration, In order to determine the number of turns between the nodes, For the total number of turns of the work path within the plot, Is the residual electric quantity deviation value and has , Is the residual electric quantity when the jth power change point is reached, For the optimum value of the remaining power when the point is shifted, For the total number of times of whole-course power change in operation, For the total number of times of adding medicine in the whole process of operation, Is a weight coefficient.
7. The end-edge collaboration-based plant protection unmanned aerial vehicle path planning method of claim 1, wherein the method of generating a plurality of initial path planning schemes comprises: clustering the plots based on the positions, and generating a first partial scheme based on the clustering result; generating a second partial solution by random perturbation based on the first partial solution; integrating the first partial scheme and the second partial scheme to obtain an initial path planning scheme; the method for screening the initial path planning scheme by utilizing the improved genetic particle algorithm and combining the multi-objective optimization function comprises the following steps: Screening the initial path planning scheme based on the multi-objective optimization function to obtain a plurality of alternative path planning schemes; encoding the alternative path planning scheme and generating a population comprising a plurality of particles based on the encoding; dividing the particles into dominant particles or disadvantaged particles based on the fitness value pairs of the particles; Performing iterative optimization on the dominant particles through a dynamic particle swarm optimization algorithm, performing iterative optimization on the disadvantaged particles through an improved genetic algorithm, and performing information interaction and cross variation on part of the dominant particles and part of the disadvantaged particles in the iterative optimization process; After the iteration is finished, generating a pareto optimal path set based on the particles; and screening again in the pareto optimal path set to obtain an optimal path planning scheme.
8. The end-to-end coordination-based plant protection unmanned aerial vehicle path planning method of claim 1, wherein the dynamic early warning information comprises high priority early warning information, medium priority early warning information and low priority early warning information, wherein the high priority early warning information comprises obstacle collision early warning, low electric quantity early warning and low drug quantity early warning, the medium priority early warning information comprises path deviation early warning and wind speed overrun early warning, and the low priority early warning information comprises low electric quantity early warning and low drug quantity early warning.
9. The method for planning a path of a plant protection unmanned aerial vehicle based on end-to-end coordination according to claim 8, wherein the method for optimizing the optimal path planning scheme in real time based on dynamic early warning information comprises: And regenerating an optimal path planning scheme when the high-priority early warning information is triggered, locally adjusting the optimal path planning scheme when the medium-priority early warning information is triggered, and maintaining the optimal path planning scheme when the low-priority early warning information is triggered.
10. Plant protection system, characterized by comprising a plurality of edge servers and a plurality of plant protection unmanned aerial vehicle terminals, wherein the edge servers are used for drawing out an optimal path planning scheme based on the plant protection unmanned aerial vehicle path planning party regulation based on end-to-end cooperation according to any one of claims 1-9, and the plant protection unmanned aerial vehicle terminals are used for sailing based on the optimal path planning scheme and spraying pesticides to plants.

Description

Plant protection unmanned aerial vehicle path planning method and plant protection system based on end edge cooperation Technical Field The invention relates to the technical field of plant protection unmanned aerial vehicles, in particular to a plant protection unmanned aerial vehicle path planning method and a plant protection system based on end-to-side cooperation. Background The plant protection unmanned aerial vehicle is important equipment for efficiently developing operations such as pest control, crop fertilization and the like in modern agricultural production, and can finish accurate pesticide application operations in farmland, orchards and other areas by carrying pesticide application devices, so that the operation efficiency is greatly improved, and the labor cost is reduced. Along with the development of wisdom agriculture, put forward higher requirement to plant protection unmanned aerial vehicle's operation precision, efficiency and security, and the path planning is as the core link of plant protection unmanned aerial vehicle operation, directly influences operation quality and cost. The existing plant protection unmanned aerial vehicle path planning method is mostly based on single-dimension optimization, and if only the shortest range or the minimum operation time is considered, the unmanned aerial vehicle power consumption and the dynamic change of the drug consumption are not fully combined, and the real-time data of the edge server and the local decision capability of the unmanned aerial vehicle end are not effectively fused. In actual operation, unmanned aerial vehicle often causes the outage in the middle of because of the electric quantity estimates inaccurately, and the dose planning is unreasonable causes the operation to break, and when facing construction area and dynamic barriers such as bird that appear temporarily, unable quick adjustment route not only influences operating efficiency, still probably causes the incident. Disclosure of Invention In order to solve the defects in the prior art, the invention provides a plant protection unmanned aerial vehicle path planning method and a plant protection system based on end-to-side coordination, which can realize optimization of an operation path while solving the problems of multi-constraint matching, dynamic path adjustment and end-to-side data coordination in the operation of the plant protection unmanned aerial vehicle, thereby achieving the effects of improving the operation efficiency, reducing the operation cost and guaranteeing the operation safety. In order to achieve the purpose, the invention adopts the specific scheme that the plant protection unmanned aerial vehicle path planning method based on end-to-side cooperation comprises the following steps: Acquiring offline map data of an area to be worked, and constructing a multi-layer planning map based on the offline map data, wherein the multi-layer planning map comprises a plurality of power change points and a plurality of medicine adding points; establishing a residual electric quantity model and a residual electric quantity model based on characteristic parameters of the plant protection unmanned aerial vehicle; constructing a multi-objective optimization function based on the residual electric quantity model, the residual electric quantity model and navigation constraint conditions; generating a plurality of initial path planning schemes, and screening the initial path planning schemes by utilizing an improved genetic particle algorithm in combination with a multi-objective optimization function to obtain an optimal path planning scheme; And in the sailing process of the plant protection unmanned aerial vehicle based on the optimal path planning scheme, the optimal path planning scheme is optimized in real time based on the dynamic early warning information. As a further optimization of the plant protection unmanned aerial vehicle path planning method based on end-edge cooperation, the method for constructing the multi-layer planning map based on the offline map data comprises the following steps: Noise reduction is carried out on offline map data of an area to be worked, and topographic data, obstacle data, land block boundary data and functional node data are analyzed from the offline map data, wherein the functional node data comprise a plurality of electricity changing points and a plurality of medicine adding points; Constructing a local rectangular coordinate system of the area to be worked, and converting the offline map data into the local rectangular coordinate system to obtain a basic planning map; And carrying out layer division on the basic planning map to obtain a multi-layer planning map, wherein the multi-layer planning map comprises a basic terrain layer, an obstacle layer, a functional node layer and an operation land parcel layer. After analyzing the functional node data, determining a first attribute of a power change point and a second attribut