CN118816899A - A UAV path planning method, device, terminal and storage medium - Google Patents

Abstract

The present invention discloses a method, device, terminal and storage medium for drone path planning. The method determines a candidate viewpoint set according to a two-dimensional base map and a safe flight altitude; generates an initial viewpoint set according to the candidate viewpoint set based on plane quality and viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane; optimizes the initial viewpoint set using a multi-objective optimization algorithm to determine an optimized viewpoint set; and determines the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set. The present invention obtains a viewpoint set capable of reconstructing high-quality textures by introducing plane quality and viewpoint-plane quality, thereby effectively solving the problem that the prior art does not consider how to collect high-quality image data for texture reconstruction, resulting in the inability to achieve an ideal texture effect when the collected image data is used for texture mapping.

Description

A UAV path planning method, device, terminal and storage medium

Technical Field

The present invention relates to the field of computer graphics, and in particular to a method, device, terminal and storage medium for unmanned aerial vehicle path planning.

Background Art

The construction of large-scale digital twin cities is usually based on image materials collected by unmanned aerial vehicles (UAVs) to generate structured models with realistic textures. Current UAV acquisition methods mainly aim at high-quality geometric reconstruction. For example: (1) A volume hierarchical representation is used to estimate the cost of reconstructing each part of the scene, balancing the two goals of maximizing the information gain of the viewpoint and minimizing the path length; (2) A scene coverage model is used to formulate a sub-module optimization method to determine the best direction for each candidate viewpoint, and an integer linear program is used to select the best path from the candidate viewpoints to maximize the scene coverage while obtaining a shorter path length; (3) A reconstructible metric is introduced to predict the reconstruction quality, and a heuristic-based continuous optimization method is used to maximize the reconstructibility of all viewpoints in the scene; (4) An adaptive UAV path planning algorithm without field investigation is used to estimate a 2.5-dimensional model for the scene based on the relationship between buildings and corresponding shadows, and the minimum viewpoint set is selected using a maximum-minimum optimization method to maximize the reconstruction quality with the same number of viewpoints; (5) A real-time UAV path planning method for urban scene reconstruction uses a bird's-eye view to generate an initial path, estimates the height of the building through a SLAM (simultaneous localization and mapping) framework, and takes close-up photos to reveal the building details, minimizing the path length at both the inter-building and intra-building levels. These UAV path planning research works aim to better and faster reconstruct complete and accurate dense models.

However, these methods do not consider how to collect high-quality image data for texture reconstruction, resulting in the inability to achieve ideal texture effects when the collected image data is used for texture mapping. When the collected images are used for texture reconstruction, due to the rich straight line features and clear component structures on the surface of urban buildings, the texture is very prone to cognitive dissonance problems, such as inconsistent perspective relationships of windows on a single building facade.

Therefore, the existing technology still needs to be improved and developed.

Summary of the invention

The technical problem to be solved by the present invention is that, in view of the above-mentioned defects of the prior art, a method, device, terminal and storage medium for drone path planning are provided, aiming to solve the problem that the prior art does not consider how to collect high-quality image data for texture reconstruction, resulting in the collected image data not being able to achieve the ideal texture effect when used for texture mapping.

The technical solution adopted by the present invention to solve the problem is as follows:

In a first aspect, an embodiment of the present invention provides a method for planning a path for an unmanned aerial vehicle, wherein the method comprises:

Acquire a two-dimensional base map and a safe flight altitude of a target scene, and determine a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude;

Based on the plane quality and the viewpoint-plane pair quality, an initial viewpoint set is generated according to the candidate viewpoint set, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane;

Using a multi-objective optimization algorithm to optimize the initial viewpoint set to determine an optimized viewpoint set;

The drone shooting path is determined according to the candidate viewpoint set and the optimized viewpoint set.

In an implementation method, the generating an initial viewpoint set according to the candidate viewpoint set based on the plane quality and the viewpoint-plane pair quality includes:

Constructing a plurality of viewpoint plane pairs based on the candidate viewpoint set;

Setting an initial sight direction for a viewpoint in each of the viewpoint plane pairs based on the plane quality;

The viewpoint-plane pair quality corresponding to each of the viewpoint-plane pairs is calculated, and the initial viewpoint set is determined according to the viewpoint-plane pair quality.

In an implementation method, setting an initial sight line direction for a viewpoint in each viewpoint plane pair based on the plane quality includes:

Sampling within a 180° range centered on the reverse direction of the normal direction of the midpoint plane to determine a number of candidate sight lines;

The candidate sight line direction corresponding to the maximum plane quality is selected as the initial sight line direction of the viewpoint centered on the viewpoint plane.

In one implementation method, the method for calculating the plane quality includes:

,

in, and Represent the perspective quality and image quality respectively, and is the weight;

Perspective quality Defined as:

,

in, and is the weight, is the viewpoint set The number of viewpoints, Metric Unit Vector and The distance between express Middle Viewpoint The direction of sight, express The average viewing direction of the midpoint, Representation plane The normal vector of

Image Quality Defined as:

,

in, and is the weight, Represents line of sight distance The normalized value, express The sight distance, express The average value of the sight distance of the midpoint, , represent Mid-viewpoint coverage plane The width of the area, represent Width.

In one implementation method, the method for calculating the viewpoint plane pair quality includes:

,

in, and Represent the perspective quality and image quality respectively, and is the weight;

Perspective quality Defined as:

,

and is the weight, Metric Unit Vector and The distance between Indicates viewpoint The direction of sight, Represents a viewpoint set The average viewing direction of the midpoint, Representation plane The normal vector of

Image Quality Defined as:

,

and is the weight, Represents line of sight distance The normalized value, Indicates viewpoint The sight distance, , represent Mid-viewpoint coverage plane The width of the area, Representative plane The width of Representative plane Only the viewpoint The proportion of the area where the observations were made.

In one implementation method, the adopting a multi-objective optimization algorithm to optimize the initial viewpoint set to determine the optimized viewpoint set includes:

Sampling the position and sight direction of each viewpoint in the neighborhood of the initial viewpoint set to obtain candidate viewpoint poses corresponding to each viewpoint;

Determine a target viewpoint pose among the candidate viewpoint poses based on a multi-objective optimization algorithm;

The initial viewpoint set is screened according to the target viewpoint pose corresponding to each viewpoint to determine the optimized viewpoint set.

In one implementation method, determining the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set includes:

Constructing a fully connected graph according to the candidate viewpoint set and the optimized viewpoint set;

The minimum cost path corresponding to the fully connected graph is solved based on the traveling salesman problem as the drone shooting path.

In a second aspect, an embodiment of the present invention further provides a drone path planning device, wherein the drone path planning device comprises:

A data acquisition module, used to acquire a two-dimensional base map and a safe flight altitude of a target scene, and determine a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude;

A viewpoint generation module, configured to generate an initial viewpoint set according to the candidate viewpoint set based on a plane quality and a viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane;

A viewpoint optimization module, used to optimize the initial viewpoint set using a multi-objective optimization algorithm to determine an optimized viewpoint set;

A path determination module is used to determine the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set.

In a third aspect, an embodiment of the present invention further provides a terminal, comprising a memory and one or more processors; the memory stores one or more programs; the program comprises instructions for executing any of the above-described drone path planning methods; and the processor is used to execute the program.

In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium on which a plurality of instructions are stored, wherein the instructions are suitable for being loaded and executed by a processor to implement any of the above-mentioned drone path planning methods.

Beneficial effects of the present invention: The embodiments of the present invention obtain a two-dimensional base map and a safe flight altitude of the target scene, and determine a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude; generate an initial viewpoint set according to the candidate viewpoint set based on the plane quality and the viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of the viewpoint-plane pair constructed by a single viewpoint and a plane; optimize the initial viewpoint set using a multi-objective optimization algorithm to determine the optimized viewpoint set; determine the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set. The present invention obtains a viewpoint set capable of reconstructing high-quality textures by introducing plane quality and viewpoint-plane quality, thereby effectively solving the problem that the prior art does not consider how to collect high-quality image data for texture reconstruction, resulting in the inability to achieve ideal texture effects when the collected image data is used for texture mapping.

BRIEF DESCRIPTION OF THE DRAWINGS

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

FIG1 is a schematic flow chart of a method for UAV path planning according to an embodiment of the present invention.

FIG. 2 is a diagram showing the generation of a plane dot matrix and the division of downward viewpoints provided by an embodiment of the present invention.

FIG. 3 is a schematic diagram of all planes observable from viewpoints provided by an embodiment of the present invention.

FIG. 4 is a schematic diagram of all viewpoints from which a plane can be observed provided by an embodiment of the present invention.

FIG. 5 is a schematic diagram of a downward viewpoint selection result provided by an embodiment of the present invention.

FIG. 6 is a schematic diagram of an initial viewpoint optimization result provided by an embodiment of the present invention.

FIG. 7 is a schematic diagram of a texture mapping result provided by an embodiment of the present invention.

FIG8 is a schematic diagram of the internal modules of the UAV path planning device provided in an embodiment of the present invention.

FIG. 9 is a functional block diagram of a terminal provided in an embodiment of the present invention.

DETAILED DESCRIPTION

The present invention discloses a method, device, terminal and storage medium for planning a path for a drone. To make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and/or" used herein includes all or any unit and all combinations of one or more associated listed items.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as herein.

The current UAV acquisition method mainly aims at high-quality geometric reconstruction, and does not consider the acquisition of high-quality image data for texture reconstruction. Therefore, the acquired image data cannot achieve ideal texture effects when used for texture mapping.

In view of the above-mentioned defects of the prior art, the present invention provides a method, device, terminal and storage medium for drone path planning, wherein the method obtains a two-dimensional base map and a safe flight altitude of the target scene, determines a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude; generates an initial viewpoint set according to the candidate viewpoint set based on the plane quality and the viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of the viewpoint-plane pair constructed by a single viewpoint and a plane; optimizes the initial viewpoint set using a multi-objective optimization algorithm to determine an optimized viewpoint set; and determines the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set. The present invention obtains a viewpoint set capable of reconstructing high-quality textures by introducing plane quality and viewpoint-plane quality, thereby effectively solving the problem that the prior art does not consider how to collect high-quality image data for texture reconstruction, resulting in the inability to achieve an ideal texture effect when the collected image data is used for texture mapping.

Exemplary Methods

As shown in FIG1 , the method comprises:

Step S100, obtaining a two-dimensional base map and a safe flight altitude of a target scene, and determining a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude;

The two-dimensional base map is a two-dimensional top-down base map corresponding to the target scene. In the two-dimensional top-down base map, each contour polygon represents a building, and the edge of the contour polygon represents the building facade. The safe flight altitude is the lowest safe flight altitude of the drone. In this embodiment, at a fixed safe flight altitude H, a plane dot matrix completely covering the target scene is generated based on the two-dimensional base map. , the plane lattice It is used as a candidate viewpoint set for subsequent viewpoint screening.

As shown in FIG1 , the method further includes:

Step S200, based on the plane quality and the viewpoint-plane pair quality, generate an initial viewpoint set according to the candidate viewpoint set, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane.

Specifically, the plane is a building facade, and this embodiment defines two quality indicators: plane quality and viewpoint plane quality. Among them, the plane quality measures the final texture quality of the plane based on a set of two-dimensional viewpoints, and the viewpoint plane pair quality measures the quality of the viewpoint plane pair constructed by the two-dimensional viewpoint and the plane. These two qualities are used to guide the selection of viewpoints. Based on the plane quality and the viewpoint plane quality, this embodiment generates an initial viewpoint set according to the candidate viewpoint set, so that the initial viewpoint set can maximize the coverage of the building facade in the target scene, and can ensure that the viewpoints in the initial viewpoint set have high perspective consistency and high image quality.

In one implementation, plane quality is used to measure the The texture quality, given a set of viewpoints , the calculation method of the plane quality includes:

,

in, and Represent the perspective quality and image quality respectively, and is the weight;

Perspective quality is used to measure The consistency of the sight direction and the alignment of the sight direction of the midpoint The degree of perspective quality Defined as:

,

in, and are the weights of the line of sight consistency and the line of sight facing item, yes The number of viewpoints, Metric Unit Vector and The distance between express Middle Viewpoint The direction of sight, express The average viewing direction of the midpoint, Representation plane The normal vector of

Image quality measures the image quality of a flat texture, including the high definition of the image material, the consistency of the image resolution, the completeness of the image coverage plane, and the indispensability of the image coverage plane. ,measure Image quality is mainly measured Relative to resolution consistency and coverage.

Image Quality Defined as:

,

in, and are the weights of resolution consistency and coverage, yes The number of viewpoints, Represents line of sight distance The normalized value, express The sight distance, express The average value of the sight distance of the midpoint, , represent Mid-viewpoint coverage plane The width of the area, represent Width.

In one implementation, given Single viewpoint ,measure With plane The constructed viewpoint plane pair. The method for calculating the quality of the viewpoint plane pair includes:

,

in, and Represent the perspective quality and image quality respectively, and is the weight;

Perspective quality measurement The sight direction and the average sight direction consistency, and Directly opposite The degree of perspective quality Defined as:

,

and is the weight, Metric Unit Vector and The distance between Indicates viewpoint The direction of sight, Represents a viewpoint set The average viewing direction of the midpoint, Representation plane The normal vector of

Image quality mainly considers the high definition, coverage and indispensability of the viewpoint relative to the plane. Image quality Defined as:

,

and is the weight, Represents line of sight distance The normalized value, Indicates viewpoint The sight distance, , represent Mid-viewpoint coverage plane The width of the area, Representative plane The width of Representative plane Only the viewpoint The proportion of the area where the observations were made.

In one implementation, the generating an initial viewpoint set according to the candidate viewpoint set based on the plane quality and the viewpoint-plane pair quality includes:

Step S201, constructing a plurality of viewpoint plane pairs based on the candidate viewpoint set;

Step S202, setting an initial sight line direction for each viewpoint in the viewpoint plane pair based on the plane quality;

Step S203: Calculate the viewpoint-plane pair quality corresponding to each viewpoint-plane pair, and determine the initial viewpoint set according to the quality of each viewpoint-plane pair.

Specifically, the steps of generating an initial viewpoint set based on a candidate viewpoint set include: (1) constructing viewpoint plane pairs; (2) initializing viewpoint directions; and (3) selecting downward viewpoints.

(1) Viewpoint plane pair construction. A method for constructing a plurality of viewpoint plane pairs based on a candidate viewpoint set includes: At each viewing angle position in , five drone shooting angles are set, including a vertical downward angle and four oblique angles; based on the candidate viewpoint set , the drone is designed to start from the height H and vertically dive to generate a downward viewpoint for photographing the building facade at a low height. As shown in Figure 2, according to the minimum distance from the viewpoint to the plane polygon and the threshold of the safety distance, The viewpoints in the image are divided into viewpoints that can be explored and viewpoints that cannot be explored. For each viewpoint that can be explored, the plane that can be observed by the viewpoint is detected, and the viewpoint-plane pair is constructed, as shown in Figures 3 and 4. , all observed The viewpoint, composition The downward viewpoint set In one implementation, a non-descendable viewpoint and a plane may be used to form a viewpoint-plane pair.

(2) Viewpoint direction initialization. After constructing the viewpoint plane pair, the sight direction of the viewpoint in the viewpoint plane pair is initialized, and an initial sight direction is selected for each viewpoint. The steps of viewpoint direction initialization include: The reverse direction of the normal direction is uniformly sampled within a 180° range as the center to determine several candidate sight lines. ; Select the candidate sight line direction corresponding to the maximum plane quality as the initial sight line direction of the viewpoint in the viewpoint plane The viewpoint direction initialization step further includes assigning the same viewpoint direction to each viewpoint before determining the candidate viewpoint direction.

(3) Downward viewpoint selection. Downward viewpoint selection is used to delete redundant viewpoints in the viewpoint set corresponding to the plane. This embodiment adopts an iterative deletion strategy. In each round of iteration, the quality of the viewpoint plane pair corresponding to each viewpoint plane pair is first calculated. , Viewpoint The quality of the viewpoint The corresponding plane The sum of the masses of the constructed viewpoint plane pairs is calculated as:

,

In each round of iteration, the viewpoint with the lowest quality value is deleted until there are no redundant viewpoints, as shown in FIG5 , then the iteration is terminated and the remaining viewpoints are determined as the initial viewpoint set.

As shown in FIG1 , the method further includes:

Step S300: optimizing the initial viewpoint set using a multi-objective optimization algorithm to determine an optimized viewpoint set.

For the viewpoints in the initial viewpoint set, a multi-objective optimization algorithm is used to optimize the position and direction of each viewpoint, thereby optimizing the plane quality and reducing the time cost of drone image acquisition. In this embodiment, the optimization target item of the multi-objective optimization algorithm includes the number C of three-dimensional hovering points required for the entire target scene. In addition, in order to improve the texture quality of the building facade, the optimization target also includes the viewpoint Observable building facades The corresponding plane quality The objective items of multi-objective optimization are:

,

The method for calculating the number of three-dimensional hovering points is as follows: for a given viewpoint (two-dimensional downward viewpoint) and plane ,exist The position is uniformly sampled vertically from top to bottom to generate a series of vertical three-dimensional viewpoints ,in, represents the position of the 3D viewpoint, Represents the viewing direction of a 3D viewpoint. Ensure that the building facade is fully covered with a certain overlap in the height direction. The two-dimensional probe points at the same position can have different sight directions and are used to observe different building facades. The vertical 3D viewpoints corresponding to the position of can be fine-tuned in the vertical position to merge the viewpoints that are closer in the vertical direction. After the vertical merging, the number of 3D hovering points of the entire target scene can be calculated.

In one implementation, the adopting a multi-objective optimization algorithm to optimize the initial viewpoint set to determine the optimized viewpoint set includes:

Step S301, sampling the position and sight direction of each viewpoint in the neighborhood of the initial viewpoint set to obtain candidate viewpoint poses corresponding to each viewpoint;

Step S302, determining a target viewpoint pose among the candidate viewpoint poses based on a multi-objective optimization algorithm;

Step S303: Screening the initial viewpoint set according to the target viewpoint pose corresponding to each viewpoint to determine the optimized viewpoint set.

Specifically, this embodiment adopts the following iterative optimization process to achieve the goal of minimizing the multi-objective optimization algorithm. In each round of iteration, first, at the viewpoint In the neighborhood of , uniform sampling of positions and sight directions is performed to obtain candidate positions and sight directions, and the candidate positions and sight directions are combined as candidate viewpoint poses; all candidate combinations are searched through a multi-objective optimization algorithm to obtain the optimal viewpoint position and sight direction combination as the optimized viewpoint pose, i.e., the target viewpoint pose. Secondly, based on the target viewpoint pose and the viewpoint plane pair quality, the redundancy of other viewpoints is re-judged and redundant viewpoints are deleted. When all the downward viewpoints no longer change, the optimization is judged to have converged, and the optimized viewpoint set is determined, as shown in Figure 6.

As shown in FIG1 , the method further includes:

Step S400: determining a drone shooting path according to the candidate viewpoint set and the optimized viewpoint set.

In this embodiment, a total drone viewpoint set is formed by a candidate viewpoint set and an optimized viewpoint set. A path is generated based on the total drone viewpoint set to obtain a drone shooting path.

In one implementation, determining the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set includes:

Step S401, constructing a fully connected graph according to the candidate viewpoint set and the optimized viewpoint set;

Step S402: solving the minimum cost path corresponding to the fully connected graph based on the traveling salesman problem as the drone shooting path.

In order to reduce the time cost of drone image acquisition, this embodiment constructs a fully connected graph based on the candidate viewpoint set and the optimized viewpoint set and generates a drone flight path that connects all viewpoints in series. By solving the traveling salesman problem, a path that traverses all nodes and has the minimum cost is obtained. In this embodiment, the viewpoints in the candidate viewpoint set and the optimized viewpoint set are used as nodes in the graph to construct a fully connected graph, and the method of solving the traveling salesman problem is used to solve the minimum cost path of the fully connected graph.

Connecting viewpoint pairs in a fully connected graph The cost value of the edge The definition is as follows:

,

in, Representative Viewpoint To Viewpoint The shortest straight-line distance in the safe area, represent and The angle of sight direction, Represents the weight value. The path with the minimum cost in the fully connected graph is the optimal drone shooting path that traverses all viewpoints.

The scene reconstruction step includes structured geometry reconstruction and texture reconstruction. Photo collection is performed based on the generated drone shooting path for scene geometry and texture reconstruction. In the geometry reconstruction step, ContextCapture software is first used to reconstruct a high-precision dense geometry model based on the collected photos. Secondly, a structure-aware reconstruction algorithm is used to obtain a structured model of the dense geometry model. Finally, the TwinTex method (geometry-aware texture generation method for abstract three-dimensional architectural models) is used to generate high-quality texture maps for the structured model, as shown in Figure 7.

By conducting experiments on urban scenes of different complexity, when the image material collected by the drone shooting path obtained by the method of this embodiment is used for texture mapping of the structured model, the obtained result has several advantages: (1) The viewpoint set corresponding to the same building facade has a more consistent perspective and faces the plane. Therefore, when the collected image material is used for texture mapping of the structured model, a texture with consistent perspective and facing the plane can be generated; (2) The viewpoint corresponding to the building facade has a high-definition feature. Therefore, when the collected image material is used for texture mapping, a high-definition texture can be generated; (3) The generated texture has fewer splicing marks, blur and deformation.

Based on the above embodiments, the present invention further provides a UAV path planning device, as shown in FIG8 , the device includes:

Data acquisition module 01, used to acquire a two-dimensional base map and a safe flight altitude of a target scene, and determine a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude;

A viewpoint generation module 02, configured to generate an initial viewpoint set according to the candidate viewpoint set based on a plane quality and a viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane;

A viewpoint optimization module 03 is used to optimize the initial viewpoint set by using a multi-objective optimization algorithm to determine an optimized viewpoint set;

The path determination module 04 is used to determine the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set.

Based on the above embodiments, the present invention also provides a terminal, whose principle block diagram can be shown in Figure 9. The terminal includes a processor, a memory, a network interface, and a display screen connected through a system bus. Among them, the processor of the terminal is used to provide computing and control capabilities. The memory of the terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the terminal is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for drone path planning is implemented. The display screen of the terminal can be a liquid crystal display or an electronic ink display.

Those skilled in the art will understand that the principle block diagram shown in FIG9 is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the terminal to which the solution of the present invention is applied. The specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

In one implementation, the terminal has one or more programs stored in its memory, and is configured to be executed by one or more processors, wherein the one or more programs include instructions for performing a method for planning a drone path.

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and/or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

In summary, the present invention discloses a method, device, terminal and storage medium for drone path planning, wherein the method obtains a two-dimensional base map and a safe flight altitude of a target scene, determines a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude; generates an initial viewpoint set according to the candidate viewpoint set based on the plane quality and the viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of the viewpoint-plane pair constructed by a single viewpoint and a plane; optimizes the initial viewpoint set using a multi-objective optimization algorithm to determine an optimized viewpoint set; and determines the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set. The present invention obtains a viewpoint set capable of reconstructing high-quality textures by introducing plane quality and viewpoint-plane quality, thereby effectively solving the problem that the prior art does not consider how to collect high-quality image data for texture reconstruction, resulting in the inability to achieve an ideal texture effect when the collected image data is used for texture mapping.

It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims (10)

1. A method for unmanned aerial vehicle path planning, characterized in that the method comprises: Acquire a two-dimensional base map and a safe flight altitude of a target scene, and determine a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude; Based on the plane quality and the viewpoint-plane pair quality, an initial viewpoint set is generated according to the candidate viewpoint set, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane; Using a multi-objective optimization algorithm to optimize the initial viewpoint set to determine an optimized viewpoint set; The drone shooting path is determined according to the candidate viewpoint set and the optimized viewpoint set. 2. The method for UAV path planning according to claim 1, characterized in that the generating of the initial viewpoint set according to the candidate viewpoint set based on the plane quality and the viewpoint plane pair quality comprises: Constructing a plurality of viewpoint plane pairs based on the candidate viewpoint set; Setting an initial sight direction for a viewpoint in each of the viewpoint plane pairs based on the plane quality; The viewpoint-plane pair quality corresponding to each of the viewpoint-plane pairs is calculated, and the initial viewpoint set is determined according to the viewpoint-plane pair quality. 3. The method for UAV path planning according to claim 2, characterized in that the step of setting an initial sight line direction for each viewpoint in each viewpoint plane pair based on the plane quality comprises: Sampling within a 180° range centered on the reverse direction of the normal direction of the midpoint plane to determine a number of candidate sight lines; The candidate sight line direction corresponding to the maximum plane quality is selected as the initial sight line direction of the viewpoint centered on the viewpoint plane. 4. The UAV path planning method according to claim 1, wherein the method for calculating the plane quality comprises: , in, and Represent the perspective quality and image quality respectively, and is the weight; Perspective quality Defined as: , in, and is the weight, is the viewpoint set The number of viewpoints, Metric Unit Vector and The distance between express Middle Viewpoint The direction of sight, express The average viewing direction of the midpoint, Representation plane The normal vector of Image Quality Defined as: , in, and is the weight, Represents line of sight distance The normalized value, express The sight distance, express The average value of the sight distance of the midpoint, , represent Mid-viewpoint coverage plane The width of the area, represent Width. 5. The UAV path planning method according to claim 1, characterized in that the method for calculating the viewpoint plane pair quality comprises: , in, and Represent the perspective quality and image quality respectively, and is the weight; Perspective quality Defined as: , and is the weight, Metric Unit Vector and The distance between Indicates viewpoint The direction of sight, Represents a viewpoint set The average viewing direction of the midpoint, Representation plane The normal vector of Image Quality Defined as: , and is the weight, Represents line of sight distance The normalized value, Indicates viewpoint The sight distance, , represent Mid-viewpoint coverage plane The width of the area, Representative plane The width of Representative plane Only the viewpoint The proportion of the area where the observations were made. 6. The method for UAV path planning according to claim 1, characterized in that the step of optimizing the initial viewpoint set by using a multi-objective optimization algorithm to determine the optimized viewpoint set comprises: Sampling the position and sight direction of each viewpoint in the neighborhood of the initial viewpoint set to obtain candidate viewpoint poses corresponding to each viewpoint; Determine a target viewpoint pose among the candidate viewpoint poses based on a multi-objective optimization algorithm; The initial viewpoint set is screened according to the target viewpoint pose corresponding to each viewpoint to determine the optimized viewpoint set. 7. The method for drone path planning according to claim 1, wherein determining the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set comprises: Constructing a fully connected graph according to the candidate viewpoint set and the optimized viewpoint set; The minimum cost path corresponding to the fully connected graph is solved based on the traveling salesman problem as the drone shooting path. 8. A drone path planning device, characterized in that the device comprises: A data acquisition module, used to acquire a two-dimensional base map and a safe flight altitude of a target scene, and determine a candidate viewpoint set according to the two-dimensional base map and the safe flight altitude; A viewpoint generation module, configured to generate an initial viewpoint set according to the candidate viewpoint set based on a plane quality and a viewpoint-plane pair quality, wherein the plane quality is used to measure the texture quality of a single plane, and the viewpoint-plane pair quality is used to measure the quality of a viewpoint-plane pair constructed by a single viewpoint and a plane; A viewpoint optimization module, used to optimize the initial viewpoint set using a multi-objective optimization algorithm to determine an optimized viewpoint set; A path determination module is used to determine the drone shooting path according to the candidate viewpoint set and the optimized viewpoint set. 9. A terminal, characterized in that the terminal includes a memory and one or more processors; the memory stores one or more programs; the program contains instructions for executing the drone path planning method as described in any one of claims 1-7; and the processor is used to execute the program. 10. A computer-readable storage medium having a plurality of instructions stored thereon, wherein the instructions are suitable for being loaded and executed by a processor to implement the steps of the drone path planning method described in any one of claims 1 to 7.