Multi-rotor unmanned aerial vehicle coverage path planning method and system

Abstract

The application discloses a multi-rotor unmanned aerial vehicle coverage path planning method and system, and the method comprises the steps of constructing a coverage path of a multi-rotor unmanned aerial vehicle to a polygonal region; constructing a plurality of evaluation indexes for the coverage path, and performing weighted summation on all the evaluation indexes to obtain a total evaluation index function; based on a biological heuristic method, the total evaluation index function is optimized and solved to obtain a coverage direction, a flight height and a path point corresponding to each flight state of the multi-rotor unmanned aerial vehicle when the multi-rotor unmanned aerial vehicle executes a coverage task; and based on the optimization and solving result, the multi-rotor unmanned aerial vehicle starts to execute the coverage task. The method can plan a full coverage path for any specified polygonal region, the coverage range of adjacent straight line paths has a certain overlap degree, and a certain margin is reserved on the boundary to improve the quality of image modeling in the later period. The biological heuristic method is used to optimize the coverage direction and the flight height, so that the time for executing the task is shortened, the energy consumption for executing the task is reduced, and the quality of mapping is improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) path planning technology, and in particular to a method and system for planning the coverage path of multi-rotor UAVs. Background Technology

[0002] The statements in this section merely refer to the background art related to this invention and do not necessarily constitute prior art.

[0003] Multi-rotor drones, with their wide field of view, high flexibility, good adaptability, and low operating costs, are widely used in surveying and mapping, key area monitoring, and search and rescue. In these applications, multi-rotor drones are required to carry visual cameras to perform full-coverage scanning of large designated areas. This necessitates that the drones be able to plan reasonable scanning paths in advance to achieve requirements such as short mission completion time, low energy consumption, and high mapping quality.

[0004] Traditional coverage path planning methods, including the "ox-plowing" decomposition method and the grid-based method, all handle coverage path planning for arbitrary polygonal regions by segmentation. However, the selection of coverage direction is relatively arbitrary, and each sub-region is planned separately after segmentation, resulting in fragmented paths. For UAV mapping applications, too many segments and fragmented paths are very detrimental to subsequent image stitching, and too many turns also cause excessive energy consumption. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a multi-rotor UAV coverage path planning method and system. It can perform full-coverage path planning for specified arbitrary polygonal regions, including concave polygonal regions. The scanning path adopts a serpentine, zigzag pattern, with a certain degree of overlap in the coverage areas of adjacent straight paths, and retains a certain margin at the boundaries to improve the quality of subsequent image modeling. Based on bio-inspired methods, the coverage direction and flight altitude are optimized, resulting in varying degrees of optimization of the total length of the coverage path, the number of turns, the area coverage rate, and image quality indicators. Therefore, it can shorten task execution time, reduce task energy consumption, and improve mapping quality.

[0006] In a first aspect, the present invention provides a method for planning coverage paths for multi-rotor unmanned aerial vehicles;

[0007] Multi-rotor UAV coverage path planning methods include:

[0008] Construct the coverage path of a multi-rotor UAV over a polygonal region;

[0009] For the coverage path, several evaluation indicators are constructed, and all evaluation indicators are weighted and summed to obtain the total evaluation indicator function;

[0010] Based on a bio-inspired approach, the overall evaluation index function is optimized to obtain the optimal coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs a coverage task. Based on the optimization results, the multi-rotor UAV begins to perform the coverage task.

[0011] Secondly, this invention provides a multi-rotor unmanned aerial vehicle (UAV) coverage path planning system;

[0012] A multi-rotor UAV coverage path planning system, including:

[0013] The path building module is configured to: build a coverage path for a multi-rotor UAV over a polygonal region;

[0014] The indicator construction module is configured to: construct several evaluation indicators for the coverage path, and perform a weighted summation of all evaluation indicators to obtain the total evaluation indicator function;

[0015] The optimization solution module is configured to: optimize the overall evaluation index function based on a bio-inspired approach to obtain the optimal coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs the coverage task; based on the optimization solution results, the multi-rotor UAV begins to perform the coverage task.

[0016] Compared with the prior art, the beneficial effects of the present invention are:

[0017] This invention adopts a serpentine scanning path pattern, with a certain degree of overlap in the coverage of adjacent straight paths, which meets the needs of subsequent stitching and mapping.

[0018] The two endpoints of the straight path in this invention are determined by taking the upper and lower boundaries after solving a series of key point sets, thus satisfying the coverage requirements of concave polygon regions.

[0019] The present invention retains a certain coverage margin at the end of each straight path and on both sides in the scanning direction, thus satisfying the requirement of full coverage;

[0020] This invention integrates four evaluation indicators: total length of the coverage path, number of bends in the coverage path, area coverage, and ground resolution, to construct a general objective function. The optimal coverage direction and flight altitude are solved using a bio-inspired method, thus ensuring the quality of the planned path.

[0021] The bio-inspired method of this invention is the longhorn beetle whisker search optimization algorithm, which globally searches for the optimal solution in the decision space. It uses a single search agent, which consumes low computational resources and has high solution efficiency. Attached Figure Description

[0022] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0023] Figure 1 This is a schematic diagram showing the relationship between the flight altitude, scanning width, ground resolution, and camera parameters of the multi-rotor UAV in Embodiment 1 of the present invention.

[0024] Figure 2 This is a schematic diagram of the method for constructing the coverage path according to Embodiment 1 of the present invention;

[0025] Figure 3 This is a schematic diagram of the final planned full-coverage path according to Embodiment 1 of the present invention;

[0026] Figures 4(a)-4(d) This is a schematic diagram of the search process in the decision space of the beetle whisker search optimizer according to Embodiment 1 of the present invention;

[0027] Figure 5 This is a schematic diagram of the search trajectory of the longhorn beetle whisker search optimizer in the decision space according to Embodiment 1 of the present invention.

[0028] Figure 6 This is a complete algorithm flowchart of Embodiment 1 of the present invention. Detailed Implementation

[0029] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0030] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0032] All data acquisition in this embodiment is carried out in accordance with laws and regulations and with user consent, and the data is used legally.

[0033] Example 1

[0034] This embodiment provides a method for planning coverage paths for multi-rotor unmanned aerial vehicles (UAVs).

[0035] Multi-rotor UAV coverage path planning methods include:

[0036] S101: Construct the coverage path of a multi-rotor UAV over a polygonal region;

[0037] S102: For the coverage path, construct several evaluation indicators, and perform a weighted summation of all evaluation indicators to obtain the total evaluation indicator function;

[0038] S103: Based on a bio-inspired approach, the overall evaluation index function is optimized to obtain the optimal coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs the coverage task; based on the optimization results, the multi-rotor UAV begins to perform the coverage task.

[0039] Further, S101: Constructing the coverage path of the multi-rotor UAV over the polygonal region specifically includes:

[0040] S101-1: Determine the flight altitude of the multi-rotor UAV based on camera parameters and the set ground resolution;

[0041] S101-2: Based on camera parameters and the flight altitude of the multi-rotor UAV, determine the width of the coverage path during the flight of the multi-rotor UAV; the coverage path is a rectangular strip area;

[0042] S101-3: Assuming that the multi-rotor UAV scans the polygonal region in a serpentine manner, the spacing between the center lines of adjacent coverage paths is determined based on the width of the coverage path and the overlap between adjacent coverage paths.

[0043] S101-4: Determine the distribution of the coverage path in the scanning direction based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths.

[0044] S101-5: Based on the coverage margin set in the coverage direction during mapping, determine the two endpoints of each coverage path in the coverage direction.

[0045] S101-6: Connect the endpoints of all coverage paths in sequence using a serpentine zigzag method, then connect the starting point of the coverage path to the take-off point of the multi-rotor UAV, and connect the ending point of the coverage path to the landing point of the multi-rotor UAV to obtain a complete coverage path.

[0046] For example, see Appendix Figure 1As shown, step S101-1: Determining the flight altitude of the multi-rotor UAV based on camera parameters and the set ground resolution, specifically includes:

[0047] The flight altitude h is determined by the camera's pixel size I. pix Focal length f cam and the set ground resolution range [GSD] min GSD max And determined:

[0048]

[0049] For example, see Appendix Figure 1 As shown, step S101-2: Based on camera parameters and the flight altitude of the multi-rotor UAV, determine the width of the coverage path during the flight of the multi-rotor UAV; the coverage path is a rectangular strip area, specifically including:

[0050] Width of the covered path fov x The size of the camera's image sensor I x Focal length f cam Determined by the flight altitude h:

[0051]

[0052] For example, see Appendix Figure 2 As shown, S101-3: Assuming the multi-rotor UAV performs path scanning of the polygonal region in a serpentine manner, the spacing between the centerlines of adjacent coverage paths is determined based on the width of the coverage path and the overlap between adjacent coverage paths, specifically including:

[0053] The spacing d between adjacent straight paths is determined by the path's coverage width fov. x The overlap ov required for mapping x And it is determined that:

[0054] d=fov x *(1-ov x ).

[0055] Further, S101-4: Based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths, the distribution of the coverage path in the scanning direction is determined, specifically including:

[0056] Construct a Cartesian coordinate system, where the horizontal axis of the Cartesian coordinate system represents the scanning direction of the multi-rotor UAV, and the vertical axis of the Cartesian coordinate system represents the coverage direction of the multi-rotor UAV.

[0057] Project all vertices of the polygonal region onto the scanning direction to obtain the coordinates of the left and right boundary points of the polygonal region.

[0058] The distribution of the coverage path in the scanning direction is determined based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths.

[0059] For example, see Appendix Figure 2 As shown, step S101-4: Based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths, the distribution of the coverage path in the scanning direction is determined, specifically including:

[0060] Coverage path in the scanning direction Distribution on {c i}, consisting of all vertices of the polygon in the scanning direction Projecting onto the boundary point PG, find the boundary point PG. left and PG right Then, based on the coverage margin requirement δ needed for mapping... x And determine:

[0061]

[0062]

[0063] c i =c i-1 +d, i∈[2,n]

[0064] Where n represents the total number of covered paths, c i c represents the center point of the i-th coverage path in the scanning direction. i-1 c1 represents the center point of the (i-1)th coverage path in the scanning direction, and c1 represents the center point of the 1st coverage path in the scanning direction. [·] ceil This indicates rounding the value up, and ||·|| indicates finding the length of the vector.

[0065] Further, in step S101-5: based on the coverage margin set in the coverage direction during mapping, determine the two endpoints of each coverage path in the coverage direction:

[0066] First, find the intersection point of the coverage width edge of the coverage path with the edge of the polygonal region. Then, find the vertices of the polygonal region within the coverage range of the coverage path. Next, project the intersection point and the vertices in the coverage direction to obtain the key point set. Finally, based on the key point set and the set coverage margin in the coverage direction, determine the two endpoints of each coverage path.

[0067] For example, S101-5: determining the two endpoints of each coverage path in the coverage direction specifically includes:

[0068] See appendix Figure 2 As shown, the endpoints y of a single straight path min and y max The intersection of the coverage width edge of a single straight path and the edge of a polygon, and the vertices of the polygon within the coverage area of ​​the single straight path, are used in the coverage direction. After projection, the key point set {y} is formed. j After taking the upper and lower boundaries, the allowance in the coverage direction is superimposed, requiring δ. y , deducting field of view fov y / 2, thus determining the two endpoints of a single straight path:

[0069]

[0070]

[0071] For example, S101-6: Connect the endpoints {p} of all covered paths sequentially in a serpentine zigzag manner. i |i=1,…2n}, then connect the starting point p1 of the coverage path to the takeoff point of the multi-rotor UAV, and connect the ending point p of the coverage path to the takeoff point of the multi-rotor UAV. 2n By connecting with the landing point of the multi-rotor drone, a complete coverage path is obtained, specifically including:

[0072] See appendix Figure 3 As shown, connecting the endpoints of all straight paths in a serpentine pattern, along with the takeoff and landing points S of the multi-rotor UAV, constitutes a complete path. Notably, the takeoff and landing points can also be different locations in space.

[0073] Further, in step S102: for the coverage path, several evaluation indicators are constructed, and all evaluation indicators are weighted and summed to obtain a total evaluation indicator function, specifically including:

[0074] S102-1: Calculate the total length of the complete coverage path, calculate the quotient of the total length of the complete coverage path and the perimeter of the polygonal region, and use the quotient as the first dimensionless evaluation index.

[0075] S102-2: Calculate the number of bends in the complete coverage path, use principal component analysis algorithm to obtain the maximum and minimum number of bends, and normalize the number of bends in the complete coverage path using the maximum and minimum number of bends as the second dimensionless evaluation index;

[0076] S102-3: Calculate the area coverage rate of the complete coverage path and use the area coverage rate as the third dimensionless evaluation index;

[0077] S102-4: Calculate the ground resolution and normalize it using the minimum and maximum ground resolutions as the fourth dimensionless evaluation index.

[0078] S102-5: The first, second, third and fourth dimensionless evaluation indices are weighted and summed to obtain the overall evaluation index function.

[0079] For example, S102-1: Calculate the total length of the complete coverage path, calculate the quotient of the total length of the complete coverage path and the perimeter of the polygonal region, and use the quotient as the first dimensionless evaluation index, specifically including:

[0080] Calculate the total path length L of the covered path:

[0081]

[0082] The total path length L of the above covered path is equal to the sum of the distance from the starting point to the first path point, the distance from the ending point to the last path point, and the length of all serpentine paths.

[0083] The total path length of the coverage path is divided by the perimeter of the polygonal region to obtain a dimensionless evaluation index.

[0084] For example, S102-2: Calculate the number of bends in the complete coverage path, use principal component analysis to obtain the maximum and minimum number of bends, and normalize the number of bends as a second dimensionless evaluation index, including:

[0085] Calculate the number of bends in the coverage path:

[0086] N = 2*n,

[0087] Where n represents the number of covered paths, and N is the maximum number of bends obtained from principal component analysis of the polygon vertex set. max and minimum number of bends N min To standardize:

[0088]

[0089] This can be transformed into a dimensionless evaluation index.

[0090] For example, S102-3: Calculating the area coverage rate of the complete coverage path, and using the area coverage rate as the third dimensionless evaluation index, specifically includes:

[0091] Calculate the area coverage ratio R:

[0092] A i =||y i,max -y i,min ||*fov x

[0093]

[0094] Among them, A i Let A be the area of ​​the i-th covering path, A be the area of ​​the polygon, and R be a dimensionless evaluation index. cover This indicates the total coverage area.

[0095] For example, S102-4: calculating the ground resolution, normalizing it using the lowest and highest ground resolutions, and using it as the fourth dimensionless evaluation index, specifically includes:

[0096] Calculate the ground sampling distance (GSD) using the lowest and highest ground sampling distances. min GSD max Standardization:

[0097]

[0098] This can be transformed into a dimensionless evaluation index.

[0099] For example, step S102-5: A weighted summation of the first, second, third, and fourth dimensionless evaluation indices is performed to obtain the overall evaluation index function, specifically including:

[0100] Overall evaluation index function:

[0101] J = w1 * L norm +w2*N norm +w3*R norm +w4*GSD norm ;

[0102] Among them, w1 to w4 are the weights of the four evaluation indicators, which are adjusted by the user according to their decision-making preferences.

[0103] The evaluation index J is a function of the coverage direction α and the UAV flight altitude h, and exhibits strong nonlinearity. Therefore, a bio-inspired approach is used for optimization.

[0104] Further, S103: Based on a bio-inspired approach, the overall evaluation index function is optimized to obtain the coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs a coverage task, specifically including:

[0105] S103-1: The overall evaluation index function is used as the fitness function of the beetle whisker search optimizer. The independent variables of the fitness function are the coverage direction and the flight altitude.

[0106] Set the parameters of the longhorn beetle whisker search optimizer, including: left and right whisker spacing, search step size, decay coefficient, and convergence criterion;

[0107] S103-2: Randomize the tentacles of the agent in the decision space and calculate the positions of the left and right tentacles in the decision space;

[0108] S103-3: Calculate the fitness of the left and right wings, and determine the direction of the agent's movement in the decision space by comparison;

[0109] S103-4: The agent moves to a new position according to the direction and step size, and updates the agent's whisker spacing and search step size according to the decay coefficient;

[0110] S103-5: Determine whether the convergence criterion has been met. If yes, output the optimized coverage direction and flight altitude. If no, return to S103-2.

[0111] It should be understood that the convergence criterion means that a predetermined number of iterations is reached, or the optimal solution output no longer changes. It should be understood that, as... Figures 4(a)-4(d) As shown, the longhorn beetle whisker search optimizer requires only one search agent to search the decision space; its search agent mimics the foraging method of longhorn beetles in the biological world, iteratively searching for the optimal solution in the decision space; the search agent contains a pair of whiskers, and in one iteration, the direction of the whiskers is first randomized. Then calculate the positions of the left and right whiskers:

[0112]

[0113]

[0114] Where, x t d is the position of the agent's centroid in the decision space at the t-th iteration. t x is the spacing at the t-th iteration. l It is the position of the left limb in the decision space, x r It is the position of the right side in the decision-making space.

[0115] By comparing the fitness f(x) of the left and right whiskers l ) and f(x rThe direction of movement of the agent in this iteration is determined by the [factor], and the new position of the agent after this iteration is:

[0116]

[0117] Where, δ t It is the step size at the t-th iteration, and sign(·) is the sign function, which takes the value 1 or -1.

[0118] The agent's tentacle length d t and search step size δ t In the next iteration, the decay will be synchronized according to the decay coefficient:

[0119] d t+1 =η*d t ,

[0120] δ t+1 =η*δ t

[0121] Where η is the attenuation coefficient, which is generally taken as 0.95 to 0.995.

[0122] See appendix Figure 5 As shown, under the set convergence criteria, the algorithm will converge to the global optimal solution.

[0123] Appendix Figure 6 This is a complete flowchart of the multi-rotor UAV coverage path planning method.

[0124] The coverage path planning method for multi-rotor UAVs proposed in this invention can perform full coverage path planning for a specified arbitrary polygonal region, including concave polygonal regions. This allows for optimization of the total length of the coverage path, the number of turns, the area coverage rate, and the ground resolution to varying degrees, thereby shortening the task execution time, reducing the energy consumption of the task execution, and improving the quality of mapping.

[0125] Example 2

[0126] This embodiment provides a multi-rotor unmanned aerial vehicle (UAV) coverage path planning system;

[0127] A multi-rotor UAV coverage path planning system, including:

[0128] The path building module is configured to: build a coverage path for a multi-rotor UAV over a polygonal region;

[0129] The indicator construction module is configured to: construct several evaluation indicators for the coverage path, and perform a weighted summation of all evaluation indicators to obtain the total evaluation indicator function;

[0130] The optimization solution module is configured to: optimize the overall evaluation index function based on a bio-inspired approach to obtain the optimal coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs the coverage task; based on the optimization solution results, the multi-rotor UAV begins to perform the coverage task.

[0131] It should be noted that the path construction module, index construction module, and optimization solution module mentioned above correspond to steps S101 to S103 in Embodiment 1. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should be noted that these modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0132] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-rotor UAV coverage path planning method, characterized by: include: Construct the coverage path of a multi-rotor UAV over a polygonal region; For the coverage path, several evaluation indicators are constructed, and all evaluation indicators are weighted and summed to obtain the total evaluation indicator function; Based on a bio-inspired approach, the overall evaluation index function is optimized to obtain the optimal coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs a coverage task. Based on the optimization results, the multi-rotor UAV begins to perform the coverage task. Specifically, for the coverage path, several evaluation indicators are constructed, and all evaluation indicators are weighted and summed to obtain the overall evaluation indicator function, which includes: Calculate the total length of the complete coverage path, and calculate the quotient of the total length of the complete coverage path to the perimeter of the polygonal region. Use the quotient as the first dimensionless evaluation index. The number of bends in the complete coverage path is calculated. Principal component analysis algorithm is used to obtain the maximum and minimum number of bends. The number of bends in the complete coverage path is normalized using the maximum and minimum number of bends and used as the second dimensionless evaluation index. Calculate the area coverage rate of the complete coverage path and use the area coverage rate as the third dimensionless evaluation index; Calculate the ground resolution and normalize it using the minimum and maximum ground resolutions as the fourth dimensionless evaluation index. The first, second, third, and fourth dimensionless evaluation indicators are weighted and summed to obtain the overall evaluation indicator function; Specifically, the optimization of the overall evaluation index function based on a bio-inspired approach yields the coverage direction, flight altitude, and path points corresponding to each flight state of the multi-rotor UAV performing a coverage mission. (1) The overall evaluation index function is used as the fitness function of the beetle whisker search optimizer. The independent variables of the fitness function are the coverage direction and the flight altitude. The parameters of the beetle whisker search optimizer are set, including: the left and right whisker spacing, the search step size, the decay coefficient and the convergence criterion. (2) Randomize the tentacles of the agent in the decision space and calculate the positions of the left and right tentacles in the decision space; (3) Calculate the fitness of the left and right sides, and determine the direction of the agent's movement in the decision space by comparison; (4) The agent moves to a new position according to the direction and step size, and updates the agent's whisker spacing and search step size according to the decay coefficient; (5) Determine whether the convergence criterion has been met. If yes, output the optimized coverage direction and flight altitude. If no, return to (2).

2. The multi-rotor UAV coverage path planning method as described in claim 1, characterized in that, Constructing the coverage path of a multi-rotor UAV over a polygonal region specifically includes: The flight altitude of the multi-rotor UAV is determined based on the camera parameters and the set ground resolution. Based on camera parameters and the flight altitude of the multi-rotor UAV, the width of the coverage path during the flight of the multi-rotor UAV is determined; the coverage path is a rectangular strip area. Assuming a multi-rotor UAV scans a polygonal region in a serpentine manner, the spacing between the centerlines of adjacent coverage paths is determined based on the width of the coverage path and the overlap between adjacent coverage paths. The distribution of the coverage path in the scanning direction is determined based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths. Based on the coverage margin set in the coverage direction in the mapping, determine the two endpoints in the coverage direction of each coverage path; Connect the endpoints of all coverage paths sequentially in a serpentine manner, then connect the starting point of the coverage path to the take-off point of the multi-rotor UAV, and connect the ending point of the coverage path to the landing point of the multi-rotor UAV to obtain the complete coverage path.

3. The multi-rotor UAV coverage path planning method as described in claim 2, characterized in that, Based on camera parameters and the set ground resolution, the flight altitude of the multi-rotor UAV is determined, specifically including: flight altitude. Determined by the camera's pixel size ,focal length and the set ground resolution range And it is determined that: 。 4. The multi-rotor UAV coverage path planning method as described in claim 2, characterized in that, The width of the coverage path during the flight of the multi-rotor UAV is determined based on camera parameters and the flight altitude of the multi-rotor UAV; the coverage path is a rectangular strip area, specifically including: Width of the coverage path Due to the size of the camera's image sensor ,focal length and flight altitude And it is determined that: 。 5. The multi-rotor UAV coverage path planning method as described in claim 2, characterized in that, Assuming a multi-rotor UAV scans a polygonal region using a serpentine weaving motion, the spacing between the centerlines of adjacent coverage paths is determined based on the width of the coverage path and the overlap between adjacent coverage paths. Specifically, this includes: Spacing between adjacent straight paths The coverage width of the path Overlap required for mapping And it is determined that: 。 6. The multi-rotor UAV coverage path planning method as described in claim 2, characterized in that, The determination of the distribution of the coverage path in the scanning direction based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths specifically includes: Construct a Cartesian coordinate system, where the horizontal axis of the Cartesian coordinate system represents the scanning direction of the multi-rotor UAV, and the vertical axis of the Cartesian coordinate system represents the coverage direction of the multi-rotor UAV. Project all vertices of the polygonal region onto the scanning direction to obtain the coordinates of the left and right boundary points of the polygonal region. The distribution of the coverage path in the scanning direction is determined based on the coordinates of the left and right boundary points of the polygonal region, the coverage margin set in the scanning direction during mapping, the width of the coverage path, and the spacing between the center lines of adjacent coverage paths.

7. The multi-rotor UAV coverage path planning method as described in claim 2, characterized in that, Determine the two endpoints of each coverage path in the coverage direction, specifically including: First, find the intersection point of the coverage width edge of the coverage path with the edge of the polygonal region. Then, find the vertices of the polygonal region within the coverage range of the coverage path. Next, project the intersection point and the vertices in the coverage direction to obtain the key point set. Finally, based on the key point set and the set coverage margin in the coverage direction, determine the two endpoints of each coverage path.

8. A multi-rotor unmanned aerial vehicle (UAV) coverage path planning system, characterized in that: The multi-rotor UAV coverage path planning method as described in claims 1-7 includes: The path building module is configured to: build a coverage path for a multi-rotor UAV over a polygonal region; The indicator construction module is configured to: construct several evaluation indicators for the coverage path, and perform a weighted summation of all evaluation indicators to obtain a total evaluation indicator function; specifically, the indicator construction module is configured to: calculate the total length of the complete coverage path, calculate the quotient of the total length of the complete coverage path and the perimeter of the polygonal region, and use the quotient as the first dimensionless evaluation indicator; calculate the number of bends of the complete coverage path, use principal component analysis to obtain the maximum and minimum number of bends, and normalize the number of bends of the complete coverage path using the maximum and minimum number of bends as the second dimensionless evaluation indicator; calculate the area coverage rate of the complete coverage path, and use the area coverage rate as the third dimensionless evaluation indicator; calculate the ground resolution, and normalize the ground resolution using the minimum and maximum ground resolution as the fourth dimensionless evaluation indicator; and perform a weighted summation of the first, second, third, and fourth dimensionless evaluation indicators to obtain the total evaluation indicator function; The optimization solution module is configured to: optimize the overall evaluation index function based on a bio-inspired approach to obtain the optimal coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs the coverage task; based on the optimization solution results, the multi-rotor UAV begins to perform the coverage task; wherein, the optimization solution of the overall evaluation index function based on a bio-inspired approach to obtain the coverage direction, flight altitude, and path points corresponding to each flight state when the multi-rotor UAV performs the coverage task specifically includes: (1) using the overall evaluation index function as the fitness function of the beetle whisker search optimizer, the fitness function's self The variables are coverage direction and flight altitude; set the parameters of the beetle whisker search optimizer, including: left and right whisker spacing, search step size, decay coefficient and convergence criterion; (2) randomize the whisker direction of the agent in the decision space and calculate the position of the left and right whiskers in the decision space; (3) calculate the fitness of the left and right whiskers and obtain the movement direction of the agent in the decision space by comparison; (4) the agent moves to the new position according to the movement direction and step size, and updates the whisker spacing and search step size of the agent according to the decay coefficient; (5) determine whether the convergence criterion is met. If yes, output the optimized coverage direction and flight altitude. If no, return (2).

Images