UAV path planning method based on Hippocampus optimization algorithm

Abstract

The present invention relates to the technical field of drone path planning, and in particular to a drone path planning method based on a hippocampus optimization algorithm, which constructs a three-dimensional space grid of the drone flight area, maps the geographic information of the drone flight area to the three-dimensional space grid, and constructs a flight cost function constrained by different factors of the flight path, wherein each track point in the flight path corresponds to a three-dimensional space grid; sets corresponding cost weights for the flight costs constrained by different factors, performs weighted fusion of the flight costs constrained by different factors according to the cost weights, and constructs an objective function of the drone flight path; and optimizes and solves the objective function of the drone flight path to determine the left and right flight paths of the drone. The present invention effectively improves the accuracy of path planning, and the planned path is more in line with the real scene.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and in particular to a UAV path planning method based on a Hippocampus optimization algorithm. Background Art

[0002] In recent years, drone technology has developed rapidly and has been widely used in civil and commercial fields. With the increasing application of drones in various tasks, how to achieve efficient and safe path planning and flight control under automatic unmanned operation has become the focus of drone technology research. During the flight, drones need to avoid various obstacles, no-fly zones and signal interference areas, while ensuring flight stability and energy efficiency. Therefore, multi-constraint path planning technology for complex environments is particularly important.

[0003] At present, the technologies for UAV path planning mainly include: path planning based on graph search algorithms such as A* algorithm and Dijkstra algorithm, and algorithms based on genetic algorithm and particle swarm optimization. However, such technologies usually have the following defects:

[0004] 1. It only focuses on exploring the shortest path while avoiding obstacles, without providing a detailed description of the flight scene. The description of the drone's flight area mostly stays on a simple three-dimensional model. This simplification leads to a lack of accuracy in the planned path, which in turn leads to poor usability in real situations.

[0005] 2. Although it involves obstacle avoidance, it ignores the importance of safety redundancy, which may increase the flight risk of the drone when it approaches the ground or obstacles;

[0006] 3. Most of them simply rely on the shortest path or obstacle avoidance principle, while ignoring the comprehensive balance of multiple factors. This will lead to the inability to achieve global optimization between safety, energy consumption and flight stability in path planning, and the lack of universality for different tasks and aircraft models;

[0007] 4. Failure to carry out detailed optimization for drones of different models and mission requirements results in poor adaptability and robustness of the planning algorithm, especially when it comes to different types of operations or complex flight missions, which can easily lead to performance bottlenecks;

[0008] 5. Using traditional optimization algorithms for path planning can easily lead to local optimal solutions, making it difficult to effectively explore the global optimal path. When dealing with multiple constraints, complex parameter adjustments are often required, resulting in reduced computational efficiency and performance bottlenecks when facing large-scale search spaces. Summary of the invention

[0009] In order to solve the above technical problems, the purpose of this application is to provide a UAV path planning method based on the Hippocampus optimization algorithm. The technical solutions adopted are as follows:

[0010] In a first aspect, the present invention provides a method for UAV path planning based on a Hippocampus optimization algorithm, comprising the following steps:

[0011] Construct a three-dimensional spatial grid of the UAV flight area, map the geographic information of the UAV flight area to the three-dimensional spatial grid, and construct a flight cost function constrained by different factors of the flight path, where each track point in the flight path corresponds to a three-dimensional spatial grid; the flight costs constrained by different factors at least include: the flight cost constrained by basic factors, the flight cost constrained by the NIMBY effect, and the flight cost constrained by the land use of the flight path;

[0012] Setting corresponding cost weights for flight costs constrained by different factors, weightedly integrating the flight costs constrained by different factors according to the cost weights, and constructing an objective function of the UAV flight path;

[0013] The objective function of the UAV flight path is optimized and solved to determine the optimal flight path of the UAV.

[0014] In combination with the first aspect above, in some possible implementations, the calculation formula of the flight cost of the NIMBY effect constraint is:

[0015] ;

[0016] In the formula, Indicates Flight paths The flight cost constrained by the NIMBY effect; Indicates Flight paths The number of demographic rasters within the buffer of the current track point; Indicates Flight paths The current track point in the buffer population count for each demographic raster; Indicates the flight path The current track point in the buffer The demographic grid and Flight paths The area of ​​the intersection of the current track point buffer.

[0017] In combination with the first aspect above, in some possible implementations, the process of determining the flight cost of the flight path land constraint includes:

[0018] Set corresponding land use index weights for different types of land;

[0019] Determine the distance between each trackpoint in the flight path and the next trackpoint after it;

[0020] According to the land use index weight corresponding to the land type of each track point in the flight path, the distances corresponding to each track point in the flight path are weighted and fused to obtain the flight cost constrained by the land use of the flight path.

[0021] In combination with the first aspect above, in some possible implementations, the calculation formula of the flight cost of the flight path land constraint is:

[0022] ;

[0023] In the formula, Indicates Flight paths Flight costs due to flight path land constraints; Indicates Flight paths The Track points The land use index weight corresponding to the land type; Indicates Flight paths The Track points With indivual The Euclidean distance between Indicates Flight paths The total number of track points in .

[0024] In combination with the above-mentioned first aspect, in some possible implementations, the flight cost constrained by the basic factors includes at least: the flight cost constrained by the flight distance, the flight cost constrained by the safety threat area, and the flight cost constrained by the flight angle.

[0025] In combination with the first aspect above, in some possible implementations, the calculation formula of the flight cost of the flight distance constraint is:

[0026] ;

[0027] In the formula, Indicates Flight paths The flight cost constrained by the flight distance; Indicates Flight paths The Track points and Track points The Euclidean distance between Indicates Flight paths The total number of track points in .

[0028] In combination with the first aspect above, in some possible implementations, the calculation formula of the flight cost constrained by the security threat zone is:

[0029] ;

[0030] In the formula, Indicates Flight paths The flight cost constrained by the security threat zone; Indicates Flight paths The total number of track points in ; Indicates the total number of obstacles in the flight environment; Indicates Flight paths Segments in Relative The threat cost of an obstacle; Defined as:

[0031] ;

[0032] In the formula, represents the obstacle avoidance penalty coefficient; Indicates the flight environment The distance between the center of mass of an obstacle and its surface; Indicates the set security threat distance; Indicates the first obstacle centroid to flight segment The shortest distance; and Respectively represent Flight paths The Track points and Track points.

[0033] In combination with the first aspect above, in some possible implementations, the calculation formula of the flight cost of the flight angle constraint is:

[0034] ;

[0035] ;

[0036] ;

[0037] In the formula, Indicates Flight paths The flight cost constrained by the flight angle; and represents the penalty term; Indicates Flight paths The The horizontal turning angle corresponding to each track point; and Respectively represent Flight paths The Track points and The vertical pitch angle corresponding to each track point; Indicates Flight paths The total number of track points in ; and Respectively represent The flight path Track points , No. Track points and Track points The two consecutive flight segments formed and exist Projection on a plane; and Respectively expressed in Flight paths The Track points and Track points The vertical coordinate in the three-dimensional coordinates of ; Indicates the vector modulus length symbol.

[0038] In combination with the first aspect above, in some possible implementations, optimizing and solving the objective function of the flight path of the drone to determine the left and right flight paths of the drone includes:

[0039] An artificial potential field method is used to generate a set of initial path solutions;

[0040] Based on the initial path solution, the Hippocampus optimization algorithm is used to optimize and solve the objective function of the UAV flight path to determine the left and right flight paths of the UAV.

[0041] In combination with the first aspect above, in some possible implementations, the calculation formulas for the repulsive potential field and the gradient of the repulsive force in the repulsive potential field in the artificial potential field method are:

[0042] ;

[0043] In the formula, Indicates Flight paths The current Track points The corresponding repulsive potential field; represents the repulsion coefficient; Indicates Flight paths The current Track points The first Obstacles distance; Indicates the flight environment Obstacles The distance of the center of mass from its surface; Indicates the set security threat distance;

[0044] ;

[0045] In the formula, Indicates Flight paths The current Track points The corresponding gradient of the repulsive force; Indicates the total number of obstacles in the flight environment.

[0046] In a second aspect, the present invention further provides a drone path planning device based on a hippocampus optimization algorithm, the device comprising:

[0047] A flight cost function acquisition module is used to construct a three-dimensional space grid of the UAV flight area, map the geographic information of the UAV flight area to the three-dimensional space grid, and construct a flight cost function constrained by different factors of the flight path. Each track point in the flight path corresponds to a three-dimensional space grid; the flight costs constrained by different factors include at least: the flight cost constrained by basic factors, the flight cost constrained by the NIMBY effect, and the flight cost constrained by the land use of the flight path;

[0048] An objective function acquisition module is used to set corresponding cost weights for flight costs constrained by different factors, perform weighted fusion on the flight costs constrained by different factors according to the cost weights, and construct an objective function for the flight path of the UAV;

[0049] The flight path acquisition module is used to optimize and solve the objective function of the UAV flight path to determine the optimal flight path of the UAV.

[0050] In a third aspect, the present invention further provides a drone path planning system based on the Hippocampus optimization algorithm, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method in the first aspect or any possible implementation of the first aspect.

[0051] In a fourth aspect, the present invention further provides a computer program product, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0052] In a fifth aspect, the present invention further provides a computer-readable storage medium storing a computer program code. When the computer program code runs on a computer, the computer executes the method in the first aspect or any possible implementation of the first aspect.

[0053] The present invention has the following beneficial effects: the present invention constructs a three-dimensional space grid, maps the geographic information of the UAV flight area to the three-dimensional space grid, constructs a flight cost function constrained by different factors of the flight path, thereby constructing an objective function of the UAV flight path, optimizing and solving the objective function of the UAV flight path, and finally determining the optimal flight path of the UAV. The present invention realizes the detailed modeling of the complex flight environment and the integration of multiple constraints by constructing a flight cost function constrained by different factors of the flight path. Compared with the rough three-dimensional description of the flight area in the prior art, the accuracy of path planning is effectively improved, so that the planned flight path is more in line with the real scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 A flowchart of a method for UAV path planning based on the Hippocampus optimization algorithm according to an embodiment of the present invention;

[0056] Figure 2Schematic diagram of the structure of a UAV path planning device based on the Hippocampus optimization algorithm according to an embodiment of the present invention;

[0057] Figure 3 Schematic diagram of the structure of a UAV path planning system based on the Hippocampus optimization algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to clearly illustrate the technical features of the present invention, the present invention is described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0059] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0060] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0061] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0062] It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0063] Although operations or steps are described in a specific order in the drawings in the embodiments of the present invention, it should not be understood that it is required to perform these operations or steps in the specific order shown or in a serial order, or that all the operations or steps shown must be performed to obtain the desired results. In the embodiments of the present invention, these operations or steps may be performed in series; these operations or steps may also be performed in parallel; or some of these operations or steps may be performed.

[0064] At the same time, it is understood that the data involved in the technical solution of the present invention (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of the relevant laws, regulations and relevant provisions. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meanings as those generally understood by technicians in the technical field of the present invention, and all parameters or indicators in the formulas involved in the present invention are normalized values ​​that eliminate the influence of dimensions.

[0065] In order to solve the problem that the existing planned UAV flight paths cannot achieve global optimization, which leads to poor availability in actual situations, an embodiment of the present invention provides a UAV path planning method based on the Hippocampus optimization algorithm. The method constructs a three-dimensional space grid, maps the geographic information of the UAV flight area to the three-dimensional space grid, constructs a flight cost function constrained by different factors of the flight path, and thus constructs an objective function of the UAV flight path. The objective function of the UAV flight path is optimized and solved, and finally the optimal flight path of the UAV is determined.

[0066] A UAV path planning method based on the Hippocampus optimization algorithm provided by an embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0067] Figure 1 A basic flow chart of a UAV path planning method based on the Hippocampus optimization algorithm provided by an embodiment of the present invention is shown. Figure 1 As shown, the method specifically comprises the following steps:

[0068] Step S100: construct a three-dimensional spatial grid of the UAV flight area, map the geographic information of the UAV flight area to the three-dimensional spatial grid, and construct a flight cost function constrained by different factors of the flight path.

[0069] Construct a three-dimensional coordinate system in the flight area of ​​the drone. The single side length of the flight area of ​​the drone can be set to 3km-7km according to the mission type, and divide the flight area of ​​the drone into three-dimensional space grids with a resolution of 1m. Each three-dimensional space grid is represented by a unique three-dimensional coordinate, and each grid represents a 1m × 1m × 1m three-dimensional space unit. All three-dimensional space grids in the flight area of ​​the drone are used as candidate flight path points of the drone. Definition The collection of all grids in the flight area of ​​the drone: ,in, The kth three-dimensional space grid representing the flight area of ​​the drone, The three-dimensional coordinates of the kth three-dimensional space grid representing the flight area of ​​the drone, The total number of three-dimensional spatial grids representing the drone's flight area.

[0070] Import the geographic information of the flight area. Import the indicator data into the geographic information system (GIS). If it is raster data, it needs to be converted into vector data first to improve the accuracy of calculation and the expression ability of spatial topological relationship. When projecting the data, ensure that it is consistent with the defined three-dimensional spatial grid and use the same geographic coordinate system. Map the regional information including terrain, land use type, no-fly zone, building information, and signal interference zone to the grid structure, and perform modeling processing to achieve visualization. The signal interference zone refers to the area in a specific area where the communication, navigation or control signal of the drone is affected due to interference from electromagnetic waves or other signal sources (such as communication towers, high-voltage wires, radar equipment, etc.), which interferes with the normal operation of the drone. Furthermore, the spatial unit occupied by the mandatory constraint area is marked as occupied, and it is regarded as an inaccessible area. The drone path point is prohibited from overlapping with it. Through the above process, the integration of accurate description of the three-dimensional space and constraint conditions is realized, so that the path planning algorithm can generate a feasible and safe flight path based on this structure.

[0071] Constrained flight altitude. Introduce constraints on the flight altitude of drones to ensure the flight safety of drones and compliance with airworthiness regulations. Specifically, the minimum flight altitude is closely related to the safety threat area of ​​the terrain, that is, the flight altitude of the drone must not be lower than the safety redundancy distance of the terrain. The maximum flight altitude is differentiated according to the type of drone and mission requirements, ranging from 120 meters for small drones to 300 meters for large drones. The altitude setting must comply with the airworthiness regulations of various types of drones, while taking into account the actual needs of the flight mission. The flight altitude of the drone can be freely adjusted between the minimum altitude and the maximum altitude to achieve path optimization.

[0072] Construct the path length index, i.e. the flight cost constrained by the flight distance. In the UAV path planning problem, the core task is to determine the optimal path from the starting point to the target point, i.e. to find the shortest flight path solution.

[0073] Usually, each track point in the flight path can be represented by its three-dimensional coordinates. Flight paths The The three-dimensional coordinates of the track points are: , then Flight paths The track point sequence is represented as , where the track point It is a specific stereo grid, that is, a three-dimensional space grid, whose three-dimensional coordinates satisfy , Respectively represent the track point in three-dimensional space Coordinates. Entire flight path It can be expressed as A sequence of three-dimensional coordinates of track points.

[0074] To measure the distance between adjacent track points in the path, the Euclidean distance formula is used to calculate the distance between any two adjacent track points. Flight paths The Track points and +1 track point As an example, the Euclidean distance can be defined as:

[0075] ;

[0076] In the formula, Indicates Flight paths The Track points and +1 track point The Euclidean distance between and Respectively represent Flight paths The Track points and +1 track point The three-dimensional coordinates of .

[0077] The distance between any two adjacent track points reflects the actual flight distance of the UAV from one track point to the next track point in three-dimensional space. Flight paths The total path length can be expressed by summing the Euclidean distances of all adjacent track points. The path length index, i.e., the flight cost of the flight distance constraint, can be defined as the sum of the Euclidean distances between all adjacent points:

[0078] ;

[0079] In the formula, Indicates Flight paths The flight cost constrained by the flight distance; Indicates Flight paths The Track points and Track points The Euclidean distance between Indicates Flight paths The total number of track points in .

[0080] Flight cost with flight distance constraint Used to quantify the entire flight path The total distance is one of the goals that need to be minimized in optimizing UAV trajectory planning.

[0081] Construct the flight cost of the security threat index, that is, the security threat zone constraint. When constructing the constraint condition for the minimum security threat, it is defined that there is a security threat zone around the area that the drone cannot enter. The security threat zone refers to a buffer area surrounding inaccessible areas such as obstacles and no-fly zones. It is equivalent to a kind of safety redundancy, which is used to provide additional protection when the drone is flying. In the security threat area, the risk will increase when it is too close to inaccessible areas such as obstacles or no-fly zones. When the flight path of the drone approaches this area, a certain security threat cost will be introduced. Once the flight path enters this threat zone, the threat cost will begin to increase. The closer the path is to the inaccessible area, the higher the threat cost.

[0082] Specifically, the safety threat index, i.e., the flight cost constrained by the safety threat zone, can be expressed as:

[0083] ;

[0084] In the formula, Indicates Flight paths The flight cost constrained by the flight distance; Indicates Flight paths The total number of track points in ; Indicates the total number of obstacles in the flight environment; Indicates Flight paths Segments in Relative The threat cost of an obstacle; Defined as:

[0085] ;

[0086] In the formula, It represents the obstacle avoidance penalty coefficient. The recommended value range is 1-10, and 5±2 is usually used as the starting point for testing. Indicates The distance between the center of mass of an obstacle and its surface; Indicates the set security threat distance; Indicates that from obstacle centroid to flight segment The shortest distance; and Respectively represent Flight paths The Track points and Track points.

[0087] The flight cost according to the above flight distance constraint From the calculation formula, we can see that when the UAV path is a certain distance away from the threat area, the threat cost is zero, indicating that the path is in safe flight; but when the path approaches or enters the threat area, the threat cost rises rapidly as the distance decreases. Reducing the security threat index is used to ensure that the UAV avoids these high-risk areas and ensure flight safety.

[0088] The NIMBY index, i.e., the flight cost constrained by the NIMBY effect, is constructed. The NIMBY index indicates the degree of avoidance of densely populated areas of human activity near the track point to reduce the interference of drone flights on residents.

[0089] Specifically, the calculation of the NIMBY index, i.e., the flight cost constrained by the NIMBY effect, is as follows:

[0090] ;

[0091] In the formula, Indicates Flight paths The flight cost constrained by the NIMBY effect; Indicates Flight paths The number of demographic grids within the current track point buffer. The current track point buffer is the number of demographic grids within the current track point buffer. Flight paths The area within the circle with each track point as the center and r as the radius, where the radius r is an adjustable parameter. The demographic grid refers to the microscopic population data obtained based on mobile phone signaling data, presented in the form of a grid, usually with a fixed spatial resolution; Indicates Flight paths The current track point in the buffer population count for each demographic raster; Indicates Flight paths The current track point in the buffer Demographic rasters and flight paths The area of ​​the intersection of the current track point buffer.

[0092] The flight cost according to the above NIMBY effect constraint The calculation formula shows that when the drone path is close to the dense human activity area, people's aversion to drones increases. Reducing the NIMBY index is used to ensure that fewer residents are disturbed by drones.

[0093] Construct the flight angle index, that is, the flight cost of the flight angle constraint. In the UAV path planning, the flight angle control parameters mainly involve the horizontal turning angle and the vertical pitch angle, which are crucial to maintaining the smoothness and physical feasibility of the flight trajectory. If the actual flight path turning angle of the UAV exceeds its maneuverability range (different models of UAVs have different functional parameters), the aircraft will not be able to follow the path planning, resulting in an unfeasible trajectory design. Therefore, it is necessary to ensure that these two turning angles meet the physical constraints of the UAV. In addition to the physical constraints of the turning angle, too frequent turning angles and too large turning angles will also bring additional energy consumption. When the UAV makes a sharp turn, it needs to consume more energy to change the flight direction, which further affects the energy consumption performance of the UAV.

[0094] set up Respectively represent the drone in the Flight paths Three consecutive track points in and Indicates Two consecutive segments of a flight path, where The projections on the plane are and . No. Flight paths The Track points The corresponding horizontal turning angle and vertical pitch angle As a control parameter, it is defined as the turning angle between flight segments. According to the vector calculation, the horizontal turning angle and vertical pitch angle It can be calculated by the following formula:

[0095] ;

[0096] ;

[0097] In the formula, and Respectively expressed in Flight paths The Track points and Track points The vertical coordinate in the three-dimensional coordinates of ; Indicates the vector modulus length symbol.

[0098] In order to prevent the horizontal turning angle and vertical pitch angle of the drone from exceeding their physical limits, a penalty term is designed and To constrain these corners. Finally, the flight angle index, i.e., the flight cost of the flight angle constraint, can be expressed as:

[0099] ;

[0100] In the formula, Indicates Flight paths The flight cost constrained by the flight angle; Indicates Flight paths The The vertical pitch angle corresponding to each track point; Indicates Flight paths The total number of track points in .

[0101] Construct a flight path land use index, which is the flight cost constrained by the flight path land use. In drone path planning, the flight path land use index is introduced to quantify the impact of drones flying over different land use types. This index aims to minimize interference with densely populated areas and sensitive land, minimize the impact of drones on the ground when they need to make an emergency landing, and ensure that the path passes through appropriate land types. Based on their functions and population density, they are divided into the following categories, and each type of land is assigned a corresponding land use index weight. , the weights are set according to their suitability in academic and practical applications, and the reference values ​​are given here:

[0102] ;

[0103] In the formula, and Respectively represent the first land type and the second land type. The first land type refers to farmland, and the second land type refers to green land or park. The first land type and the second land type are suitable for flying, and the index is low. Set the reference value of the land use index weight corresponding to the first land type and the second land type ; and Indicates the third and fourth land types. The third land type refers to industrial areas, and the fourth land type refers to office areas. Although the third and fourth land types are acceptable for flight, safety and noise must be considered. The index is medium. Set the reference value of the land use index weight corresponding to the third and fourth land types. ; and Respectively represent the fifth and sixth land types. The fifth land type refers to residential areas, and the sixth land type is a sensitive area. Flying over the fifth and sixth land types should be avoided as much as possible to reduce interference and privacy issues. The index is higher. The reference value of the land use index weight corresponding to the fifth and sixth land types .

[0104] Assume that the drone Flight paths The Track points The location can be mapped to a specific land type , define each flight segment Land weight The total impact on the path is:

[0105] ;

[0106] In the formula, Indicates Flight paths Flight costs due to flight path land constraints; Indicates Flight paths The Track points The land use index weight corresponding to the land type; Indicates Flight paths Track points in and The Euclidean distance between Indicates Flight paths The total number of track points.

[0107] Step S200: setting corresponding cost weights for flight costs constrained by different factors, performing weighted fusion on the flight costs constrained by different factors according to the cost weights, and constructing an objective function of the UAV flight path.

[0108] Construct the objective function of the UAV flight path. Combining the flight cost constrained by the above different factors, the final objective function of the UAV flight path is It can be expressed as:

[0109] + ;

[0110] In the formula, They are the cost weights of path length index, security threat index, NIMBY index, flight turning angle index, and flight path land index.

[0111] Step S300: optimizing and solving the objective function of the UAV flight path to determine the optimal flight path of the UAV.

[0112] In the above objective function of the UAV flight path In the calculation formula, by adjusting the cost weight, the multiple objectives of drone path planning can be balanced in different mission scenarios to achieve more optimized flight performance. For example, for small drones, whose flight performance is more flexible, the cost weights of the flight angle index and the flight path land index can be appropriately reduced; while for large drones with longer endurance, the cost weights of the security threat index and the flight angle index need to be increased. The specific value of the weight needs to be determined based on the mission model, expert evaluation and flight data analysis.

[0113] The goal of UAV flight path planning is to optimize the objective function , making it as small as possible to meet the scene requirements and achieve reasonable planning of the UAV flight path. According to the objective function The definition of objective function There is no linear relationship between the terms in , which is a nonlinear and non-convex combinatorial optimization problem. Involving multiple discrete decision variables means that multiple local optimal solutions may be encountered during the optimization process. The above characteristics of the traditional optimization method usually cannot effectively solve this problem for the following reasons:

[0114] In non-convex optimization problems, traditional optimization methods can easily fall into local optimal solutions because these methods rely on gradients or other local information and cannot effectively escape from local optimal solutions.

[0115] The search space for combinatorial optimization problems is very large, especially when the number of potential alternative flight trajectories is large, and the number of possible solutions grows exponentially with the increase in positions and constraints.

[0116] Based on the above problems, we choose to use the Sea-horse optimizer (SHO) in the heuristic algorithm to optimize the objective function. The Sea-horse optimizer is a heuristic optimization algorithm based on swarm intelligence, inspired by the movement, predation and reproduction behavior of seahorses in nature. The specific implementation process is as follows:

[0117] Initial path generation. Artificial Potential Field: The movement of a mobile object in an environment is designed as a movement in an abstract artificial virtual force field. Use the artificial potential field method to generate a set of initial path solutions for UAV planning. In the virtual force field, the target point exerts gravity on the drone, while the obstacle exerts repulsion on the drone. Under the combined effect of gravity and repulsion, the drone gradually moves to generate a more reasonable initial path. The basic settings for path initialization are as follows:

[0118] The gravitational potential field consists of the gravitational force and its gradient. Generates gravitational force on the drone, attracting it to approach the end point. The gravitational potential field is defined as:

[0119] ;

[0120] In the formula, Indicates Flight paths The current Track points The corresponding gravitational potential field; Indicates the end point of the path; is the gravitational coefficient; Indicates Flight paths The current Track points With path end The Euclidean distance of .

[0121] The gradient of gravity (i.e. the direction of gravity) is defined as:

[0122] ;

[0123] In the formula, Represents the gradient of gravity (i.e. the direction of gravity).

[0124] The repulsive potential field consists of the repulsive force and the gradient of the repulsive force. Apply repulsive force to the drone to avoid collision. The repulsive potential field is defined as:

[0125]

[0126] In the formula, Indicates Flight paths The current Track points The corresponding repulsive potential field; represents the repulsion coefficient; For the Flight paths The current Track points With Obstacles distance; Indicates Obstacles The distance of the center of mass from its surface; Indicates the set security threat distance.

[0127] The gradient of the repulsive force (i.e. the direction of the repulsive force) is defined as:

[0128] ;

[0129] In the formula, Indicates Flight paths The current Track points The corresponding repulsive force gradient (i.e., repulsive force direction); Indicates the total number of obstacles in the flight environment.

[0130] The initial path is generated by the artificial potential field method. Each time the path is updated, the path points are adjusted according to the resultant force, and finally an initial solution path from the starting point to the end point is obtained.

[0131] Sea-horse Optimizer (SHO): is a meta-heuristic optimization algorithm inspired by the natural social behavior of seahorses. It simulates the movement, predation and reproduction behaviors of seahorses and is used to solve complex optimization problems. Based on this initial solution path, the Sea-horse Optimizer is implemented to achieve path optimization.

[0132] The three key components of the seahorse optimization algorithm include the spiral motion, predation behavior and reproduction process of the seahorse. These three behaviors work together to enable the seahorse optimization algorithm to effectively solve the complex multi-objective optimization problem in UAV path planning and achieve a good balance between local exploitation and global exploration.

[0133] The spiral motion is mainly used for local development, that is, to further search for a better solution in the neighborhood of the current optimal solution. This behavior simulates the spiral movement of the seahorse around an optimal position (such as the optimal path point). The spiral motion enables the drone to accurately optimize the path point by continuously reducing the search radius. At the same time, Lévy Flight is used to generate random step sizes, so that the search can be adaptively expanded or reduced. The spiral motion is mainly used for local development. The randomness of Lévy Flight ensures that the search process has a certain global exploration ability and avoids local optimization traps.

[0134] Specifically, the spiral motion simulates the path adjustment of the drone through a dynamic update mechanism, which is mathematically expressed as follows:

[0135] ;

[0136] In the formula, Indicates Flight paths The The track point is The position after iterations; They represent the coordinate components of the spiral motion in three-dimensional space, and these three components represent the position changes of the UAV in three-dimensional space during the spiral motion; Indicates Flight paths The The position of the track point at the tth iteration; represents the optimal solution found in the current t-th iteration, that is, the optimal path point; represents the Levy flight distribution, which is a random step generation mechanism; Represents the radius of the spiral motion, which determines the precision of the waypoint search.

[0137] Among them, Levy flight is a heavy-tailed distribution that allows long jump searches, which helps global exploration. The mathematical representation of Levy flight distribution is:

[0138] ;

[0139] In the formula, and Represents a random number, A parameter to control the step size, usually between (1,2); is the standard deviation of the distribution; Represents the displacement scale in path search, which is used to adjust the size of the generated random step and control the locality or jumpiness of the search process. Through this Levy distribution, the algorithm search process has non-uniformity, which can improve the probability of finding the global optimal solution.

[0140] By adjusting the radius of the spiral motion , the algorithm can flexibly switch between large-scale global exploration and small-scale local search. The radius of the spiral motion Defined as:

[0141] ;

[0142] In the formula, and is a constant that controls the compactness and amplitude of the spiral; is a A random angle within , which determines how much the spiral rotates.

[0143] Predation behavior is a dynamic adjustment mechanism that adaptively switches between global search and local search. It is considered a combination of local and global optimization strategies, which ensures that the drone can strike a balance between exploring new possible paths (global search) and making precise adjustments on the current path (local development). The predation behavior in the seahorse optimization algorithm simulates the strategy of seahorses when catching prey in nature. It adjusts the exploration and development direction of the path based on the random success or failure of predation results. The core idea is to simulate the drone's exploration of potential optimal path points. When the drone successfully "captures" a better path, the algorithm will tend to continue to optimize according to the current local search optimal solution; when predation fails, the drone will abandon the current path and switch to global search to find new path points. This strategy effectively avoids the drone from falling into local optimality and promotes global search.

[0144] Specifically, its mathematical expression is as follows:

[0145] ;

[0146] In the formula, Indicates Flight paths The The track point is The position after iterations; Indicates that in the current In the iteration, the The candidate solution with the best performance among the track points corresponds to the path with the lowest objective function value; Represents a A random number uniformly distributed within; Indicates the current In the iteration, The first path The coordinates of the track points, i.e. the solution before the predation behavior update; It represents the parameter that controls the search step size. It decreases with the number of iterations. It determines the update amplitude between the current solution and the optimal solution. It is used to adjust the update speed of the UAV flight path points in path planning. Indicates that in the interval The random number within determines the success or failure of predation.

[0147] when When , the predation is successful, indicating that the drone has found a better path segment or path point. At this time, the algorithm tends to move the path point to the current optimal path. In path planning, successful predation is equivalent to the drone finding an ideal track point at a certain location, which can significantly reduce the flight distance, reduce the turning angle or avoid obstacles. At this time, the drone will accelerate to approach this more optimal track point and reduce further exploration by reducing the search step size, focusing on refining the path in this local area.

[0148] when When , the predation fails, indicating that the current path is not optimal, or the drone has failed to find a better path point. In this case, the drone needs to continue the global search and explore new path points. This behavior is equivalent to the drone giving up the current path segment or track point in path planning. The algorithm is more inclined to increase the search range and step size, thereby improving the global search capability to find a better global solution.

[0149] Reproduction behavior mainly generates new candidate solutions by combining the excellent characteristics of individual seahorse parents, thereby maintaining population diversity and enhancing global search capabilities. In drone path planning, reproduction behavior is equivalent to generating new possible paths based on the experience of historical paths. In three-dimensional scenarios, when faced with multiple paths, complex obstacle distribution, and multi-dimensional constraints, reproduction behavior will introduce new track points and inherit the excellent characteristics from different paths through crossover and mutation processes. Explore potential excellent paths globally, thereby providing drones with more flight options and avoiding falling into local optimality.

[0150] Specifically, its mathematical expression is as follows:

[0151] ;

[0152] In the formula, Indicates Flight paths No. The sub-solution of the track points is the new track points of the UAV; and Respectively represent the parent path The coordinates of each track point are usually selected based on the objective function in path planning to select the best individual in the group as the parent solution; Indicates that in the interval The random number between represents the influence weight of the father individual and the mother individual in the offspring solution. Its randomness ensures the diversity of the solution, so that the generated offspring solution can cover a wider solution space.

[0153] For the selection of three optimization strategies, after the initial path is generated, the optimization algorithm performs spiral motion and predation behavior. The two mechanisms are flexibly selected in each iteration according to random parameters and search progress, and are executed alternately or mixed. The Hippocampus optimization algorithm uses normal distribution random numbers to decide which strategy to execute. After each round of predation behavior, the algorithm divides the population into parent solutions and mother solutions according to the fitness of the path solution. The offspring path and the parent solution set form a new population to replace the path with poor fitness. After continuous iterations, the objective function gradually approaches the global optimal solution.

[0154] Termination conditions and rules of the Hippocampus optimization algorithm:

[0155] Maximum number of iterations: When the number of iterations reaches the preset maximum number of iterations T, the algorithm stops.

[0156] Convergence judgment: If the objective function of the path solution has no obvious improvement after multiple iterations, it will be terminated early. The mathematical expression is as follows:

[0157] ;

[0158] In the formula, represents the latest objective function; Represents the last objective function of the latest objective function; Indicates the preset convergence threshold.

[0159] It should be noted that the maximum number of iterations and convergence threshold It can be set by experts or according to task requirements and historical experience. The range of this value varies significantly in different tasks and scenarios.

[0160] In the optimized UAV path planning scheme, the output results determine the optimized UAV path planning scheme , outputs the minimized objective function value , whose value is used to measure the optimization degree of the UAV path planning solution, and outputs the actual number of iterations Based on the above output results, the system can further adjust the optimization parameters and recalculate to provide a feasible solution to the UAV path planning problem.

[0161] Compared with the prior art, the UAV path planning method based on the Hippocampus optimization algorithm provided in the embodiment of the present invention has the following advantages:

[0162] 1. Geographic information integration under multiple constraints. By constructing a 1-meter resolution three-dimensional grid, the terrain, buildings, no-fly zones, signal interference zones, and security threat zones are accurately mapped to the flight area model, achieving detailed modeling of complex flight environments and integration of multiple constraints. This design is superior to the rough three-dimensional description of the flight area in existing technologies, effectively improving the accuracy of path planning and more in line with real-life scenarios.

[0163] 2. Assign safety threat areas to obstacles. By assigning safety threat areas to all obstacles, it ensures that the drone maintains sufficient safety redundancy during flight. Compared with only considering obstacle avoidance, considering the potential risks of the drone is more in line with the real scene.

[0164] 3. Path planning with multi-index comprehensive optimization. By comprehensively considering the impact of path length, safety threats, flight turning angles, NIMBY effect, and land type, a multi-objective optimization model is constructed to achieve a balance between flight safety, energy consumption, stability, and resident interference. Unlike existing technologies that only focus on the shortest path or simple obstacle avoidance, this model is more suitable for complex mission scenarios and ensures the global optimality of the flight path.

[0165] 4. Hippocampus optimization algorithm solves combinatorial optimization problems. The Hippocampus optimization algorithm is introduced to solve the nonlinear and non-convex combinatorial optimization problems of UAV path planning. It achieves a balance between global exploration and local development through spiral motion, predation behavior and reproduction process, so as to more effectively explore a wider solution space, avoid falling into the local optimal solution, and provide a more efficient solution. Compared with traditional algorithms such as A* and genetic algorithms, the Hippocampus optimization algorithm has a higher convergence speed and search efficiency when dealing with nonlinear and multi-objective problems.

[0166] 5. Weight adjustment mechanism. By adjusting the weights of various indicators in path planning according to the use scenarios and purposes of the drone, a flexible path planning mechanism is provided, which enables the drone system to show higher flexibility and robustness in multi-task and dynamic environments, and effectively solves the problem that traditional algorithms are difficult to adapt to different scenarios in tasks and have low flexibility.

[0167] Based on the same inventive concept, the embodiment of the present invention also provides a UAV path planning device based on the Hippocampus optimization algorithm, such as Figure 2 As shown, the device comprises:

[0168] A flight cost function acquisition module is used to construct a three-dimensional space grid of the UAV flight area, map the geographic information of the UAV flight area to the three-dimensional space grid, and construct a flight cost function constrained by different factors of the flight path. Each track point in the flight path corresponds to a three-dimensional space grid; the flight costs constrained by different factors include at least: the flight cost constrained by basic factors, the flight cost constrained by the NIMBY effect, and the flight cost constrained by the land use of the flight path;

[0169] An objective function acquisition module is used to set corresponding cost weights for flight costs constrained by different factors, perform weighted fusion on the flight costs constrained by different factors according to the cost weights, and construct an objective function for the flight path of the UAV;

[0170] The flight path acquisition module is used to optimize and solve the objective function of the UAV flight path to determine the optimal flight path of the UAV.

[0171] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0172] Based on the same inventive concept, the embodiment of the present invention also provides a drone path planning system based on the Hippocampus optimization algorithm, such as Figure 3 As shown, the system includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the system can execute any of the drone path planning methods based on the Hippocampus optimization algorithm introduced above.

[0173] The embodiment of the present invention can divide the system into functional modules according to the above method example. For example, each functional module can be corresponded to, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0174] Based on the same inventive concept, an embodiment of the present invention also provides a computer program product, which includes: a computer program code, when the computer program code runs on a computer, enables the computer to execute any one of the drone path planning methods based on the Hippocampus optimization algorithm introduced above.

[0175] Based on the same inventive concept, an embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program code. When the computer program code runs on a computer, the computer executes any one of the drone path planning methods based on the Hippocampus optimization algorithm introduced above.

[0176] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A UAV path planning method based on the Hippocampus optimization algorithm, characterized in that: The following steps are involved: Construct a three-dimensional spatial grid of the UAV flight area, map the geographic information of the UAV flight area to the three-dimensional spatial grid, and construct a flight cost function constrained by different factors of the flight path, where each track point in the flight path corresponds to a three-dimensional spatial grid; the flight costs constrained by different factors at least include: the flight cost constrained by basic factors, the flight cost constrained by the NIMBY effect, and the flight cost constrained by the land use of the flight path; Setting corresponding cost weights for flight costs constrained by different factors, weightedly integrating the flight costs constrained by different factors according to the cost weights, and constructing an objective function of the UAV flight path; Optimizing and solving the objective function of the UAV flight path to determine the optimal flight path of the UAV; The flight cost constrained by basic factors at least includes: the flight cost constrained by flight distance, the flight cost constrained by safety threat area, and the flight cost constrained by flight angle; The calculation formula of the flight cost of the flight angle constraint is: ； ； ； In the formula, Indicates Flight paths The flight cost constrained by the flight angle; and represents the penalty term; Indicates Flight paths The The horizontal turning angle corresponding to each track point; and Respectively represent Flight paths The Track points and The vertical pitch angle corresponding to each track point; Indicates Flight paths The total number of track points in ; and Respectively represent The flight path Track points , No. Track points and Track points The two consecutive flight segments formed and exist Projection on a plane; and Respectively expressed in Flight paths The Track points and Track points The vertical coordinate in the three-dimensional coordinates of ; Indicates the vector modulus length symbol.

2. The method for UAV path planning based on the Hippocampus optimization algorithm according to claim 1, characterized in that: The calculation formula of the flight cost of the NIMBY effect constraint is: ； In the formula, Indicates Flight paths The flight cost constrained by the NIMBY effect; Indicates Flight paths The number of demographic rasters within the buffer of the current track point; Indicates Flight paths The current track point in the buffer population count for each demographic raster; Indicates the flight path The current track point in the buffer The demographic grid and Flight paths The area of ​​the intersection of the current track point buffer.

3. The UAV path planning method based on the Hippocampus optimization algorithm according to claim 1 is characterized in that: The process of determining the flight cost of the flight path land constraint includes: Set corresponding land use index weights for different types of land; Determine the distance between each trackpoint in the flight path and the next trackpoint after it; According to the land use index weight corresponding to the land type of each track point in the flight path, the distances corresponding to each track point in the flight path are weighted and fused to obtain the flight cost constrained by the land use of the flight path.

4. The method for UAV path planning based on the Hippocampus optimization algorithm according to claim 3 is characterized in that: The calculation formula of the flight cost of the flight path land constraint is: ； In the formula, Indicates Flight paths Flight costs due to flight path land constraints; Indicates Flight paths The Track points The land use index weight corresponding to the land type; Indicates Flight paths The Track points With indivual The Euclidean distance between Indicates Flight paths The total number of track points in .

5. The method for UAV path planning based on Hippocampus optimization algorithm according to claim 1, characterized in that: The calculation formula of the flight cost of the flight distance constraint is: ； In the formula, Indicates Flight paths The flight cost constrained by the flight distance; Indicates Flight paths The Track points and Track points The Euclidean distance between Indicates Flight paths The total number of track points in .

6. The method for UAV path planning based on Hippocampus optimization algorithm according to claim 1, characterized in that: The calculation formula of the flight cost constrained by the security threat zone is: ； In the formula, Indicates Flight paths The flight cost constrained by the security threat zone; Indicates Flight paths The total number of track points in ; Indicates the total number of obstacles in the flight environment; Indicates Flight paths Segments in Relative The threat cost of an obstacle; Defined as: ； In the formula, represents the obstacle avoidance penalty coefficient; Indicates the flight environment The distance between the center of mass of an obstacle and its surface; Indicates the set security threat distance; Indicates the first obstacle centroid to flight segment The shortest distance; and Respectively represent Flight paths The Track points and Track points.

7. The method for unmanned aerial vehicle path planning based on the Hippocampus optimization algorithm according to claim 1, characterized in that: The objective function of the UAV flight path is optimized and solved to determine the left and right flight paths of the UAV, including: An artificial potential field method is used to generate a set of initial path solutions; Based on the initial path solution, the Hippocampus optimization algorithm is used to optimize and solve the objective function of the UAV flight path to determine the left and right flight paths of the UAV.

8. The method for unmanned aerial vehicle path planning based on the Hippocampus optimization algorithm according to claim 7, characterized in that: The calculation formulas for the repulsive potential field and the gradient of the repulsive force in the repulsive potential field in the artificial potential field method are: ； In the formula, Indicates Flight paths The current Track points The corresponding repulsive potential field; represents the repulsion coefficient; Indicates Flight paths The current Track points The first Obstacles distance; Indicates the flight environment Obstacles The distance of the center of mass from its surface; Indicates the set security threat distance; ； In the formula, Indicates Flight paths The current Track points The corresponding gradient of the repulsive force; Indicates the total number of obstacles in the flight environment.

Images