Unmanned aerial vehicle multi-task-point cruise path planning method and device

Abstract

The invention relates to a multi-task-point cruise path planning method and device for an unmanned aerial vehicle. The method comprises the following steps: acquiring flight data of the unmanned aerial vehicle in multiple flight stages; the flight data comprises corner data when the flight stage is a corner stage; constructing a flight energy consumption model according to the undirected graph, the flight data and the rotation angle data; the energy consumption of each task point in the cruising process of the unmanned aerial vehicle is obtained; performing iterative optimization on the energy consumption of the plurality of task points based on a wild dog optimization algorithm to obtain a path planning result, and controlling the unmanned aerial vehicle to fly according to the path planning result; energy consumption of different flight stages is quantified by constructing a flight energy consumption model, including energy consumption of rotation angle data in a rotation angle stage, so that the complex environment adaptability is improved, iterative optimization is performed on a path in combination with a wild dog optimization algorithm, the problems of inaccurate energy consumption modeling and easy falling into local optimum are solved, and the energy consumption modeling efficiency is improved. The method has the advantages that the energy consumption optimization efficiency and accuracy of path planning are improved, and the local optimum problem is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle path planning, and particularly relates to a method and device for unmanned aerial vehicle multi-task point cruise path planning. BACKGROUND

[0002] The rapid development of technology and the increasing social demand make unmanned aerial vehicle technology gradually become a tool for solving various problems, such as data collection, logistics distribution, security patrol, and disaster response. With the expansion of application fields and the increase of complexity, the demand for unmanned aerial vehicle path planning in multi-task point cruise operations is increasingly prominent, which includes efficiently completing multiple tasks within a limited time, optimizing flight paths to reduce energy consumption, and ensuring the safety and accuracy of tasks. Therefore, the multi-task point path planning problem of unmanned aerial vehicles has become a research hotspot, and it is particularly important to improve the flexibility and optimization performance of path planning algorithms in order to efficiently and accurately perform tasks in a multi-task point operation environment. Optimization of multi-task unmanned aerial vehicle path planning algorithms has become a core direction in the field of unmanned aerial vehicle task allocation.

[0003] Existing unmanned aerial vehicle path planning algorithms can be mainly divided into classical algorithms and heuristic algorithms. The classical algorithms are not suitable for high-dimensional complex path planning problems. The heuristic algorithms still have limitations when dealing with multi-task point cruise operations.

[0004] Therefore, it is urgent to propose a method and device for unmanned aerial vehicle multi-task point cruise path planning to solve the technical problem that the existing unmanned aerial vehicle path planning algorithm has limitations when performing multi-task point cruise operations and cannot take into account the adaptability of complex environments. SUMMARY

[0005] Therefore, it is necessary to provide a method and device for unmanned aerial vehicle multi-task point cruise path planning to solve the technical problem that the existing unmanned aerial vehicle path planning algorithm has limitations when performing multi-task point cruise operations and cannot take into account the adaptability of complex environments.

[0006] To solve the above problems, in a first aspect, the present application provides a method for unmanned aerial vehicle multi-task point cruise path planning, comprising: setting multiple task points and obtaining flight data of the unmanned aerial vehicle in multiple flight stages; the flight data includes turn angle data when the flight stage is a turn angle stage; converting the cruise operation of the unmanned aerial vehicle at the multiple task points into an undirected graph, and constructing a flight energy consumption model according to the undirected graph, the flight data, and the turn angle data; obtaining the energy consumption of each task point of the unmanned aerial vehicle in the cruise process according to the flight energy consumption model; The energy consumption of the plurality of task points is iteratively optimized based on a wild dog optimization algorithm to obtain a path planning result, and the UAV is controlled to fly according to the path planning result.

[0007] In a possible implementation, the flight energy consumption model is constructed according to the undirected graph, the flight data and the turning angle data, including: According to the flight data, a basic force of each flight phase is obtained; According to the basic force, the flight data and the turning angle data, a total aerodynamic power required by the UAV when flying at a constant speed is obtained; The flight energy consumption model is constructed according to the undirected graph and the total aerodynamic power.

[0008] In a possible implementation, the plurality of flight phases further include a take-off phase, a horizontal flight phase, a descent phase and a hovering phase; and the total aerodynamic power required by the UAV when flying at a constant speed is obtained according to the basic force and the flight data, including: According to the basic force and the flight data, a parasitic power, an induced power, a profile power and a climb power corresponding to each flight phase are obtained, and a turning energy consumption is obtained according to the turning angle data; A first power is obtained according to the parasitic power, the induced power, the profile power and the climb power of the take-off phase; A second power is obtained according to the parasitic power, the induced power and the profile power of the horizontal flight phase; A third power is obtained according to the parasitic power, the induced power, the profile power and the climb power of the descent phase; A fourth power is obtained according to the parasitic power, the induced power and the profile power of the hovering phase; The total aerodynamic power of the UAV when flying between adjacent task points is obtained according to the first power, the second power, the third power, the fourth power, the flight data and the turning energy consumption.

[0009] In a possible implementation, the energy consumption of the plurality of task points is iteratively optimized based on a wild dog optimization algorithm to obtain a path planning result, including: A mapping relationship between a number of each task point and a position thereof is set; The number and the position of the plurality of task points of the navigation path of the UAV in the undirected graph are initialized according to the mapping relationship to obtain initialization data; The initial data is iteratively optimized based on the Wild Dog optimization algorithm, the adaptive t-distribution mutation strategy, and the energy consumption of the multiple task points to obtain the path planning result.

[0010] In one possible implementation, the iterative optimization of the initial data based on the Wild Dog optimization algorithm, the adaptive t-distribution mutation strategy, and the energy consumption of the multiple task points to obtain the path planning result includes: The initial data is optimized based on the wild dog optimization algorithm and the energy consumption of the multiple task points to obtain the optimal individual position in the current population; According to the adaptive t Adaptive strategies in distribution mutation The distribution function and the optimal individual position are used to perturb the location of the wild dog to obtain the perturbed target wild dog location; Determine whether the current iteration count of the population is the maximum iteration count; If not, the location of the target wild dog is iteratively optimized according to the wild dog optimization algorithm. If so, the path planning result is obtained based on the optimal individual position output in each iteration.

[0011] In one possible implementation, the initialization of the numbering and position of the plurality of task points on the flight path of the UAV in the undirected graph according to the mapping relationship to obtain initialization data includes: Confirm the randomly generated initial position, and determine the target task point based on the principle of proximity according to the initial position; Substitute the position data of the target task point in the mapping relationship into the SPM chaotic mapping function for calculation to obtain the iterative position of the target task point; Determine whether the task point in the undirected graph has completed iteration; If not, the iteration position is calculated according to the SPM chaotic mapping function to obtain the updated iteration position; If so, the initialization data is obtained based on all iteration positions calculated in the iterative process.

[0012] In one possible implementation, the flight data includes hovering time, takeoff altitude, climb angle, UAV flight speed, descent altitude, and Euclidean distance between adjacent mission points; obtaining the total aerodynamic power of the UAV flying between adjacent mission points based on the first power, the second power, the third power, the fourth power, the flight data, and the turning energy consumption includes: The takeoff time is obtained based on the takeoff altitude, the climb angle, and the UAV flight speed. The horizontal flight time is obtained based on the takeoff altitude, the climb angle, the UAV flight speed, the descent altitude, and the Euclidean distance between adjacent mission points; The descent time is obtained based on the descent altitude, the climb angle, and the UAV flight speed; The total aerodynamic power of the UAV flying between adjacent mission points is obtained based on the first power, the second power, the third power, the fourth power, the takeoff time, the horizontal flight time, the descent time, the hovering time, and the turning energy consumption.

[0013] In one possible implementation, the SPM chaotic mapping function is as follows:

[0014] In the formula, Represents the randomly generated first... The initial location of the stray dog; Indicates the first The position of the wild dog in the iteration after the first iteration; Represents a random number between 0 and 1; when and At that time, the function is in a chaotic state.

[0015] In one possible implementation, the flight energy consumption model is as follows:

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

[0022] In the formula, C represents the set of task points. , , Indicates task point arrive The distance is calculated by the task point. arrive It is obtained by the Euclidean distance between them. This represents a route consisting of task point numbers; when the drone departs from the task point... Moving to task point Time 1 if the task point is reached and only once, otherwise 0; Indicates that each task point is reached and only once; Indicates that each task point is left and only once; Indicates whether the edge (i, j) is selected in the optimal solution; Indicates that the sub-loop is eliminated; Indicates that the battery capacity of the UAV is greater than the sum of the flight energy consumption and the turning energy consumption; Indicates that the distance the UAV needs to fly forward before adjusting the flight attitude is greater than or equal to the minimum distance .

[0023] In a second aspect, the present application also provides a UAV multi-task point cruise path planning device, comprising: A data acquisition module is configured to set multiple task points and acquire flight data of the UAV in multiple flight stages; the flight data includes turning data when the flight stage is a turning stage; A model construction module is configured to convert the cruise operation of the UAV at the multiple task points into an undirected graph, and construct a flight energy consumption model according to the undirected graph, the flight data and the turning data; An energy consumption confirmation module is configured to obtain the energy consumption of each task point of the UAV in the cruise process according to the flight energy consumption model; An iterative optimization module is configured to perform iterative optimization on the energy consumption of the multiple task points based on a wild dog optimization algorithm, obtain a path planning result, and control the UAV to fly according to the path planning result.

[0024] The present application has the advantages that multiple task points are set, flight data of the UAV in multiple flight stages is acquired, turning data when the flight stage is a turning stage is acquired, the cruise operation of the UAV at the multiple task points is converted into an undirected graph, a flight energy consumption model is constructed according to the undirected graph and the flight data, the energy consumption of each task point of the UAV in the cruise process is obtained according to the flight energy consumption model, iterative optimization is performed on the energy consumption of the multiple task points based on a wild dog optimization algorithm, a path planning result is obtained, and the UAV is controlled to fly according to the path planning result; the energy consumption of different flight stages is quantified by constructing the flight energy consumption model, including the energy consumption of the turning data when the turning stage, thereby improving the adaptability to complex environments, and the path is iteratively optimized in combination with the wild dog optimization algorithm, thereby solving the problems of inaccurate energy consumption modeling and being prone to local optimization in the prior art, improving the energy consumption optimization efficiency and accuracy of path planning, avoiding the local optimization problem, and improving the endurance of the UAV. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 An embodiment flow chart of the UAV multi-task point cruise path planning method provided by the present application is shown in the figure; Figure 2 An embodiment structure diagram of the UAV flight phase provided by the present application is shown in the figure; Figure 3 An embodiment structure diagram of the UAV cruise operation provided by the present application is shown in the figure; Figure 4 An embodiment flow chart of the total aerodynamic power calculation process provided by the present application is shown in the figure; Figure 5 An embodiment structure diagram of the wild dog position to actual route mapping process provided by the present application is shown in the figure; Figure 6 An embodiment structure diagram of the two-dimensional distribution of task points provided by the present application is shown in the figure; Figure 7 An embodiment structure diagram of the total energy consumption of the optimal path planning cruise provided by the present application is shown in the figure; Figure 8 An embodiment structure diagram of the corner energy consumption of the optimal path planning provided by the present application is shown in the figure; Figure 9 An embodiment structure diagram of the cruise average energy consumption of different task points provided by the present application is shown in the figure; Figure 10 An embodiment structure diagram of the UAV multi-task point cruise path planning device provided by the present application is shown in the figure. DETAILED DESCRIPTION

[0026] The preferred embodiments of the present application will be specifically described below in conjunction with the accompanying drawings, wherein the drawings form a part of the present application, and are used to explain the principles of the embodiments of the present application, and are not used to limit the scope of the present application.

[0027] As shown in the figure, one specific embodiment of the present application discloses a UAV multi-task point cruise path planning method, which comprises: Figure 1 S101, a plurality of task points are set, and flight data of a UAV in a plurality of flight phases is acquired; the flight data comprises corner data when the flight phase is a corner phase.

[0028] ​In the unmanned aerial vehicle multi-task point cruise path planning method provided by the embodiments of the application, the unmanned aerial vehicle multi-task point cruise path planning system can be a software system running on a terminal device. The terminal device can be a server, a tablet computer, an augmented reality (AR) / virtual reality (VR) device, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a mobile phone, or the like. The embodiments of the application do not limit the specific type of the terminal device.

[0029] In specific embodiments, the plurality of flight stages can include four stages, namely a take-off stage, a horizontal flight stage, a descent stage, and a hovering stage, as shown in Figure 2 As shown in the figure, first, the unmanned aerial vehicle takes off from the ground at a uniform speed to a certain height, so that the unmanned aerial vehicle avoids the interference of obstacles in a complex environment, then maintains a uniform horizontal flight, and then descends at a uniform speed to a flight height suitable for operation when reaching the task point, and then hovers above the task point to perform the operation. The flight stage considers the use of a single unmanned aerial vehicle to perform multi-task point cruise operation. During the flight, the unmanned aerial vehicle needs to turn between the task points. Therefore, in addition to the four basic states, the unmanned aerial vehicle also needs to consider the turning stage, so as to obtain the data of all states of the unmanned aerial vehicle during the flight. Considering that the unmanned aerial vehicle needs to turn between the task points, a turning state is introduced.

[0030] S102, converting the cruise operation of the unmanned aerial vehicle at the plurality of task points into an undirected graph, and constructing a flight energy consumption model according to the undirected graph, flight data, and turning data.

[0031] The undirected graph refers to a mathematical model for describing the connection relationship between the task points. Specifically, the connection relationship between the task points can be realized by adopting a mode in which a node represents a task point and an edge represents a flight path, and the reachability between the task points is reflected through a topological structure. The flight energy consumption model refers to a mathematical expression for quantifying the energy consumption of the unmanned aerial vehicle in different flight stages. Specifically, the flight energy consumption model can be realized by calculating the product of the aerodynamic power and time in stages, and is used to accurately evaluate the total energy consumption of the voyage.

[0032] In specific embodiments, when the unmanned aerial vehicle performs the multi-task point cruise operation, as shown in Figure 3As shown, the UAV will take off from the site, sequentially traverse each task point, hover over each task point for work, continue to fly to the next task point after the work is completed, and return to the starting point after all task points are completed. Considering the energy constraint of the UAV flight, in order to improve the cruising ability of the UAV, how to plan an efficient flight path to complete the work of each task point becomes the key. The mathematical model of path planning can be represented by an undirected graph . represents a set of vertices, where is a special work point, representing a base station (starting point and ending point). is a set of edges, representing the path link between work points; is the battery capacity limit value of the UAV.

[0033] S103, obtaining the energy consumption of the UAV at each task point in the cruising process according to the flight energy consumption model.

[0034] After the flight energy consumption model is constructed, the energy consumption of each task point can be predicted when the UAV is cruising.

[0035] S104, iteratively optimizing the energy consumption of multiple task points based on the wild dog optimization algorithm, obtaining a path planning result, and controlling the UAV to fly according to the path planning result.

[0036] Wherein, the energy consumption of multiple task points is iteratively optimized by the wild dog optimization algorithm, and the wild dog optimization algorithm is an improved wild dog optimization algorithm, which introduces SPM chaotic mapping method and adaptive t-distribution variation strategy, so as to enhance the search ability of the algorithm in complex solution space.

[0037] Specifically, the cruising path of the UAV between multiple task points is abstracted as nodes and edges in an undirected graph, and by collecting flight time and force data in the take-off, horizontal flight, descent and hovering stages, the parasitic power, induced power and other aerodynamic power components are calculated in stages. Based on the product of the power and duration of each stage, the flight energy consumption model is constructed to obtain the energy consumption value of the flight between task points. Then, the improved wild dog optimization algorithm is used to iteratively optimize the task point sequence, generate an initial solution set through chaotic mapping, and introduce adaptive disturbance to jump out of local optimum, and finally output the flight path with the lowest energy consumption.

[0038] Compared with the prior art, the embodiment provided sets multiple task points, obtains flight data of the unmanned aerial vehicle in multiple flight stages, converts cruising operation of the unmanned aerial vehicle at the multiple task points into an undirected graph, and constructs a flight energy consumption model according to the undirected graph and the flight data; the energy consumption of each task point in the cruising process of the unmanned aerial vehicle is obtained according to the flight energy consumption model; the energy consumption of the multiple task points is iteratively optimized based on a wild dog optimization algorithm to obtain a path planning result, and the unmanned aerial vehicle is controlled to fly according to the path planning result; the energy consumption of different flight stages is quantified by constructing the flight energy consumption model, including the energy consumption of the turning angle data in the turning angle stage, thereby improving the complex environment adaptability, and the path is iteratively optimized in combination with the wild dog optimization algorithm, thereby solving the problems of inaccurate energy consumption modeling and being prone to local optimization in the prior art, and the energy consumption optimization efficiency and accuracy of path planning are improved, the problem of local optimization is avoided, and the endurance of the unmanned aerial vehicle is improved.

[0039] In some embodiments of the application, step S102 comprises: According to the flight data, the basic force of each flight stage is obtained.

[0040] The basic force can include gravity, lift, drag and thrust, the gravity is generated by the mass of the unmanned aerial vehicle and the earth's gravity, and acts downward. The lift is used to overcome the gravity to maintain the stable position of the unmanned aerial vehicle in the air. In horizontal flight, the lift and the gravity need to be balanced to prevent the unmanned aerial vehicle from falling. The drag is the air reaction force received by the unmanned aerial vehicle, and the thrust is used to offset the drag to maintain the constant speed of the unmanned aerial vehicle to achieve dynamic balance.

[0041] According to the basic force, the flight data and the turning angle data, the total aerodynamic power required by the unmanned aerial vehicle in uniform flight is obtained.

[0042] The total aerodynamic power refers to the energy required to overcome air resistance to maintain the uniform flight of the unmanned aerial vehicle, which can be specifically realized by vector product operation of the basic force and the flight speed. This parameter reflects the energy demand characteristics of the unmanned aerial vehicle in different flight attitudes.

[0043] In specific embodiments, the force components can be vector operated with the real-time flight speed to derive parasitic power, induced power and other aerodynamic power components. The turning angle data can also be processed to obtain the energy consumption in the turning angle stage.

[0044] According to the undirected graph and the total aerodynamic power, a flight energy consumption model is constructed.

[0045] According to the undirected graph and the total aerodynamic power, a mathematical model of single unmanned aerial vehicle multi-task point cruising operation problem including turning angle energy consumption factors is established, that is, the flight energy consumption model, as shown in formulas (1)-(7): (1) (2) (3) (4) (5) (6) (7) In the formula, C represents a set of task points , , represents the distance from the task point to , which is obtained by calculating the Euclidean distance between the task points and , represents a set of routes composed of task point numbers; when the unmanned aerial vehicle moves from the task point to the task point , is 1, otherwise 0; represents that each task point is reached and only once; represents that each task point is left and only once; represents whether the edge (i, j) is selected in the optimal solution; represents the elimination of sub-loops; represents that the battery capacity of the unmanned aerial vehicle is greater than the sum of the range energy consumption and the corner energy consumption; represents that the distance to be flown in front of the unmanned aerial vehicle before adjusting the flight attitude is greater than or equal to the minimum distance .

[0046] Specifically, in the construction of the flight energy consumption model, first, the speed, acceleration, and attitude angle parameters in the flight data are analyzed, and the lift, drag, and lateral force components corresponding to each flight stage are calculated using fluid mechanics formulas. Then, the force components are vector operated with the real-time flight speed, and the aerodynamic power components such as parasitic power and induced power are derived. Further, combining the navigation path topological relationship described by the undirected graph, the aerodynamic power components of each navigation section are spatially superimposed, and finally an energy consumption quantification model covering the entire navigation section is established.

[0047] Through the above technical scheme, the present application realizes the fine energy consumption modeling of complex flight conditions in multi-task point cruise operation. By calculating the aerodynamic power components in stages and superimposing spatial topological constraints, the energy consumption estimation error problem caused by the traditional method without distinguishing flight stages is solved, providing accurate energy consumption data basis for subsequent path optimization algorithms.

[0048] In some embodiments of the application, the plurality of flight phases further include a take-off phase, a horizontal flight phase, a descent phase, and a hovering phase. Figure 4 As shown, the total aerodynamic power required by the UAV when flying at a constant speed is obtained according to the basic acting force and flight data, including: S401, according to the basic acting force and flight data, the parasitic power, induced power, profile power and climbing power corresponding to each flight phase are obtained, and the turning energy consumption is obtained according to the turning angle data.

[0049] Wherein, the plurality of flight phases include a take-off phase, a horizontal flight phase, a descent phase, and a hovering phase; the parasitic power refers to the energy required by the UAV to overcome air resistance, which can be calculated by multiplying the air density, the square of the flight speed and the reference area, and is used to quantify the energy loss caused by aerodynamic resistance in the flight process. The induced power refers to the energy required to generate lift, which can be calculated by the function relationship between the rotor speed and the blade angle, and is used to represent the power consumption required to maintain the flight height. The profile power refers to the energy consumption corresponding to the additional resistance caused by the shape of the UAV, which can be calculated by using the empirical formula based on the fuselage streamline coefficient, and reflects the influence of fuselage design on aerodynamic efficiency. The climbing power refers to the energy required by the UAV to resist gravity during vertical ascent or descent, which can be calculated by multiplying the flight height change rate and gravity, and is used to evaluate the energy demand in the height adjustment stage. The turning energy consumption refers to the additional energy consumption caused by centrifugal force when changing the flight direction, which can be calculated by using the relationship between the turning radius in the turning angle data and the square of the angular velocity, and is used to quantify the power loss caused by heading adjustment.

[0050] In specific embodiments, to simplify the flight model of the UAV, it is assumed to fly at a constant speed, mainly including the total aerodynamic power of uniform linear motion and the turning angle power of uniform turning motion.

[0051] The calculation process of the total aerodynamic power required by the UAV when maintaining a constant speed is as follows: (1) Parasitic power Parasitic power is the dot product of body drag and velocity vector, and its calculation method is shown in formula (8): (8) In the formula, , is the reference area, is the body drag coefficient, is the airspeed.

[0052] (2) Induced power Caused by rotor lift, the specific calculation method is shown in formula (9): (9) In the formula, thrust , This represents the wash coefficient.

[0053] (3) Line power Caused by rotor airfoil drag, the specific calculation formula is shown in formula (10): (10) In the formula, atmospheric density =1.125kg / m3; Rotor disk area Tip speed ; Indicates the number of blades; Indicates the rotor's mean chord; It is the average lift coefficient of the blade; Rotor radius It refers to the number of rotors; the solid-to-disk ratio. ; It is the average drag coefficient of the blade section; propulsion rate. .

[0054] (4) Climbing power The dot product of velocity vector and velocity is calculated using the formula shown in formula (11): (11) When a drone maintains a constant speed of flight, turning consumes energy. This embodiment of the invention uses a small drone, and the formula for calculating the energy consumption (i.e., turning energy consumption) during its turning process is shown in formula (12): (12) In the formula, θ Indicates the turning angle.

[0055] S402. Based on the parasitic power, induced power, line power, and climb power during the takeoff phase, the first power is obtained.

[0056] in, This represents the first power during the takeoff phase, which is the sum of the parasitic power, induced power, profile power, and climb power calculated during the takeoff process of the UAV.

[0057] S403. Based on the parasitic power, induced power, and profile power during the level flight phase, the second power is obtained.

[0058] in, This represents the second power during the horizontal flight phase, which is the sum of the parasitic power, induced power, and profile power calculated by the UAV during horizontal flight.

[0059] S404. Based on the parasitic power, induced power, profile power, and climbing power during the descent phase, the third power is obtained.

[0060] in, This represents the third power during the uniform descent phase, which is the sum of the parasitic power, induced power, profile power, and climb power calculated by the UAV during its descent.

[0061] S405. Based on the parasitic power, induced power, and profile power during the hovering phase, the fourth power is obtained.

[0062] in, This represents the fourth power during the hovering phase, which is the sum of the parasitic power, induced power, and profile power calculated by the UAV during hovering.

[0063] S406. Based on the first power, second power, third power, fourth power, flight data, and turning energy consumption, the total aerodynamic power of the UAV during flight between adjacent mission points is obtained.

[0064] Specifically, during takeoff, the UAV must overcome both gravity and air resistance simultaneously. At this stage, parasitic power, induced power, profile power, and climb power together constitute the first power. During level flight, maintaining level flight is sufficient; the second power consists of parasitic power, induced power, and profile power. During descent, controlling the gravitational potential energy conversion process is crucial; the third power includes parasitic power, induced power, profile power, and reverse climb power. During hovering, maintaining static lift is essential; the fourth power consists of parasitic power, induced power, and profile power. The power of each stage is multiplied by the corresponding time parameters and then summed with the turning energy consumption to ultimately form the total aerodynamic power calculation model between adjacent mission points.

[0065] Through the above technical solution, this application solves the problem of range prediction failure caused by inaccurate energy consumption estimation in traditional path planning. By establishing a phased power calculation model, the actual energy consumption of the UAV under different motion states can be accurately reflected, providing accurate energy consumption data input for subsequent optimization algorithms, thereby ensuring that the path planning results meet the constraints of actual flight capabilities.

[0066] In some embodiments of the present invention, flight data includes hovering time, takeoff altitude, climb angle, UAV flight speed, descent altitude, and Euclidean distance between adjacent mission points; step S406 includes: The takeoff time is obtained based on the takeoff altitude, climb angle, and drone flight speed.

[0067] Flight data can include data from the drone at the mission point. Arrive at the mission point takeoff time Horizontal flight time descent time and hovering time Departure time The calculation is shown in formula (13): (13) In the formula, Indicates the altitude at which the aircraft takes off; Indicates the climb angle; This indicates the drone's flight speed, which is constant.

[0068] The horizontal flight time is obtained based on the takeoff altitude, climb angle, UAV flight speed, descent altitude, and Euclidean distance between adjacent mission points.

[0069] Among them, horizontal flight time The calculation is shown in formula (14): (14) In the formula, Indicates the descent altitude; This represents the Euclidean distance between task point i and task point j.

[0070] The descent time is obtained based on the descent altitude, climb angle, and drone flight speed.

[0071] Among them, descent time The calculation is shown in formula (15): (15) Based on the first power, second power, third power, fourth power, takeoff time, horizontal flight time, descent time, hovering time, and turning energy consumption, the total aerodynamic power of the UAV flying between adjacent mission points is obtained.

[0072] Among them, hovering time The calculation is shown in formula (16): (16) The flight power of a drone varies depending on its flight status during flight. Arrive at the mission point The total aerodynamic power of flight energy consumption is calculated as shown in formula (17): (17) In some embodiments of the present invention, step S104 includes: Configure a mapping relationship between the number of each task point and its location.

[0073] The mapping relationship refers to establishing a correspondence between task point numbers and spatial coordinates, which can be achieved using coordinate encoding. A unique identifier is bound to the geographical location to ensure the uniqueness and identifiability of task points during path planning. After modeling, an improved traditional Wild Dog optimization algorithm is used to find the optimal task point visit sequence. Because the solution space of the UAV path planning problem is discrete, the generated solution... It should be a route composed of node numbers. However, the Wild Dog Optimization Algorithm requires continuity of the variable space. Considering the conflict between these requirements, this embodiment of the invention discretizes the wild dog position vectors when using the Wild Dog Optimization Algorithm to solve the path planning problem. For the position of each wild dog, a... dimensional vector express, , Indicates the number of nodes. A set of positional components. Match an actual node to realize the mapping relationship from the location of the wild dog to the actual route.

[0074] The initialization data is obtained by initializing the numbering and position of multiple task points on the flight path of the UAV in the undirected graph according to the mapping relationship.

[0075] The initialization data refers to the generation of the initial population through chaotic mapping. Specifically, the SPM chaotic mapping function can be used to iteratively calculate the random initial positions, and the ergodicity and randomness of the chaotic system can be used to provide a variety of initial solution sets for the optimization algorithm.

[0076] The initial data is iteratively optimized based on the Wild Dog optimization algorithm, the adaptive t-distribution mutation strategy, and the energy consumption of multiple task points to obtain the path planning results.

[0077] Among them, the Wild Dog Optimization Algorithm refers to an intelligent optimization algorithm that simulates the hunting behavior of wild dog packs. Specifically, it updates individual positions by simulating behaviors such as encirclement, pursuit, and prey allocation, achieving a balance between global and local search. The Adaptive t-distribution mutation strategy refers to a position perturbation mechanism based on a probability distribution model. Specifically, it can apply random perturbations to the optimal individual position through a t-distribution function with the number of iterations as the degrees of freedom, enhancing the algorithm's ability to escape local optima.

[0078] Specifically, in the path planning process, a mapping table between task point numbers and locations is first established using coordinate encoding to clarify the spatial distribution characteristics of each task point. Then, the SPM chaotic mapping function is used to iteratively calculate the randomly generated initial positions, generating an initial population with uniform distribution characteristics, providing diverse candidate path schemes for subsequent optimization. The initial population is iteratively updated based on the wild dog optimization algorithm, simulating the cooperative hunting behavior of wild dog packs. The energy consumption is calculated to evaluate the path's merits, gradually approaching the optimal solution. In each iteration, an adaptive t-distribution mutation strategy is used to perturb the current optimal individual position, dynamically balancing exploration and development capabilities by adjusting the degree of freedom parameters, thus preventing the algorithm from prematurely converging to a suboptimal solution.

[0079] In some embodiments of the present invention, the numbering and position of multiple task points on the flight path of the UAV in the undirected graph are initialized according to the mapping relationship to obtain initialization data, including: Confirm the randomly generated initial position, and then determine the target task point based on the principle of proximity; Substitute the position data of the target task point in the mapping relationship into the SPM chaotic mapping function for calculation to obtain the iterative position of the target task point; Determine whether the task node in an undirected graph has completed iteration; If not, the iteration position is calculated based on the SPM chaotic mapping function to obtain the updated iteration position; If so, the initialization data is obtained based on all iteration positions calculated in the iterative process.

[0080] The SPM chaotic mapping function is a method for generating chaotic sequences based on sine and polynomial mappings. Specifically, it can be implemented by superimposing a sine function and a quadratic polynomial function, generating a chaotic sequence through nonlinear iteration. This sequence is used to generate an initial solution with ergodicity and randomness during the initialization phase. The task point iteration completion judgment condition refers to the state determination that all task points in the undirected graph have completed the chaotic mapping calculation. This can be implemented by comparing a loop counter with the total number of task points, controlling the iteration termination condition of the chaotic mapping.

[0081] Specifically, in the population initialization phase of DOA, the initial population obtained suffers from uneven distribution of individual locations, inconsistent quality, and poor diversity. Using such an initial population for iteration may lead to a series of problems in finding quality, such as decreased optimization efficiency and coarsened optimization accuracy. To improve the quality of the search, this embodiment of the invention introduces the SPM chaotic mapping method to process the randomly generated initial solution. Optimization is performed to generate an initial population with high coverage and high quality. The specific SPM chaotic mapping function is shown in formula (18): (18) In the formula, Represents the randomly generated first... The initial location of the stray dog; Indicates the first The position of the wild dog in the iteration after the first iteration; Represents a random number between 0 and 1. When and At this point, the function is in a chaotic state. Experiments show that when... and At that time, the sequence generated by the SPM mapping has better ergodicity and randomness, and its chaotic orbital state value range is (0,1).

[0082] In the initialization process of UAV path planning, an initial position vector is first randomly generated as the input value for chaotic mapping. This initial position is then input into the SPM chaotic mapping function (i.e., in formula (18)) for nonlinear iterative calculation to obtain a position sequence with chaotic characteristics. After each iteration, it is checked whether the number of currently processed task points has reached the total number of task points in the undirected graph. If not, the chaotic mapping calculation continues, generating an initial position set covering different regions of the solution space through multiple iterations. When all task points have completed chaotic mapping processing, the final iterative positions are integrated into the initialization data to provide a highly diverse initial population for subsequent Wild Dog optimization algorithms. For example, such as Figure 5 As shown, a vector express, , The lengths are all , Indicates the number of nodes. A set of positional components. , To map the location of the wild dog to the actual route, a real node is matched. First, an index is assigned to the known node (task point). Its position is determined by coordinates Indicates, such as J 1 (0.4, 1.2) J 2 (1.1, 1.7) J 3 (2.2, 1.9), etc. The location components of wild dogs are... , ,like J 1 (0.1, 0.4) J 2 (0.6, 0.8) J 3 (1.3, 1.3), etc. For each group of positional components of the wild dog. Assign a nearest node, such as Figure 5 The coordinate system is shown. The most recently unassigned node is , The most recently unassigned node is And so on, the resulting lines are as follows: Figure 5 As shown by the connecting lines, the location of the stray dog ​​is mapped to the actual route. Then, the position information of the nodes in the route can be substituted into formula (18) for calculation to obtain the iterative position of each node after initialization.

[0083] In some embodiments of the present invention, the initial data is iteratively optimized based on the Wild Dog optimization algorithm, the adaptive t-distribution mutation strategy, and the energy consumption of multiple task points to obtain path planning results, including: The initial data is optimized based on the wild dog optimization algorithm and the energy consumption of multiple task points to obtain the optimal individual position in the current population; In this process, the fitness value of each wild dog is calculated according to the defined objective function to evaluate their performance in task planning. The fitness values ​​of wild dogs are compared among themselves, and the position Xbest of the current population is recorded as a reference for subsequent position updates. The objective function for calculating fitness is shown in formula (19): (19) In the formula, This represents the set of task points. If the drone starts from a task point... Move to the mission point ,but It is 1 if it is true, otherwise it is 0.

[0084] The location of wild dogs is updated by selecting one of three behaviors: siege, pursuit, or scavenging. The siege behavior causes individual wild dogs to search near the globally optimal location; the pursuit behavior simulates the trajectory of a wild dog hunting prey alone; and the scavenging behavior causes wild dogs to wander randomly in search of carrion. This flexible selection of these three behaviors improves search efficiency and global optimization capabilities.

[0085] (1) Besieging behavior A certain number of wild dogs are randomly selected to search near the global optimal position and obtain a new position. The position update is shown in formula (20): (20) In the formula, Indicates the first During the nth iteration The location of the stray dog This refers to the group of wild dogs that participated in this round of attacks. Indicates the first During the nth iteration The location of the stray dog This indicates the position of the best solution found in the previous iteration. This represents a uniformly generated random number within [-2, 2], and is a scaling factor that can change the size of the stray dog's trajectory. Indicates in Random integers generated in the process, Indicates the size of the population.

[0086] (2) Pursuit behavior The pursuit behavior simulates the trajectory of a wild dog hunting prey alone, and the calculation is shown in formula (21): (twenty one) In the formula, This represents a uniformly generated random number within the interval [-1, 1]. Indicates the randomly selected first Only stray dogs, among them .

[0087] (3) Carrion-eating behavior Wild dogs will roam freely in their habitats in search of carrion, and their corresponding locations will be updated as shown in formula (22): (twenty two) In the formula, This represents a binary number randomly generated by an algorithm. .

[0088] Then, the survival rate of each wild dog is calculated through survival behavior to assess its adaptability in the current environment. When the survival rate of an individual wild dog is less than or equal to 0.3, these individuals with low survival rates are updated using formula (23), introducing new search directions and maintaining population diversity. Formula (23) is shown below: (twenty three) Recalculate the fitness value of each wild dog after the update and evaluate its performance in the new location. Compare the fitness values ​​between individuals again and record the position Xbest of the best individual in the current population to ensure continuous optimization of the algorithm.

[0089] Based on the adaptive t-distribution mutation strategy The distribution function and the optimal individual position are used to perturb the location of the wild dog, and the perturbed target wild dog location is obtained. To prevent population homogenization from causing a decline in search ability, avoid premature convergence, and expand the search space of solutions, an adaptive t-distribution mutation strategy was adopted, which improves the search performance of the algorithm by introducing small-amplitude random changes.

[0090] The calculation in the adaptive t-distribution mutation strategy is shown in Equation (24): (twenty four) In the formula, Indicates adaptive Location of wild dogs after distribution disturbance; For stray dogs in the first The position of the next iteration; This represents an adaptive system with the number of iterations of the population as the degree of freedom parameter. Distribution function.

[0091] To reduce the computation time of the algorithm and leverage its advantages, the use of the adaptive distribution mutation operator is adjusted by selecting the probability of exponential decay. This allows the algorithm to perturb individual positions with a higher probability in the early stages of iteration, thereby mitigating the impact of premature convergence on the algorithm's optimization ability. The exponential decay is shown in formula (25): (25) In the formula, The maximum number of iterations, This represents the current iteration number. This determines the upper limit of the dynamic selection probability, which is 0.5. This determines the magnitude of the change in the dynamic selection probability, with a value of 0.1.

[0092] Determine if the current number of iterations for the population is the maximum number of iterations; If not, the target wild dog location is iteratively optimized according to the wild dog optimization algorithm; If so, the path planning result is obtained based on the optimal individual position output in each iteration.

[0093] The algorithm determines whether the current iteration count of the population has reached a preset termination condition, such as whether the current iteration count has reached the maximum iteration count Tmax. If the termination condition is reached, the algorithm generates the AUV task point access order (i.e., path planning result) based on the final recorded global optimum value Xbest, and outputs this order and its corresponding optimal fitness value, completing the algorithm optimization process; otherwise, it returns to "optimize the initial data based on the Wild Dog optimization algorithm and the energy consumption of multiple task points to obtain the optimal individual position in the current population" to continue optimizing the population.

[0094] In a specific embodiment of this invention, simulation experiments were conducted on the MATLAB simulation platform. Based on a case study with 13 task points, an initial and final point were determined, forming a complete closed-loop task path. The remaining task points represent the target locations where the UAV needs to perform its tasks. To concretely demonstrate the optimal path planning results of this algorithm, a 3km × 3km coordinate grid was designed to map the distribution of these task points, as shown below. Figure 6As shown, it was compared and analyzed with the PSO and IGCPA algorithms. The experimental parameters followed the settings listed in Table 1, with a population size of 100 for all three algorithms and a maximum number of iterations of 600.

[0095] Table 1. Simulation Parameters

[0096] Simulation content and results are as follows Figure 7 As shown, based on simulation data, the simulation results of the CAMWDA algorithm are represented by circular lines, the simulation results of the PSO algorithm by diamond lines, and the simulation results of the IGCPA algorithm by triangle lines. The x-axis represents the number of mission points, and the y-axis represents the total energy consumption during cruise. The total cruise energy consumption of the UAV using the CAMWDA, PSO, and IGCPA algorithms for path planning reached 2.0665. kilowatt-hours, 2.1302 kilowatt-hours, 2.2704 In terms of kilowatt-hours, the CAMWDA algorithm saves approximately 2.9% of energy compared to the PSO algorithm and approximately 8.9% compared to the IGCPA algorithm. This comparison result indicates that the CAMWDA algorithm has better optimization capabilities.

[0097] Corner energy consumption, such as Figure 8 As shown, the simulation results of the CAMWDA algorithm are represented by circular lines, the simulation results of the PSO algorithm by diamond lines, and the simulation results of the IGCPA algorithm by triangle lines. The x-axis represents the number of task points, and the y-axis represents the corner energy consumption. The corner energy consumption of the CAMWDA, PSO, and IGCPA algorithms in the figure is 1.97586. kilowatt-hours, 2.08776 kilowatt-hours, 2.62764 Kilowatt-hours. Compared with the IGCPA algorithm, the energy-saving effect is significant, reaching 5.3% compared with the PSO algorithm. This data proves the effectiveness of the CAMWDA algorithm in reducing energy consumption at UAV corners and improving path planning efficiency. In UAV path planning research, cruising distance is an important indicator for evaluating the effectiveness of path optimization, as it is related to the UAV's energy consumption and mission execution efficiency. As shown in Table 2: Table 2. Cruise distance of optimal path planning

[0098] Table 2 shows that the optimal path planning cruising distance of the CAMWDA algorithm reached 11.61 km, compared to 12.00 km and 12.84 km for the PSO and IGCPA algorithms, respectively. This algorithm reduces the UAV's flight distance. This result demonstrates the effectiveness of the algorithm in improving path planning efficiency and reducing energy consumption, thereby enhancing the overall efficiency of the UAV in performing multiple tasks.

[0099] The cruising times for the optimal path are shown in Table 3: Table 3. Cruise Time for Optimal Path Planning

[0100] In practical applications of UAV path planning, shortening the cruise time not only significantly improves mission efficiency but also plays a crucial role in reducing energy consumption. According to the provided data, the algorithm used in this embodiment of the invention achieves a minimum flight distance of 11.61 kilometers while effectively reducing the cruise time to only 1587 seconds, which is superior to the PSO algorithm's 1636 seconds and the IGCPA algorithm's 1740 seconds. This performance improvement not only reduces the power consumption of the UAV during missions but also means that in urgent or time-constrained missions, this algorithm can complete mission objectives faster.

[0101] To comprehensively evaluate the performance and effectiveness of the proposed algorithm, this embodiment of the invention sets up multiple test cases with different numbers of task points. The number of task points in the experiments was set to 11, 12, 13, 14, and 15, respectively, to examine the algorithm's adaptability and stability to varying task loads. To ensure the randomness and generalizability of the experimental results, the initial positions of the task points were randomly distributed in each experimental setting with each number of task points, and each setting underwent 50 randomization iterations. For each task point setting, each evaluation algorithm was run independently to ensure the reliability and consistency of the results. Then, the results of these runs were averaged to obtain the average performance index of each algorithm. Figure 9 The results of the above experimental setup are shown. The x-axis represents the number of cruise points, and the y-axis represents the total cruise energy consumption. This demonstrates the performance of the CAMWDA algorithm, PSO algorithm, and IGCPA algorithm under different numbers of cruise points.

[0102] To better implement the UAV multi-mission point cruise path planning method in this embodiment of the invention, correspondingly, this embodiment of the invention also provides a UAV multi-mission point cruise path planning device, such as... Figure 10 As shown, the UAV multi-task point cruise path planning device 1000 includes: The data acquisition module 1001 is used to set multiple task points and acquire flight data of the UAV in multiple flight stages; the flight data includes cornering data when the flight stage is a cornering stage; The model building module 1002 is used to convert the UAV's cruise operation at multiple mission points into an undirected graph, and to build a flight energy consumption model based on the undirected graph, flight data and turning angle data. The energy consumption confirmation module 1003 is used to obtain the energy consumption of the UAV at each mission point during the cruise process based on the flight energy consumption model. The iterative optimization module 1004 is used to iteratively optimize the energy consumption of multiple task points based on the Wild Dog optimization algorithm, obtain path planning results, and control the UAV to fly according to the path planning results.

[0103] The UAV multi-mission point cruise path planning device 1000 provided in the above embodiments can realize the technical solutions described in the above UAV multi-mission point cruise path planning method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above UAV multi-mission point cruise path planning method embodiments, and will not be repeated here.

[0104] The above provides a detailed description of the UAV multi-task point cruise path planning method and apparatus provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for multi-task point cruise path planning for unmanned aerial vehicles (UAVs), characterized in that, include: Multiple task points are set to acquire flight data of the UAV at multiple flight stages; the flight data includes cornering data when the flight stage is a cornering stage; The drone's cruise operations at multiple mission points are converted into an undirected graph, and a flight energy consumption model is constructed based on the undirected graph, the flight data, and the turning angle data. The energy consumption of the UAV at each mission point during the cruise is obtained based on the flight energy consumption model. The energy consumption of the multiple task points is iteratively optimized based on the Wild Dog optimization algorithm to obtain the path planning result, and the UAV is controlled to fly according to the path planning result.

2. The UAV multi-task point cruise path planning method according to claim 1, characterized in that, The step of constructing a flight energy consumption model based on the undirected graph, the flight data, and the turning angle data includes: Based on the flight data, the basic forces for each flight phase are obtained; Based on the basic forces, the flight data, and the rotation angle data, the total aerodynamic power required by the UAV during uniform flight is obtained. A flight energy consumption model is constructed based on the undirected graph and the total aerodynamic power.

3. The UAV multi-mission point cruise path planning method according to claim 2, characterized in that, The multiple flight phases also include a takeoff phase, a level flight phase, a descent phase, and a hovering phase; obtaining the total aerodynamic power required by the UAV for uniform flight based on the basic forces and the flight data includes: Based on the fundamental forces and the flight data, the parasitic power, induced power, profile power and climb power corresponding to each flight stage are obtained, and the turning energy consumption is obtained based on the turning angle data. The first power is obtained based on the parasitic power, the induced power, the profile power, and the climb power during the takeoff phase; The second power is obtained based on the parasitic power, the induced power, and the profile power during the horizontal flight phase; The third power is obtained based on the parasitic power, the induced power, the profile power, and the climbing power during the descent phase; The fourth power is obtained based on the parasitic power, the induced power, and the profile power during the hovering phase; The total aerodynamic power of the UAV flying between adjacent mission points is obtained based on the first power, the second power, the third power, the fourth power, the flight data, and the turning energy consumption.

4. The UAV multi-task point cruise path planning method according to claim 1, characterized in that, The iterative optimization of energy consumption at multiple task points based on the Wild Dog optimization algorithm yields path planning results, including: Configure a mapping relationship between the number of each task point and its location; The numbering and position of the multiple task points of the UAV's flight path in the undirected graph are initialized according to the mapping relationship to obtain initialization data; The initial data is iteratively optimized based on the Wild Dog optimization algorithm, the adaptive t-distribution mutation strategy, and the energy consumption of the multiple task points to obtain the path planning result.

5. The UAV multi-mission point cruise path planning method according to claim 4, characterized in that, The initialization data is iteratively optimized based on the Wild Dog optimization algorithm, the adaptive t-distribution mutation strategy, and the energy consumption of the multiple task points to obtain the path planning result, including: The initial data is optimized based on the wild dog optimization algorithm and the energy consumption of the multiple task points to obtain the optimal individual position in the current population; According to the adaptive t Adaptive strategies in distribution mutation The distribution function and the optimal individual position are used to perturb the location of the wild dog to obtain the perturbed target wild dog location; Determine whether the current iteration count of the population is the maximum iteration count; If not, the location of the target wild dog is iteratively optimized according to the wild dog optimization algorithm. If so, the path planning result is obtained based on the optimal individual position output in each iteration.

6. The UAV multi-task point cruise path planning method according to claim 4, characterized in that, The initialization of the numbering and position of the multiple task points of the UAV's flight path in the undirected graph according to the mapping relationship yields initialization data, including: Confirm the randomly generated initial position, and determine the target task point based on the principle of proximity according to the initial position; Substitute the position data of the target task point in the mapping relationship into the SPM chaotic mapping function for calculation to obtain the iterative position of the target task point; Determine whether the task point in the undirected graph has completed iteration; If not, the iteration position is calculated according to the SPM chaotic mapping function to obtain the updated iteration position; If so, the initialization data is obtained based on all iteration positions calculated in the iterative process.

7. The UAV multi-mission point cruise path planning method according to claim 3, characterized in that, The flight data includes hovering time, takeoff altitude, climb angle, UAV flight speed, descent altitude, and Euclidean distance between adjacent mission points; the process of obtaining the total aerodynamic power of the UAV flying between adjacent mission points based on the first power, the second power, the third power, the fourth power, the flight data, and the turning energy consumption includes: The takeoff time is obtained based on the takeoff altitude, the climb angle, and the UAV flight speed. The horizontal flight time is obtained based on the takeoff altitude, the climb angle, the UAV flight speed, the descent altitude, and the Euclidean distance between adjacent mission points; The descent time is obtained based on the descent altitude, the climb angle, and the UAV flight speed; The total aerodynamic power of the UAV flying between adjacent mission points is obtained based on the first power, the second power, the third power, the fourth power, the takeoff time, the horizontal flight time, the descent time, the hovering time, and the turning energy consumption.

8. The UAV multi-mission point cruise path planning method according to claim 6, characterized in that, The SPM chaotic mapping function is as follows: In the formula, Represents the randomly generated first... The initial location of the stray dog; Indicates the first The position of the wild dog in the iteration after the first iteration; Represents a random number between 0 and 1; when and At that time, the function is in a chaotic state.

9. The UAV multi-task point cruise path planning method according to claim 1, characterized in that, The flight energy consumption model is as follows: In the formula, C represents the set of task points. , , Indicates task point arrive The distance is calculated by the task point. arrive It is obtained by the Euclidean distance between them. This represents a route consisting of task point numbers; when the drone departs from the task point... Move to the mission point hour It is 1 if it is true, otherwise it is 0; This indicates that each task point is reached exactly once; This indicates that each task point is left and only left once; Indicates whether edge (i, j) is selected in the optimal solution; This indicates the elimination of sub-loops; This indicates that the drone's battery capacity is greater than the sum of its range energy consumption and cornering energy consumption; This indicates that the distance the drone needs to fly forward before adjusting its flight attitude must be greater than or equal to the minimum distance. .

10. A multi-task point cruise path planning device for unmanned aerial vehicles (UAVs), characterized in that, include: The data acquisition module is used to set multiple task points and acquire flight data of the UAV in multiple flight phases; the flight data includes cornering data when the flight phase is a cornering phase; The model building module is used to convert the UAV's cruise operation at multiple mission points into an undirected graph, and to build a flight energy consumption model based on the undirected graph, the flight data, and the turning angle data. The energy consumption confirmation module is used to obtain the energy consumption of the UAV at each mission point during the cruise process based on the flight energy consumption model. The iterative optimization module is used to iteratively optimize the energy consumption of the multiple task points based on the Wild Dog optimization algorithm, obtain the path planning result, and control the UAV to fly according to the path planning result.

Images