Agricultural unmanned aerial vehicle flight path planning method based on improved Harris eagle algorithm

Abstract

The invention is suitable for the technical field of path planning, and particularly relates to an agricultural unmanned aerial vehicle flight path planning method based on an improved Harris eagle algorithm, and the method comprises the steps: constructing a site environment model, and determining a target function based on the site environment model; initializing parameters; using Chebyshev mapping to initialize a population, initializing individuals, and reserving Hallis eagle individuals with higher fitness to form a final initial population; executing a global exploration strategy or a local development strategy; and continuously updating the current globally optimal solution by comparing the fitness values of all individuals in the whole search space, and outputting the fitness value and position of the optimal individual when a stop condition is met. According to the method, Chebyshev mapping, a non-linear change escape energy factor mechanism and a reverse learning crossover operator disturbance strategy are combined, and a safe, reliable and low-cost unmanned aerial vehicle flight path planning scheme is obtained by constructing a flight path planning model fusing constraint conditions such as an unmanned aerial vehicle flight area and a maximum climbing angle.

Description

Technical Field

[0001] The invention belongs to the technical field of path planning, and in particular relates to a track planning method for an agricultural unmanned aerial vehicle based on an improved Harris Hawk algorithm. Background Art

[0002] The recurrence of pests and diseases will have a catastrophic impact on fruit safety and will directly lead to a significant decrease in fruit quality and yield. In recent years, with the rapid development of emerging technologies such as the Internet of Things and artificial intelligence, agricultural drones have been widely used in the field of agricultural pesticide application. Compared with manual pesticide application, the application of drones to crop pests and diseases has technical advantages. Therefore, it is necessary to plan an efficient flight path in the complex terrain environment of known mountain orchards, taking into account the flight area and maximum climb angle of drones for pesticide application and other operations, so as to improve flight safety and reduce flight costs.

[0003] Many scholars use artificial intelligence algorithms for drone path planning. Among them, metaheuristic algorithms, as a new type of heuristic algorithm, have been widely used because they do not require the use of specific conditions of a certain problem, have strong global optimization capabilities, and have fewer parameter settings. A hybrid metaheuristic algorithm for three-dimensional complex unknown environments is proposed in the prior art, which combines the difference method with the cuckoo algorithm to accelerate the global convergence of the algorithm while maintaining strong robustness. The prior art also proposes an artificial bee colony optimization algorithm based on chaos operators, establishes a drone flight model, and verifies the effectiveness, feasibility, and robustness of the optimization algorithm.

[0004] There are many studies on UAV path planning algorithms, but most of the results focus on improving the algorithm model, and there are two problems: First, after the algorithm performance is improved, the strategy of enhancing the algorithm to jump out of the local optimal value is not ideal. In particular, in the method of using the group puzzle algorithm optimization, the chaotic characteristics of different mapping strategies are different in the population initialization stage, and in the exploration and development ability of the balance algorithm, the linear change factor balance is weak; second, the flight path of the UAV for applying pesticides to mountain fruit trees needs to be planned in combination with the terrain environment of the hilly orchard, the maximum climbing angle and other constraints. Summary of the invention

[0005] The purpose of the present invention is to provide a trajectory planning method for agricultural UAV based on the improved Harris Hawk algorithm, aiming to solve the problem in the prior art that the strategy of enhancing the algorithm to jump out of the local optimal value after the algorithm performance is improved.

[0006] The present invention is implemented as follows: a method for trajectory planning of an agricultural UAV based on an improved Harris Hawk algorithm, the method comprising: Obtain the environmental conditions of the site, build a site environmental model, and determine the objective function based on the site environmental model; The objective function is solved by improving the Harris Eagle algorithm: Initialization parameters, including population size , maximum number of iterations , the dimension of the objective function , initial value upper and lower bounds and ; The population is initialized using Chebyshev mapping, the individuals are initialized, and the Harris hawk individuals with greater fitness are retained to form the final initial population cluster; renew Value, calculate the prey nonlinear escape energy factor ,according to The value of executes a global exploration strategy or a local development strategy; By comparing the fitness values ​​of all individuals in the entire search space, the current global optimal solution is continuously updated, and it is determined whether the maximum number of iterations T has been reached. If the conditions are met, the fitness value and position of the optimal individual are output, otherwise the population is reinitialized.

[0007] Preferably, in the steps of obtaining the environmental conditions of the site, constructing the site environment model, and determining the objective function based on the site environment model, the mathematical model of the site environment model is expressed as: ; In the formula, (x i ,y i ) is the center coordinate of the ith peak; h i is the terrain parameter, controlling the height, x si and si are the attenuation of the ith peak along the x-axis and y-axis directions, respectively, which control the slope, and n is the total number of peaks.

[0008] Preferably, in the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, a terrain constraint is set, which is expressed as: ; In the formula, Z i is the flight altitude of the UAV, Z is the terrain function, Z(x i ,y i ) represents (x i ,y i ) When the flight altitude is less than or equal to the terrain height, the fitness will be amplified at a preset ratio.

[0009] Preferably, in the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, a position constraint is set, which is expressed as: ; In the formula, x max ,y max and z max are the maximum values ​​of each dimension in the solution space, n is a natural number. When the path point exceeds the solution space, the fitness will be amplified.

[0010] Preferably, in the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, a maximum climbing angle constraint γ is set, which is expressed as: ; in, is the maximum climb angle.

[0011] Preferably, the expression of the objective function is: ; ; Where: ; ; ; Among them, c1, c2 and c3 are the control variables of terrain constraint, position constraint and angle constraint θ respectively. max is the maximum constraint angle, and P is the solution space of the optimization problem.

[0012] Preferably, in the step of initializing the population using Chebyshev mapping, initializing the individuals, and retaining Harris hawk individuals with greater fitness to form the final initial population group, the initial position of the Harris hawk population is set by Chebyshev mapping, and the initial position of the population is: ; in, is a randomly selected individual from the current population; and For the current Harris Hawks The iteration and The position vector of the iteration; is the position of the current optimal individual, i.e. the prey; is the average position of individuals in the population; is the average position difference between the prey and the individual; and are the lower and upper bounds of the population respectively; is the population size; Z1, Z2、 Z3 and Z4 are position constraint coefficients.

[0013] Preferably, update Value, calculate the prey nonlinear escape energy factor ,according to In the step of executing the global exploration strategy or the local development strategy, The value calculation formula is expressed as: ; ; Where: is the initial value of the weight at the beginning of the iteration; is the termination value at the end of the weight iteration; t is the current iteration number; T is the maximum iteration number; m is A random number between It is the escape energy factor for the prey; for A random number.

[0014] Preferably, update Value, calculate the prey nonlinear escape energy factor ,according to In the step of executing the global exploration strategy or the local development strategy, when When , the algorithm executes the local development strategy and uses the reverse learning crossover operator perturbation strategy to update the optimal individual position information of the population under soft encirclement and hard encirclement conditions. The expression is: ; Where: Iterate the optimal solution for the population The reverse solution of ; for A random number between .

[0015] Preferably, when When , the soft encirclement strategy is implemented to update the individual positions of Harris Hawks; when When , the algorithm executes the soft encirclement strategy of progressive rapid dive, updates the individual position, and uses the optimal solution under different strategies to perform reverse learning perturbation to update the optimal position of the final population individual; when When , the hard encirclement strategy is executed and the individual position is updated; when When , a progressive rapid dive hard encirclement strategy is implemented to update the individual position, and the optimal solution under different strategies is used for reverse learning perturbation to update the optimal position of the final population individual.

[0016] The agricultural UAV trajectory planning method based on the improved Harris Eagle algorithm provided by the present invention combines Chebyshev mapping, nonlinear escape energy factor mechanism and reverse learning crossover operator perturbation strategy, and obtains a safe, reliable and low-cost UAV trajectory planning scheme by constructing a trajectory planning model that integrates constraints such as the UAV flight area and maximum climb angle. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of an environment model provided by an embodiment of the present invention; Figure 2 A flow chart of a method for agricultural UAV trajectory planning based on an improved Harris Hawk algorithm provided in an embodiment of the present invention; Figure 3 A test function convergence curve diagram provided by an embodiment of the present invention; Figure 4 An aerial topographic map of a field scene experiment provided by an embodiment of the present invention; Figure 5 A front view of a 20m radius area environment model provided by an embodiment of the present invention; Figure 6 A top view of a 20m radius area environment model provided in an embodiment of the present invention; Figure 7 A front view of a 30m radius area environment model provided by an embodiment of the present invention; Figure 8 A top view of a 30m radius area environment model provided in an embodiment of the present invention; Fig. 9 A front view of a 40m radius area environment model provided by an embodiment of the present invention; Fig.10 This is a top view of the 40m radius area environment model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script. like Figure 2FIG. 1 is a flow chart of a method for trajectory planning of an agricultural UAV based on an improved Harris Hawk algorithm provided by an embodiment of the present invention, wherein the method comprises: Obtain the environmental conditions of the site, build a site environmental model, and determine the objective function based on the site environmental model.

[0020] In this step, if Figure 1 As shown, the basis of the trajectory planning of the present invention is a specific field orchard environment model. This study simulates the obstacle area through a higher mountain model and establishes a mathematical model as formula (1): (1); In the formula, (x i ,y i ) is the center coordinate of the ith peak; h i is the terrain parameter, controlling the height; x si and si are the attenuation of the ith peak along the x-axis and y-axis directions, controlling the slope; n is the total number of peaks. The environmental model of this paper is as follows: Figure 1 shown.

[0021] Terrain constraint: To avoid collisions of UAVs during field flight in orchards, this paper sets the flight altitude to be higher than the terrain altitude. Therefore, the model of terrain constraint condition is formula (2): (2); In the formula, Z i is the flight altitude of the UAV, Z is the terrain function, Z(x i ,y i ) represents (x i ,y i ) is located at the terrain height. When the flight altitude is less than or equal to the terrain height, the fitness will be magnified at a certain ratio.

[0022] Position constraint: To prevent the UAV from colliding in an unknown environment, the position constraint model is established as formula (3): (3); In the formula, x max ,y max 、z max are the maximum values ​​in each dimension of the solution space respectively. When the path point exceeds the solution space, the fitness will be amplified.

[0023] Maximum climb angle constraint: The UAV is also subject to the maximum climb angle constraint. If the maximum climb angle is exceeded, the UAV will stall and cause danger. Therefore, the climb angle γ constraint can be expressed as formula (4): (4); In the present invention, when the flight path satisfies each constraint condition, the cost is the flight path length; on the contrary, when certain constraints are not met, the fitness of the position point will be appropriately amplified by the control variables of the corresponding constraint conditions. The specific expressions of the control variables of each constraint condition are formula (5), formula (6), and formula (7): (5); (6); (7); Among them, c1, c2, c3 are the control variables of terrain constraint, position constraint and angle constraint respectively, P is the solution space of the optimization problem, and the objective function expression based on each constraint condition is formula (8): (8); In the prior art, the Harris Hawk Algorithm (HHO) is a new meta-heuristic algorithm, the idea of ​​which is derived from the unique cooperative foraging activities of Harris Hawk populations.

[0024] Exploration phase (Strategy 1): In the global exploration phase, Harris's hawk randomly inhabits the search space and waits for opportunities to observe prey. Harris's hawk is at a certain location according to two equally probable The size of selects the corresponding strategy to explore the prey, and the update formula for the position vector of Harris Hawk in the next iteration is formula (9): (9); (10); Where: is a randomly selected individual from the current population; and For the current Harris Hawks The iteration and The position vector of the iteration; Both A random number; is the position of the current optimal individual, i.e. the prey; is the average position of individuals in the population; is the average position difference between the prey and the individual; and are the lower and upper bounds of the population respectively; For the population size.

[0025] From global exploration to local development: The HHO algorithm uses the escape energy factor of the prey The size of the global exploration phase is transformed into the local development phase. When , the HHO algorithm enters the global exploration phase, otherwise it enters the local development phase. During the prey’s escape process, its escape energy factor The expression of is formula (11): (11); Where: It is the escape energy factor for the prey; for A random number; is the current iteration number; is the maximum number of iterations.

[0026] Local development stage: according to the absolute value of escape energy and random numbers By comparing with 0.5, it can be judged that according to the Harris Hawk's hawk's encirclement method and the escape behavior of the prey, four strategies can be divided into soft encirclement, hard encirclement, soft encirclement with gradual and rapid dive, and hard encirclement with gradual and rapid dive.

[0027] Soft Siege (Strategy 2): When When the prey has enough energy and has a chance to escape, the Harris Hawk will continue to hover around the prey to consume the prey's energy and use the soft encirclement strategy to attack the prey. The position update formula is (12): (12); in: (13); (14); Where: is the difference between the optimal individual and the current individual; for The distance of the jump during the escape; for A random number.

[0028] Hard Encirclement (Strategy 3): When When , the prey has neither enough energy nor the chance to escape, and the Harris Hawk uses a hard encirclement strategy to attack the prey. The position update formula is (15): (15); Soft Encirclement with a Progressive Fast Dive (Strategy 4): When When , the prey has enough energy to escape and has a chance to escape successfully from the encirclement. At this time, the Harris Hawk needs to form a soft encirclement before attacking, and then use a gradual rapid dive strategy to capture the prey. According to formula (16), its position is updated and compared with the fitness value of the current position. If the fitness value is improved, the position is updated using formula (16); otherwise, it means that the encirclement fails, and formula (17) is used to update the position.

[0029] (16); (17); Where: The dimension of the problem; For one dimensional random vector; is the fitness function; is the Levy flight function.

[0030] At this stage, the final round-up strategy is formula (18): (18); Hard Encirclement with Progressive Fast Dive (Strategy 5): When When the prey has a chance to escape, but the energy to escape is insufficient, the Harris Hawk forms a hard encirclement before attacking, and then uses a gradual and rapid dive strategy to capture the prey. The position update formula is as follows: (19); (20); (twenty one); Among them, if the fitness value is updated, according to the average position Update and execute policies ; If the fitness value is not updated, the strategy is adopted , if the strategy fails, return to the original place.

[0031] In the present invention, an improved Harris Hawk algorithm (CSHHO) is adopted. First, Chebyshev mapping is introduced to complete population initialization, which increases the initial population diversity while expanding the global exploration space, thereby improving the convergence speed and optimization range of the algorithm; then the escape energy factor is changed by using the nonlinear logarithmic inertia weight to balance the global exploration and local development capabilities of the algorithm; finally, the ability of the optimal individual in the algorithm to escape from the local optimum is improved through the reverse learning crossover operator perturbation strategy.

[0032] Introducing Chebyshev mapping population initialization: Chaos is a deterministic, random, nonlinear dynamic system, and chaotic sequences have the characteristics of irregularity, randomness, and ergodicity. Since the last century, chaotic mapping has been widely valued in the field of optimization algorithms, and its dynamic random characteristics can enable optimization algorithms to perform more effective global searches in the search area.

[0033] Improved Harris Hawk Algorithm for Agricultural UAV Trajectory Planning: Although the standard HHO algorithm has a good convergence speed, when solving the global optimal problem, the randomly generated initial population may lead to poor population diversity, affecting the global search capability of the algorithm, resulting in slow convergence speed or unsatisfactory optimization accuracy. In order to improve the algorithm search efficiency, chaotic mapping is introduced into the population initialization stage of the algorithm. The initial individuals are evenly distributed in the search area through chaotic mapping, thereby improving the diversity of the initial population. In the selection of the chaotic mapping model, the present invention chooses to introduce the population initialization operation of the Chebyshev mapping. The traditional Chebyshev mapping can be made to enter a chaotic state by controlling relevant parameters. Compared with one-dimensional mappings such as Tent mapping and Logistic mapping, its value range is larger and has good chaotic characteristics. The expression of the chaotic sequence generated by the Chebyshev mapping is formula (22): (twenty two); Where: The order of chaos.

[0034] The initial position of the Harris hawk population is set by Chebyshev mapping, and equation (9) is updated to equation (23): (twenty three);

[0035] Nonlinear escape energy factor mechanism: In the HHO algorithm, whether the exploration behavior is successfully converted into the exploitation behavior depends on the escape energy factor The absolute value of . When the number of iterations is hour, The algorithm linearly decreases from 2 to 1, and executes the global exploration strategy; when the number of iterations is hour, The algorithm linearly decreases from 1 to 0, and executes a local development strategy. However, in nature, the process of Harris's hawk encircling prey cannot be simulated by a single linear decreasing function. This linear escape energy decreasing strategy easily leads to a poor balance between the algorithm's global exploration and local development behaviors. The algorithm is prone to fall into the local optimum in the later stages of iteration, and converges prematurely. Inspired by the inertia weight setting of the PSO algorithm, this paper uses a nonlinear logarithmic inertia weight in the energy decreasing formula to change the iterative process of the escape energy, thereby balancing the algorithm's global exploration and local development capabilities. The expression is updated to formula (24): (twenty four); (25); Where: is the initial value of the weight at the beginning of the iteration; is the termination value at the end of the weight iteration; t is the current iteration number; T is the maximum iteration number; m is A random number between .

[0036] Reverse learning perturbation strategy: In the later iteration of the HHO algorithm, the population position is updated only by relying on the guidance of the best individual in the population, which leads to a greater dependence on the global best individual. When the best individual lacks the ability to jump out of the local optimal solution, the algorithm is prone to fall into the local extreme value space. Therefore, in the CSHHO algorithm, the reverse solution in the solution space is first captured through the reverse learning strategy, and then the crossover operator strategy is introduced to perturb the current optimal solution and the reverse solution, so as to expand the global search space of the algorithm, enhance the diversity of the population, and improve the ability of the algorithm to jump out of the local optimum.

[0037] The purpose of reverse learning is to find the corresponding reverse solution in the search space based on the current solution, and save the better solution after comparison and selection. The mathematical model expression is formula (26): (26); Where: Iterate the optimal solution for the population The reverse solution of ; for A random number between .

[0038] Logical flow: Step 1: Initialize parameters, including population size , maximum number of iterations , the dimension of the objective function , initial value upper and lower bounds and .

[0039] Step 2: Use Chebyshev mapping to initialize the population. Initialize the individuals according to formula (23), and retain the Harris hawk individuals with greater fitness to form the final initial population cluster.

[0040] Step 3: Update The prey nonlinear escape energy factor is calculated according to formula (24): , realizing the stage transition between global exploration and local development.

[0041] Step 4: When When , the algorithm executes the global exploration strategy and updates the individual position according to formula (9).

[0042] Step 5: When When , the algorithm executes the local development strategy. To avoid the algorithm from falling into the local optimal solution, the reverse learning crossover operator perturbation strategy is used to update the optimal individual position information of the population under soft encirclement and hard encirclement conditions: (1) When When , the algorithm executes the soft encirclement strategy and updates the individual positions of the Harris Hawk according to formula (12); When , the algorithm executes the soft encirclement strategy of progressive rapid dive, updates the individual position according to formula (18), and uses formula (26) to perform reverse learning perturbation on the optimal solution under the two strategies to update the final optimal position of the individual in the population.

[0043] (2) When When , the algorithm executes the hard encirclement strategy and updates the individual position according to formula (15); When , the algorithm executes a hard encirclement strategy of progressive rapid dive, updates the individual position according to formula (19), and uses formula (26) to perform reverse learning perturbation on the optimal solution under the two strategies to update the final optimal position of the individual population.

[0044] Step 6: By comparing the fitness values ​​of all individuals in the entire search space, the current global optimal solution is continuously updated, and it is determined whether the maximum number of iterations has been reached. If the conditions are met, go to step 7, otherwise go to step 2.

[0045] Step 7: Output the optimal individual fitness value and position.

[0046] The objective function is solved by improving the Harris Eagle algorithm: Initialization parameters, including population size , maximum number of iterations , the dimension of the objective function , initial value upper and lower bounds and .

[0047] The population was initialized using Chebyshev mapping, the individuals were initialized, and the Harris hawk individuals with greater fitness were retained to form the final initial population cluster.

[0048] renew Value, calculate the prey nonlinear escape energy factor ,according to The value of executes a global exploration strategy or a local development strategy.

[0049] By comparing the fitness values ​​of all individuals in the entire search space, the current global optimal solution is continuously updated, and it is determined whether the maximum number of iterations T has been reached. If the conditions are met, the fitness value and position of the optimal individual are output, otherwise the population is reinitialized.

[0050] The effectiveness of the present invention is illustrated by the following experiments: In order to test the optimization ability of the improved algorithm, this paper selects 5 benchmark test functions to test the algorithm, and sets the EO algorithm, GWO algorithm, PSO algorithm, HHO algorithm and CSHHO algorithm for comparison. The relevant attribute parameters of the function are shown in Table 1. The initial population size of all algorithms is set to 30, and the number of iterations is set to 500.

[0051] Table 1 Benchmark test function table

[0052] The experiment counted the optimal value, average value and variance of each algorithm to measure the optimization ability of each algorithm. The results of CSHHO algorithm and each comparison algorithm are shown in Table 2. The test convergence curve is shown in Figure 3 As shown in the figure, the horizontal axis represents the number of iterations, and the vertical axis represents the optimal fitness value. The comparison of algorithm test results shows that compared with each control algorithm, the CSHHO algorithm has certain advantages. Under test functions F1, F4 and F5, the PSO algorithm is worse than the EO algorithm, GWO algorithm and HHO algorithm, but under test functions F2 and F3, the average and variance values ​​of the PSO algorithm are not as good as the EO algorithm, GWO algorithm and HHO algorithm, which shows that the PSO algorithm has the ability to jump out of the local optimum, but its optimization robustness is not strong. The result values ​​of the CSHHO algorithm are better than the result values ​​of the first four algorithms, indicating that the CSHHO algorithm effectively improves the convergence speed, reduces the probability of falling into the local extreme value, and has a higher global search ability compared with the control algorithm. The test function convergence curve shows that compared with each control algorithm, under the five test functions, with the increase of the number of iterations, the convergence speed of the CSHHO algorithm is shorter. In summary, compared with the other five algorithms, the CSHHO algorithm performs best.

[0053] Table 2 Comparison of algorithm test results

[0054] Field scene experiment and results: Field scene shooting was carried out in the mulberry plantation of Zhanjiang South Subtropical Botanical Garden in Mazhang District, Zhanjiang City, Guangdong Province. First, the drone took high-altitude aerial photos of the overall terrain, and then the experimental area was demarcated. Figure 4 shown.

[0055] Comparison results of algorithms based on obstacles with different radii: In order to increase the flight difficulty of the improved algorithm in this paper in the UAV trajectory planning problem, this paper generates obstacle areas with radii of 20, 30, and 40m in the constructed environment model map based on the field scene map. The larger the radius, the greater the difficulty of path planning. The experiment was carried out in Matlab 2018a, Windows 10, 16GB environment. The flight starting point was set to (45, 45, 0), the target point was set to (200, 200, 80), the population size was set to 70, and the number of iterations was set to 50. The GWO algorithm, PSO algorithm and HHO algorithm were selected for simulation experiment comparison. The comparison results are shown in the figure below. Figure 5 , Figure 6 , Figure 7 , Figure 8 , Fig. 9 and Fig.10 As shown in the figure, on the basis of satisfying the constraints such as the maximum turning angle, in the obstacle environment model with different radii, the CSHHO algorithm can avoid the obstacle area between the starting point and the end point and plan the shortest or optimal path, and can converge to the optimal track.

[0056] In view of the problems of slow convergence speed and easy falling into local extrema of HHO algorithm, this invention proposes an improved CSHHO algorithm through Chebyshev mapping strategy, nonlinear escape energy factor mechanism and reverse learning perturbation strategy, and verifies that CSHHO algorithm has better algorithm performance than the comparison algorithm based on benchmark test function. Aiming at the problem of agricultural UAV trajectory planning in the complex terrain environment of hilly orchard, a safe trajectory planning model for agricultural UAV is constructed based on the constraints of terrain, position and maximum turning angle under the environmental model. The CSHHO algorithm is applied to solve the trajectory planning model, and the trajectory planning comparison experiment of obstacle areas with different radius is designed, which shows that CSHHO algorithm has better planning effect than GWO algorithm, PSO algorithm and HHO algorithm. The experiment verifies the effectiveness of the proposed safe trajectory planning model method for UAV in the complex terrain environment of hilly orchard.

[0057] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for trajectory planning of agricultural UAV based on improved Harris Hawk algorithm, characterized in that: The method comprises: Obtain the environmental conditions of the site, build a site environmental model, and determine the objective function based on the site environmental model; The objective function is solved by improving the Harris Eagle algorithm: Initialization parameters, including population size , maximum number of iterations , the dimension of the objective function , initial value upper and lower bounds and ; The population is initialized using Chebyshev mapping, the individuals are initialized, and the Harris hawk individuals with greater fitness are retained to form the final initial population cluster; renew Value, calculate the prey nonlinear escape energy factor ,according to The value of executes a global exploration strategy or a local development strategy; By comparing the fitness values ​​of all individuals in the entire search space, the current global optimal solution is continuously updated, and it is determined whether the maximum number of iterations T has been reached. If the conditions are met, the fitness value and position of the optimal individual are output, otherwise the population is reinitialized.

2. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: In the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, the mathematical model of the site environmental model is expressed as: ； In the formula, (x i ,y i ) is the center coordinate of the ith peak; h i is the terrain parameter, controlling the height, x si and si are the attenuation of the ith peak along the x-axis and y-axis directions, respectively, which control the slope, and n is the total number of peaks.

3. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: In the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, a terrain constraint is set, which is expressed as: ； In the formula, Z i is the flight altitude of the UAV, Z is the terrain function, Z(x i ,y i ) represents (x i ,y i ) When the flight altitude is less than or equal to the terrain height, the fitness will be amplified at a preset ratio.

4. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: In the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, a position constraint is set, which is expressed as: ； In the formula, x max ,y max and z max are the maximum values ​​of each dimension in the solution space, n is a natural number. When the path point exceeds the solution space, the fitness will be amplified.

5. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: In the steps of obtaining the environmental conditions of the site, constructing the site environmental model, and determining the objective function based on the site environmental model, a maximum climbing angle constraint γ is set, which is expressed as: ； in, is the maximum climb angle.

6. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: The expression of the objective function is: ； Where: ; ； ； Among them, c1, c2 and c3 are the control variables of terrain constraint, position constraint and angle constraint θ respectively. max is the maximum constraint angle, and P is the solution space of the optimization problem.

7. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: In the step of initializing the population using Chebyshev mapping, initializing the individuals, and retaining the Harris hawk individuals with greater fitness to form the final initial population group, the initial position of the Harris hawk population is set by Chebyshev mapping, and the initial position of the population is: ； in, is a randomly selected individual from the current population; and For the current Harris Hawks The iteration and The position vector of the iteration; is the position of the current optimal individual, i.e. the prey; is the average position of individuals in the population; is the average position difference between the prey and the individual; and are the lower and upper bounds of the population respectively; is the population size; Z1, Z 2、 Z3 and Z4 are position constraint coefficients.

8. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: renew Value, calculate the prey nonlinear escape energy factor ,according to In the step of executing the global exploration strategy or the local development strategy, The value calculation formula is expressed as: ； ； Where: is the initial value of the weight at the beginning of the iteration; is the termination value at the end of weight iteration; t is the current iteration number; T is the maximum number of iterations; m is A random number between It is the escape energy factor for the prey; for A random number.

9. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1 is characterized in that: renew Value, calculate the prey nonlinear escape energy factor ,according to In the step of executing the global exploration strategy or the local development strategy, when When , the algorithm executes the local development strategy and uses the reverse learning crossover operator perturbation strategy to update the optimal individual position information of the population under soft encirclement and hard encirclement conditions. The expression is: ； Where: Iterate the optimal solution for the population The reverse solution of ; for A random number between .

10. The agricultural UAV trajectory planning method based on the improved Harris Hawk algorithm according to claim 1, characterized in that: when When , the soft encirclement strategy is implemented to update the individual positions of Harris Hawks; when When , the algorithm executes the soft encirclement strategy of progressive rapid dive, updates the individual position, and uses the optimal solution under different strategies to perform reverse learning perturbation to update the optimal position of the final population individual; when When , the hard encirclement strategy is executed and the individual position is updated; when When , a progressive rapid dive hard encirclement strategy is implemented to update the individual position, and the optimal solution under different strategies is used for reverse learning perturbation to update the optimal position of the final population individual.

Images