Unmanned aerial vehicle path planning method in mountain environment

Abstract

The invention discloses an unmanned aerial vehicle path planning method in a mountain environment, and the method comprises the steps: firstly carrying out the modeling of the mountain environment, including a planning interval, a terrain environment and a visual obstacle threat environment; constraint conditions such as flight length, flight cost and obstacle avoidance in the flight process are met in the flight process of the unmanned aerial vehicle; the optimal path of the unmanned aerial vehicle is searched for with the purpose of solving the minimum value through the target function; improving a dung beetle algorithm, adopting a composite chaotic mapping mode to improve initialization of a random population, and adopting a nonlinear convergence method to adjust a spawning boundary for breeding dung beetles and a foraging boundary of small dung beetles; introducing a gradient prompt drop population adjustment method to dynamically adjust the proportion of each dung beetle variety in the population in the iteration process; and solving the optimal path objective function of the unmanned aerial vehicle based on an improved dung beetle algorithm to obtain a path planning result. The planning path of the unmanned aerial vehicle in the mountain environment is shorter, the flight safety is higher, the flight cost is lower, and the requirement of the unmanned aerial vehicle for completing the mountain flight efficiency is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) path planning, and in particular to a UAV path tracking method in a mountainous environment. Background Art

[0002] In recent years, unmanned aerial vehicles (UAVs), as small autonomous aircraft that do not require direct piloting, can complete battlefield reconnaissance, intelligence gathering and other tasks when necessary. Compared with manned aircraft, the reconnaissance cost is lower and the safety is higher, so it has broad application prospects. In modern warfare, information warfare and intelligence warfare are undoubtedly more important. UAVs, due to their own advantages such as small size, light weight, easy operation and low cost, play an important role in early warning, interference, confrontation, communication and other aspects in small-scale airspace reconnaissance battlefields.

[0003] In the above application scenarios, the UAV is required to select a flight path with high safety, low cost and strong mission feasibility according to the task requirements, the distribution and size of known obstacles, the spatial accessibility of the UAV, the maximum flight distance and other constraints. In the UAV path planning problem in mountainous environments, the flight distance, the minimum flight altitude, that is, the safety of the flight, and the path smoothness are mainly considered. The flight distance is represented by the length of the path. By reducing the path length, on the one hand, the efficiency of the UAV in completing the task can be improved, thereby reducing the probability of being discovered by the detection equipment; on the other hand, it can also reduce energy consumption, thereby reducing costs. The flight safety cost is represented by the minimum flight altitude. In order to ensure the safety of the UAV flight and avoid collisions with static obstacles during the flight path, the UAV is guaranteed to fly in a relatively safe space. The path smoothness cost is determined according to the steep terrain characteristics of the mountainous environment area with large slope fluctuations. In this environment, the flight track of the UAV must meet its own flight angle capability so that the UAV can reach the designated mission location as safely, efficiently and quickly as possible.

[0004] In view of the problem of UAV trajectory planning in complex mountainous environments with various obstacles and constraints, swarm intelligence optimization algorithms are widely used in UAV path planning because of their strong robustness, simple structure and efficient integration with other algorithms. This patent mainly uses the improved dung beetle optimization algorithm to plan the UAV path. At present, many swarm intelligence algorithms have been applied in UAV path planning, among which the more typical intelligent optimization algorithms are particle swarm optimization (PSO), gray wolf algorithm (GWO), hippopotamus algorithm (HO), etc. Such algorithms are easy to implement, have low overhead and fast optimization speed, and in many cases can achieve high-quality path planning solutions for UAVs under many constraints. Xing Na et al. designed a new dynamic search induction operator to solve the problem of long solution time and poor optimal solution accuracy of ant colony algorithm, which can increase the diversity of the initial population in the early stage and improve the convergence speed of the algorithm in the later stage, but the convergence accuracy has not been greatly improved; Niu Qinyu et al. proposed an optimized genetic algorithm that integrates artificial potential field method and simulated annealing algorithm to solve the problem of low solution efficiency and poor convergence state when solving three-dimensional path planning in traditional genetic algorithms, which makes the planned path length lower and the convergence effect outstanding, but when the obstacle and the target point are close, the search efficiency of the algorithm will decrease; the gray wolf optimization algorithm is a new type of swarm intelligence bionic algorithm, which is generated based on the simulation of the gray wolf population hierarchy mechanism and hunting behavior in nature. It has simple principles, parallelism, low parameter adjustment requirements, and does not need to obtain the gradient information of the problem, and has strong global search capabilities. However, almost all swarm intelligence algorithms have a common problem. When solving complex engineering optimization problems, it is difficult for the algorithm to find a suitable balance between local optimality and global search. Too high will lead to poor solution-seeking ability of the algorithm, and too low will lead to premature limitation to local optimal problems. Therefore, although the above algorithms have certain advantages in solving the UAV path planning problem, the feasible path planning schemes facing complex obstacle avoidance are inefficient and are prone to fall into local optimal loops too early, thus failing to obtain effective feasible flight path schemes. Based on this, it is of great value to study how to quickly obtain an effective optimal path for UAVs in a three-dimensional flight space in a complex environment such as a mountain, and to reduce the flight cost of UAVs. Summary of the invention

[0005] Purpose of the invention: The present invention provides a UAV path tracking method in a mountainous environment to solve the optimal problem of improving UAV path planning in a mountainous environment.

[0006] Technical solution: The method for tracking the path of a UAV in a mountainous environment described in the present invention comprises the following steps:

[0007] (1) Construct an environmental model, including a mountain model and a static obstacle model, to simulate the complex mountain environment and three-dimensional obstacles in the real environment that pose a threat to the UAV;

[0008] (2) Construct multi-dimensional resource constraints for the UAV platform, including constraints on the length of the UAV flight path, the safety of the flight space, and the smoothness of the path;

[0009] (3) Under the condition of multi-dimensional resource constraints of UAVs, the optimal path objective function of the UAV is constructed with the minimum flight cost as the goal; the objective function is composed of three weighted parts: path length cost function, path smoothness cost function, and path flight safety cost function, as well as the introduction of collision cost penalty function;

[0010] (4) Improve the dung beetle algorithm: use a composite chaotic mapping method to improve the initialization of the random population, use a nonlinear convergence method to adjust the egg-laying boundary of the breeding dung beetles and the foraging boundary of the small dung beetles; introduce a gradient descent population adjustment method to dynamically adjust the proportion of each dung beetle species in the population during the iteration process; solve the optimal path objective function of the UAV based on the improved dung beetle algorithm and obtain the path planning result;

[0011] (5) The flight trajectory is processed by cubic spline interpolation smoothing method. The flight path includes the starting point p of the flight. s (x s ,y s , z s ), end point p z (x z ,y z , z z ) and passing through point p o (x o ,y o , z o ), and obtain the UAV trajectory planning, where p o (x o ,y o , z o ) is the optimal solution The corresponding three-dimensional coordinate value.

[0012] Furthermore, the mountain model in step (1) is:

[0013] Read terrain data and generate a mountain model. The minimum flight altitude of the drone is ≥1km and the maximum flight altitude is ≤3km. The mountain environment model is simulated by scatter point fitting and two-dimensional interpolation. The formula is as follows:

[0014]

[0015] Among them, F ij (x i,y j )=z ij is the value in the z direction given in the data table; coefficient a ijkl Solve the following system of equations:

[0016] F ij (x i ,y j )=z ij and F ij (x i+1 ,y j )=z (i+1)j Get the value at the endpoint in the x direction;

[0017] F ij (x i ,y j+1 )=z i(j+1) and F ij (x i+1 ,y j+1 )=z (i+1)(j+1) Get the value at the endpoint in the y direction;

[0018] In summary, the interpolation coordinates in the z direction are obtained, thereby obtaining a mountain model.

[0019] Furthermore, the static obstacle model in step (1) is:

[0020] Use n×3 matrix to simulate the threat environment of static obstacles to drones. The formula is as follows:

[0021]

[0022] In the formula, a i , b i Respectively represent the coordinate information of the i-th static obstacle in space, R i Represents the radius of the i-th static obstacle.

[0023] 4. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, wherein the unmanned aerial vehicle platform constraints in step (2) are as follows:

[0024]

[0025] Among them, l o represents the length of the oth track, L represents the flight distance of the UAV, and the total number of n navigation points constitutes the UAV flight path; γ o , is the deflection angle and pitch angle of the UAV during flight; (x o ,y o ,z o) is the coordinate of the current position of the UAV; d represents the vertical distance between the center of the static obstacle and the track formed by the two nearest adjacent nodes; R represents the obstacle radius, and M represents the UAV flight safety threshold.

[0026] Furthermore, the optimal path objective function of the UAV in step (3) is:

[0027] F C =ω1*f L +ω2*f E +ω3*f S +ω4f V

[0028] In the formula, ω1, ω2, ω3, and ω4 represent f L 、f E 、f S 、f V The weight ratio coefficient, and ω1+ω2+ω3+ω4=1;

[0029] Track length cost function:

[0030]

[0031] In the formula, l o is the length of the corresponding track segment o;

[0032] Path smoothing cost function:

[0033]

[0034] Among them, φ o , It is the deflection angle and pitch angle of the UAV during flight;

[0035] Flight safety cost function:

[0036]

[0037] Among them, f o Represents the safety cost function between two adjacent navigation points, d o represents the straight-line distance between this track and the center of the nearest obstacle, M is the set safety threshold constant, and R i Represents the radius of the static obstacle;

[0038] Collision penalty cost function:

[0039] f V =V×P f

[0040] Among them, the constant V∈(0,1) represents the penalty value; P fCharacterizes the penalty coefficient.

[0041] Furthermore, the implementation process of improving the initialization of the random population by using the composite chaotic mapping method in step (4) is as follows:

[0042] Logistic-Sine-Cosine mapping is used to improve the coordinate distribution of the random x-axis and y-axis; Tent mapping is used to improve the coordinate distribution of the z-axis:

[0043] X LSC =cos(π*4rx*(1-x)+(1π-r)*sin(πx)-0.5))

[0044] Y LSC =cos(π*4rx*(1-x)+(1-r)*sin(πx)-0.5)), r∈(0,1)

[0045] Tent Mapping:

[0046]

[0047] Furthermore, the implementation process of using the nonlinear convergence method in step (4) to adjust the egg-laying boundary of the breeding dung beetles and the foraging boundary of the young dung beetles is as follows:

[0048] According to the planned UAV path objective function, the objective function value corresponding to each dung beetle in the initial population is calculated, and the optimal value, i.e. the minimum value, is recorded as Pop * , and record the corresponding three-dimensional coordinates at this time

[0049] Let the initial iteration number t = 1, and set the size of the random number r1;

[0050] Determine the size of r1. If r1 < 0.9, the dung beetle updates its position according to the rolling ball search behavior and sets a natural random number α1, which is a random value between 0 and 1. If α1 > 0.1, set α = 1, otherwise set α = -1.

[0051] Generate a random number r2 between 0 and 0.2, and a random number r3 between 0 and 1;

[0052] The search is performed using the following search equation:

[0053]

[0054] in, represents the position of the i-th rolling dung beetle at the t+1th iteration; represents the position of the i-th rolling dung beetle at the t-th iteration; represents the position of the i-th ball-rolling dung beetle at the t-1th iteration; Δx is used to simulate natural factors such as light intensity; represents the global worst position;

[0055] If r1>0.9, the ball beetle determines the search strategy by simulating dancing, setting the random number r4 to a random integer between 0 and 180. If r4=0||90||180, the position search is performed according to the above search equation, otherwise the search is performed according to the following search equation:

[0056]

[0057] Among them, tan(aa) is the position offset of the rolling dung beetle when it encounters an obstacle;

[0058] The breeding dung beetles update their positions by breeding behavior; the boundaries of the breeding area of ​​the breeding dung beetles are set in the following way:

[0059]

[0060] Among them, Lb * represents the lower boundary of the breeding area, Ub * represents the upper boundary of the breeding area, is the position information corresponding to the current local optimal solution, f1 is the boundary convergence factor of the breeding area, and the nonlinear convergence boundary method is used. k is Lb * and Ub * Any integer between ;

[0061] The breeding dung beetles have arrived at the production area to hatch their eggs. The location has been updated as follows:

[0062]

[0063] in, represents the position of the i-th hatched egg at the t+1th iteration, represents the position of the i-th hatched egg at the t-th iteration, and are mutually independent random vectors of 1×3 dimensions;

[0064] After the eggs are hatched, the young dung beetles are produced and use foraging behavior to update their positions. The optimal foraging area is set in the following way:

[0065]

[0066] In the formula, Ub ** and Lb ** Represent the upper and lower boundaries of the optimal foraging area, respectively, and f2 represents the boundary convergence factor of the foraging area. The nonlinear convergence boundary method is also used.

[0067] The dung beetle reaches the foraging area to forage and updates its position based on the foraging behavior:

[0068] Set the random number r5 to follow the normal distribution; generate a random vector ranging from 0 to 1

[0069]

[0070] in, is the position of the i-th dung beetle at the t+1th iteration, is the position of the i-th dung beetle at the t-th iteration; another part of the dung beetles will steal and update their positions by stealing;

[0071] Randomly generate a random vector of size 1×3 that follows a normal distribution Generate a fixed constant r8;

[0072] The stealing dung beetle steals and updates its location by stealing:

[0073]

[0074] Calculate the objective function value of each dung beetle in the newly generated population Pop(i+1), and record the minimum value as Pop fit , compared with Pop fit and size, and Update to the smaller value between the two.

[0075] Furthermore, the implementation process of introducing the gradient rapid drop population adjustment method in step (4) to dynamically adjust the proportion of each dung beetle species in the population during the iteration process is as follows:

[0076] After each iteration, the best individual in the population is selected to replace the best individual in the current population; after every quarter of the total number of iterations, the proportion of rolling dung beetles is reduced, and the proportions of breeding dung beetles and small dung beetles are increased;

[0077]

[0078] In the formula, r Percent When the population is initialized, the proportion of the population occupied by the rolling ball dung beetle, v1, v2, v3 are weight constants related to the number of iterations; when the number of iterations t is equal to MAX it Will As the optimal solution.

[0079] The beneficial effects are as follows: compared with the prior art, the beneficial effects of the present invention are as follows: the present invention takes the flight path length, the flight path smoothness angle, and the flight altitude, i.e., flight safety, as performance indicators, and has established a three-dimensional improved optimization model under the constraints of static obstacles in a mountainous environment; by constructing a weight cost function and setting the weight coefficient, an optimization model of a single performance indicator is obtained; in order to be able to find the optimal solution in such a three-dimensional problem model with a high degree of application, based on the low search efficiency and easy falling into local optimal solutions of the swarm intelligence optimization algorithm that has been continuously proposed in recent years in solving three-dimensional and high-dimensional problems, an improved dung beetle algorithm with better balance, better convergence and better solution-finding ability is proposed; the optimal problem of improving the path planning of unmanned aerial vehicles in mountainous environments is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 is a flow chart of the present invention;

[0081] Figure 2 The present invention provides a method flow for the UAV path planning based on the improved dung beetle search algorithm;

[0082] Figure 3 An environmental model is proposed for the implementation of the present invention;

[0083] Figure 4 is the convergence curve f1 of the convergence factor of the breeding area;

[0084] Figure 5 is the convergence curve f2 of the foraging area convergence factor;

[0085] Figure 6 The simulation results of the present invention are shown in Figure 1, where (a) is a three-dimensional path diagram of the UAV; (b) is a top view of the UAV path; and (c) is a convergence curve diagram of the path function. DETAILED DESCRIPTION

[0086] The present invention is further described in detail below with reference to the accompanying drawings.

[0087] like Figure 1 The present invention proposes a method for tracking the path of a UAV in a mountainous environment, which solves and improves the optimal problem of UAV path planning in a mountainous environment. The specific implementation process is as follows:

[0088] Step 1: Construct an environmental model, including a mountain model and a static obstacle model, to simulate the obstacle threat posed to the UAV by complex mountain environments and three-dimensional obstacles in real environments.

[0089] Read terrain data and generate a mountain model. The minimum flight altitude of the drone is ≥1km and the maximum flight altitude is ≤3km. The mountain environment model is simulated by scatter point fitting and two-dimensional interpolation. The formula is as follows:

[0090]

[0091] Among them, F ij (x i ,y j )=z ij is the value in the z direction given in the data table. Coefficient a ijkl Solve the following system of equations:

[0092] F ij (x i ,y j )=z ij and F ij (x i+1 ,y j )=z (i+1)j Get the value at the endpoint in the x direction; F ij (x i ,y j+1 )=z i(j+1) and F ij (x i+1 ,y j+1 )=z (i+1)(j+1) The value at the endpoint in the y direction is obtained; that is, the interpolated coordinates in the z direction are obtained, thereby obtaining a mountain model.

[0093] Use n×3 matrix to simulate the threat environment of static obstacles to drones. The formula is as follows:

[0094]

[0095] In the formula, a i , b i Respectively represent the coordinate information of the i-th static obstacle in space, R i Represents the radius of the i-th static obstacle.

[0096] Step 2: Construct multi-dimensional resource constraints for the UAV platform, including constraints on the length of the UAV flight path, the safety of the flight space, and the smoothness of the path.

[0097] The drone itself has high maneuverability, so the position of the drone in space is constantly changing, and the flight altitude of the drone is not unlimited, so the maximum flight altitude of the drone is set to H. C , the flyable space C is expressed as: (x,y,z)∈{(x,y)|X min ≤x≤X max ,Y min ≤y≤Y max ,H min ≤z≤H C};X min , X max, Y min , Y max is the flight space range of the UAV, H min , H D are the minimum and maximum safe flight altitudes of the drone, and the total length of UVA’s flight path is Due to the limitation of its own weight and the size of the energy it carries, the distance of L is not infinite, L<L max , where L max It is the maximum length of a drone flying in a straight line when unloaded. Among them, (x o ,y o , z o ) is the current position coordinate of the drone; φ o Indicates the deflection angle of the drone during flight. It indicates the pitch angle of the drone during flight. The drone needs to meet its maximum climbing angle during flight, especially in a mountainous geographical environment with large slopes and interlaced peaks. If the pitch angle of the drone during flight is too large, it may exceed its maximum ascent speed, causing the drone to lose control and collide with obstacles in the mountainous environment. If the deflection angle is too large, the drone will have excessive unilateral force and load on one side, causing imbalance at both ends of the drone, which will cause great safety hazards and risks to the flight control module. Therefore, the pitch angle and deflection angle should be lower than the maximum value.

[0098] Step 3: Under the condition of multiple constraints of the UAV, taking the shortest flight time as the goal, the optimal path objective function of the UAV is constructed in combination with the path length of the UAV:

[0099] F C =ω1*f L +ω2*f E +ω3*f S +ω4f V

[0100] The objective function is composed of the flight path length cost function f L , flight path smoothness cost function f E , flight safety cost function f S And the collision cost penalty function f V Composition, ω1, ω2, ω3, ω4 represent the power coefficients of each part. In the present invention, different tasks can use different power coefficients, but it is necessary to ensure that the sum of the power coefficients is 1. In addition, it is worth noting that there can be multiple objective functions set for different UAV path planning problems.

[0101] In the process of UAV track planning, the shorter the track length, the shorter the flight time and the less energy required. The flight path length cost of the entire track segment is the sum of the path length costs of each track sub-segment. To calculate the length of the o-th track segment, it is necessary to calculate the Euclidean distance along the o-th track segment. In order to reduce the amount of calculation, in the present invention, n navigation points are set to calculate the path length cost on the entire track segment. The specific calculation formula of the flight path length cost function is as follows:

[0102]

[0103] In the formula, l o is the length of the corresponding track segment o.

[0104] The smoothness of the flight path is very important for the stability of the UAV's own flight, especially in environments with undulating terrain and steep terrain such as mountains. Generally speaking, in order to ensure the stability of the UAV when performing tasks, it is necessary to maintain a relatively appropriate flight angle. The turning angle of the UAV during flight will disturb its own stability and maneuverability. The instantaneous flight angle in the UAV trajectory planning includes the pitch angle in the vertical direction. and horizontal yaw angle φ o , and these two parameters meet the angle constraint during use. The path smoothness cost function can be described by the following formula:

[0105]

[0106] When flying in a mountainous environment, the drone also needs to avoid collisions with obstacles such as trees and rocks. Therefore, the flight safety cost function is constructed based on the distance d from the center point of each track segment to the obstacle coordinate center, the obstacle radius R and the safety threshold M. Therefore, the flight safety cost function is shown as follows:

[0107]

[0108] Among them, f o Represents the safety cost function between two adjacent navigation points, d o represents the straight-line distance between this track and the center of the nearest obstacle, M is the set safety threshold constant, and R i Represents the radius of static obstacles.

[0109] When the drone is about to encounter an obstacle during flight in a complex mountain environment, the drone will automatically choose to deviate from the straight line to avoid collision with trees and rocks. Set the violation degree V and the penalty coefficient P f .

[0110] Step 4: Use the improved dung beetle optimization search algorithm to solve the optimal path objective function of the UAV, use the optimal path objective function of the UAV as the fitness function of the dung beetle optimization search algorithm, and obtain the corresponding path planning result through iterative optimization; Figure 2 As shown, the following steps are included:

[0111] The dung beetle algorithm is inspired by the physiological behavior of dung beetles in nature. Dung beetles have the behaviors of rolling balls, dancing, foraging, stealing, and reproduction. First, the rolling ball dung beetles navigate according to the information of celestial bodies (the sun and the moon), and roll the dung ball quickly in a straight line. If the light is dim or there is no light at all, they will choose the direction of travel through dancing, so as to quickly reach the appropriate area to build nests and lay eggs. The above behavior reflects the first inspiration of the algorithm we talked about. Several dung beetles are randomly distributed in the entire search space. Each dung beetle will explore the search space in constant motion. After finding a suitable area, the breeding dung beetles will lay eggs in the breeding area. After the hatched eggs mature, small dung beetles will be produced. The small dung beetles will look for suitable foraging areas in the surrounding environment to forage. In addition, some dung beetles will steal in order to get more food in the competition.

[0112] Initialize the population size N of the dung beetle population Pop 、Rolling ball dung beetle Num , breeding dung beetles Num 、Small Dung Beetle x Num , stealing dung beetles Num , number of iterations t, maximum number of iterations MAX it .

[0113]

[0114] Among them, POP(i) represents the position of the initial population, They represent the coordinate position of the i-th (rolling, breeding, small, stealing) dung beetle in three-dimensional space, represents the position of the i-th dung beetle at the t-th iteration, They represent the upper and lower boundaries of the 3D flyable space respectively. is a 3D random vector between 0 and 1.

[0115] The composite chaotic mapping is used to improve the initial population distribution, and the tent mapping and Logistic-Sine-Cosine mapping are used to improve the randomly generated initial population position, so that the 3D coordinates are generated uniformly in space.

[0116] Logistic-Sine-Cosine Mapping:

[0117] X LSC=cos(π*4rx*(1-x)+(1-r)*sin(πx)-0.5)),

[0118] Y LSC =cos(π*4rx*(1-x)+(1-r)*sin(πx)-0.5)), r∈(0,1)

[0119] Tent mapping:

[0120]

[0121] The problem of uneven distribution of the initial population can be effectively improved by using a composite chaotic mapping. Compared with a single logistic chaotic mapping, this composite chaotic mapping has better traversability, randomness and distribution when solving 3D and higher dimensional problems.

[0122] According to the planned UAV path objective function, the objective function value corresponding to each dung beetle in the initial population is calculated, and the optimal value, i.e. the minimum value, is recorded as Pop * , and record the corresponding three-dimensional coordinates at this time

[0123] Let the initial iteration number t=1 and set the size of the random number r1.

[0124] Determine the size of r1. If r1 is less than 0.9, the rolling dung beetle updates its position through rolling behavior and sets a random number α1, which is a random value between 0 and 1. If α1 is greater than 0.1, set the natural coefficient α=1, otherwise set α=-1. When α=1, it indicates that the rolling dung beetle has not deviated from its direction, and when α=-1, it indicates that the dung beetle's search direction has shifted.

[0125] The search behavior is performed randomly in the entire set search space to find the best breeding area. This is called the search or exploration phase. The algorithm can explore the surrounding area suitable for egg laying with different step sizes and update the position of the rolling dung beetle by simulating light intensity. The specific steps of the search behavior are as follows:

[0126] Generate a random number r2 between 0 and 0.2, and generate a random number r3 between 0 and 1.

[0127] The search is performed using the following search equation:

[0128]

[0129] in, represents the position of the i-th rolling dung beetle at the t+1th iteration; represents the position of the i-th rolling dung beetle at the t-th iteration; represents the position of the i-th ball-rolling dung beetle at the t-1th iteration; Δx is used to simulate natural factors such as light intensity; represents the global worst position.

[0130] In addition to being affected by light, dung beetles are also affected by wind direction and uneven terrain, which can cause the direction of their advance to deviate. Therefore, the probability method is used to set the value of the natural coefficient α. Similarly, the higher the value of Δx, the weaker the light. r2 is used to characterize the deflection coefficient, and r3 represents a constant. The closer to the global worst position, the weaker the light, which prompts the dung beetle to move away from that position. By setting Δx, the dung beetle can perform a global search of the entire space as much as possible in the early stage of the algorithm. On the other hand, it can also enhance the search performance and reduce the probability of entering the local optimum, which helps the dung beetle search algorithm to improve the efficiency of drone path planning.

[0131] If r1>0.9, it means that the ball beetle encountered an obstacle in the process of moving forward, so it can re-search the direction of movement through dancing behavior. This behavior is to avoid static obstacles that appear in the process of drone path planning and prevent the algorithm from falling into a local optimal solution.

[0132] Set the random number r4 to a random integer between 0 and 180. If r4 = 0||90||180, perform the position search according to the above search equation. Otherwise, perform the search according to the following search equation:

[0133]

[0134] Among them, tan(aa) is the position offset of the rolling dung beetle when it encounters an obstacle, representing the new rolling direction.

[0135] In nature, dung beetles push their dung balls to a safe and suitable location and hide them there, providing a safe environment for their unhatched offspring. After finding a suitable egg-laying area, the breeding dung beetles begin laying eggs, and all hatched eggs are laid in a restricted area.

[0136] Set the boundaries of the breeding area for dung beetles by:

[0137]

[0138] Among them, Lb * represents the lower boundary of the breeding area, Ub * represents the upper boundary of the breeding area, is the position information corresponding to the current local optimal solution, f1 is the boundary convergence condition of the breeding area, and the convergence curve is as follows: Figure 4 shown.

[0139] The breeding dung beetles have arrived at the production area to hatch their eggs. The location has been updated as follows:

[0140]

[0141] in, represents the position of the i-th hatched egg at the t+1th iteration, represents the position of the i-th hatched egg at the t-th iteration, and are 1×3 dimensional independent random vectors respectively.

[0142] In the dung beetle algorithm discussed in the present invention, each breeding dung beetle only produces one hatching egg, corresponding to one small dung beetle. After the hatching egg is hatched, a small dung beetle is produced. The small dung beetle starts from the hatching area and goes to a suitable foraging area for foraging. Its foraging area is also limited to a certain range. This behavior is intended to enhance the ability of local optimization and improve the convergence of the algorithm. The specific steps of the foraging behavior are as follows:

[0143] The optimal foraging area is set by:

[0144]

[0145] In the formula, Ub ** and Lb ** Represent the upper and lower boundaries of the optimal foraging area, respectively, f2 represents the boundary convergence condition of the foraging area, and the convergence curve is shown in Figure 5 shown.

[0146] The dung beetle reaches the foraging area to forage and updates its position based on the foraging behavior:

[0147] Set the random number r5 to follow the normal distribution; generate a random vector ranging from 0 to 1

[0148]

[0149] in, is the position of the i-th dung beetle at the t+1th iteration, is the position of the i-th dung beetle at the t+1th iteration.

[0150] Some other dung beetles will steal dung balls from other dung beetles to compete for the best food source; the purpose of this behavior is to jump out of the local optimal solution and avoid the algorithm from falling into the local optimal defect. The specific steps of the stealing behavior are as follows:

[0151] Randomly generate a random vector of size 1×3 that follows a normal distribution Generates a fixed constant r8.

[0152] The stealing dung beetle steals and updates its location by stealing:

[0153]

[0154] Common population communication mechanisms include: 1) Select the best individual in the population after iteration, record the initial global worst position (i.e. the individual with the largest fitness value) to replace the best individual in the historical population; however, if this strategy selects the local best individual, it may cause all individuals in the population to fall into the local optimal solution. 2) After a certain number of iterations, adjust the proportion of various types of dung beetle individuals in the population to optimize the global search and local search capabilities of the algorithm; this strategy lacks instability, if the individuals in the sub-population are all poor individuals, the population quality will be low. 3) Collision penalty mechanism, in order to avoid the UAV from colliding with static obstacles such as trees and stones in the mountainous environment during flight, a penalty function is set. When the UAV is within the safety threshold, it will consider the obstacle avoidance selection of the UAV and improve safety by detouring. Although this strategy will improve the path length function of the UAV, it can reduce the flight safety cost function, and on the other hand, it can also ensure the safety of the UAV. 4) Boundary convergence strategy. Restrictions are set on the breeding and foraging areas of the breeding dung beetles and small dung beetles in the population. However, the linear convergence of the traditional dung beetle algorithm may lead to incomplete global search in the early stage and too fast convergence in the later stage, which may lead to the problem of falling into the local optimum. In order to better balance the global search ability and local optimization ability of the algorithm, the boundary convergence is changed to a nonlinear convergence strategy. This strategy can improve the balance between global search and local search, that is, it can strengthen the global search ability in the early stage, and increase the convergence speed in the later stage, reducing the risk of falling into the local optimum. In addition, by adjusting the proportion of various types of dung beetles in the iterative process, the purpose of optimizing the convergence performance of the algorithm is achieved, and the gradient descent method is proposed.

[0155] The nonlinear boundary convergence strategy and gradient descent method are as follows:

[0156] Breeding region boundary convergence strategy:

[0157]

[0158] Where k is Lb * and Ub * Any integer between .

[0159] Foraging area boundary convergence strategy:

[0160]

[0161] Gradient rapid descent method:

[0162]

[0163] Among them, rPercent When the population is initialized, the proportion of the rolling dung beetle in the population, v1, v2, v3 are weight constants related to the number of iterations.

[0164] Calculate the objective function value of each dung beetle in the newly generated population POP(i+1), and record the minimum value as POP fit Compare POP fit and and will Update to the smaller value between the two.

[0165] Let t = t + 1, and determine whether the number of iterations t is equal to t max , if not equal, repeat the above steps of setting the size of the random number r1 and the subsequent steps. In particular, when t satisfies one-quarter, two-quarters, and three-quarters of the total number of iterations, perform the gradient descent method. If; equal, As the optimal solution.

[0166] Step 5: The flight trajectory is processed by cubic spline interpolation smoothing method. The flight path includes the starting point p of the flight. s (x s ,y s , z s ), end point p z (x z ,y z , z z ) and passing through point p o (x o ,y o , z o ), and obtain the UAV trajectory planning. o (x o ,y o , z o ) is the optimal solution The corresponding three-dimensional coordinate value trajectory is smoothed using the cubic B-spline curve equation in which the basis function is:

[0167]

[0168] in Representing factorial, it simplifies to:

[0169]

[0170] Establish as Figure 3 The environment model shown contains six obstacle models. Figure 4 , Figure 5is the improved boundary convergence curve; compared with the convergence curve of the original algorithm (linear straight line), the modified convergence curve f1 can better perform global search in the early stage of the algorithm and better avoid falling into local optimality in the later stage; the f2 curve improves the local optimization ability of the foraging dung beetle in the later stage of the algorithm, which can make the found path function value smaller. Figure 6 The final simulation results are shown in Figure 1, where (a) is the three-dimensional path diagram of the UAV; (b) is the top view of the UAV path; and (c) is the convergence curve of the path function. It can be seen that the path curve obtained by the improved algorithm has good smoothness, fast convergence speed, and high convergence accuracy.

[0171] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for tracking the path of an unmanned aerial vehicle in a mountainous environment, characterized in that: The following steps are involved: (1) Construct an environmental model, including a mountain model and a static obstacle model, to simulate the complex mountain environment and three-dimensional obstacles in the real environment that pose a threat to the UAV; (2) Construct multi-dimensional resource constraints for the UAV platform, including constraints on the length of the UAV flight path, the safety of the flight space, and the smoothness of the path; (3) Under the condition of multi-dimensional resource constraints of UAVs, the optimal path objective function of the UAV is constructed with the minimum flight cost as the goal; the objective function is composed of three weighted parts: path length cost function, path smoothness cost function, and path flight safety cost function, as well as the introduction of collision cost penalty function; (4) Improve the dung beetle algorithm: use a composite chaotic mapping method to improve the initialization of the random population, use a nonlinear convergence method to adjust the egg-laying boundary of the breeding dung beetles and the foraging boundary of the small dung beetles; introduce a gradient descent population adjustment method to dynamically adjust the proportion of each dung beetle species in the population during the iteration process; solve the optimal path objective function of the UAV based on the improved dung beetle algorithm and obtain the path planning result; (5) The flight trajectory is processed by cubic spline interpolation smoothing method. The flight path includes the starting point p of the flight. s (x s ,y s , z s ), end point p z (x z ,y z , z z ) and passing through point p o (x o ,y o , z o ), and obtain the UAV trajectory planning, where p o (x o ,y o , z o ) is the optimal solution The corresponding three-dimensional coordinate value.

2. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The mountain model in step (1) is: Read terrain data and generate a mountain model. The minimum flight altitude of the drone is ≥1km and the maximum flight altitude is ≤3km. The mountain environment model is simulated by scatter point fitting and two-dimensional interpolation. The formula is as follows: Among them, F ij (x i ,y j )=z ij is the value in the z direction given in the data table; coefficient a ijkl Solve the following system of equations: F ij (x i ,y j )=z ij and F ij (x i+1 ,y j )=z (i+1)j Get the value at the endpoint in the x direction; F ij (x i ,y j+1 )=z i(j+1) and F ij (x i+1 ,y j+1 )=z (i+1)(j+1) Get the value at the endpoint in the y direction; In summary, the interpolation coordinates in the z direction are obtained, thereby obtaining a mountain model.

3. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The static obstacle model in step (1) is: Use n×3 matrix to simulate the threat environment of static obstacles to drones. The formula is as follows: In the formula, a i , b i Represents the coordinate information of the i-th static obstacle in space, R i Represents the radius of the i-th static obstacle.

4. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The UAV platform constraints in step (2) are as follows: Among them, l o represents the length of the oth track, L represents the flight distance of the UAV, and the total number of n navigation points constitutes the UAV flight path; γ o , is the deflection angle and pitch angle of the UAV during flight; (x o ,y o ,z o ) is the coordinate of the current position of the UAV; d represents the vertical distance between the center of the static obstacle and the track formed by the two nearest adjacent nodes; R represents the obstacle radius, and M represents the UAV flight safety threshold.

5. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The optimal path objective function of the UAV in step (3) is: F C =ω1*f L +ω2*f E +ω3*f S +ω4f V In the formula, ω1, ω2, ω3, and ω4 represent f L 、f E 、f S 、f V The weight ratio coefficient, and ω1+ω2+ω3+ω4=1; Track length cost function: In the formula, l o is the length of the corresponding track segment o; Path smoothing cost function: Among them, φ o , It is the deflection angle and pitch angle of the UAV during flight; Flight safety cost function: Among them, f o Represents the safety cost function between two adjacent navigation points, d o represents the straight-line distance between this track and the center of the nearest obstacle, M is the set safety threshold constant, and R i Represents the radius of the static obstacle; Collision penalty cost function: f V =V×P f Among them, the constant V∈(0,1) represents the penalty value; P f Characterizes the penalty coefficient.

6. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The implementation process of improving the initialization of the random population by using the composite chaotic mapping method in step (4) is as follows: Logistic-Sine-Cosine mapping is used to improve the coordinate distribution of the random x-axis and y-axis; Tent mapping is used to improve the coordinate distribution of the z-axis: X LSC =cos(π*4rx*(1-x)+(1-r)*sin(πx)-0.5)) Y LSC =cos(π*4rx*(1-x)+(1-r)*sin(πx)-0.5)),r∈(0,1) Tent Mapping:

7. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The implementation process of using the nonlinear convergence method to adjust the egg-laying boundary of the breeding dung beetles and the foraging boundary of the young dung beetles in step (4) is as follows: According to the planned UAV path objective function, the objective function value corresponding to each dung beetle in the initial population is calculated, and the optimal value, i.e. the minimum value, is recorded as Pop * , and record the corresponding three-dimensional coordinates at this time Let the initial iteration number t = 1, and set the size of the random number r1; Determine the size of r1. If r1 < 0.9, the dung beetle updates its position according to the rolling ball search behavior and sets a natural random number α1, which is a random value between 0 and 1. If α1 > 0.1, set α = 1, otherwise set α = -1. Generate a random number r2 between 0 and 0.2, and a random number r3 between 0 and 1; The search is performed using the following search equation: in, represents the position of the i-th rolling dung beetle at the t+1th iteration; represents the position of the i-th rolling dung beetle at the t-th iteration; represents the position of the i-th ball-rolling dung beetle at the t-1th iteration; Δx is used to simulate natural factors such as light intensity; represents the global worst position; If r1>0.9, the ball beetle determines the search strategy by simulating dancing, setting the random number r4 to a random integer between 0 and 180. If r4=0||90||180, the position search is performed according to the above search equation, otherwise the search is performed according to the following search equation: Among them, tan(aa) is the position offset of the rolling dung beetle when it encounters an obstacle; The breeding dung beetles update their positions by breeding behavior; the boundaries of the breeding area of ​​the breeding dung beetles are set in the following way: Among them, Lb * represents the lower boundary of the breeding area, Ub * represents the upper boundary of the breeding area, is the position information corresponding to the current local optimal solution, f1 is the boundary convergence factor of the breeding area, and the nonlinear convergence boundary method is used. k is Lb * and Ub * Any integer between ; The breeding dung beetles have arrived at the production area to hatch their eggs. The location has been updated as follows: in, represents the position of the i-th hatched egg at the t+1th iteration, represents the position of the i-th hatched egg at the t-th iteration, and are mutually independent random vectors of 1×3 dimensions; After the eggs are hatched, the young dung beetles are produced and use foraging behavior to update their positions. The optimal foraging area is set in the following way: In the formula, Ub ** and Lb ** Represent the upper and lower boundaries of the optimal foraging area, respectively, and f2 represents the boundary convergence factor of the foraging area. The nonlinear convergence boundary method is also used. The dung beetle reaches the foraging area to forage and updates its position based on the foraging behavior: Set the random number r5 to follow the normal distribution; generate a random vector ranging from 0 to 1 in, is the position of the i-th dung beetle at the t+1th iteration, is the position of the i-th dung beetle at the t-th iteration; another part of the dung beetles will steal and update their positions by stealing; Randomly generate a random vector of size 1×3 that follows a normal distribution Generate a fixed constant r8; The stealing dung beetle steals and updates its location by stealing: Calculate the objective function value of each dung beetle in the newly generated population Pop(i+1), and record the minimum value as Pop fit , compared with Pop fit and size, and will Update to the smaller value between the two.

8. The method for tracking the path of an unmanned aerial vehicle in a mountainous environment according to claim 1, characterized in that: The implementation process of introducing the gradient rapid drop population adjustment method in step (4) to dynamically adjust the proportion of each dung beetle species in the population during the iteration process is as follows: After each iteration, the best individual in the population is selected to replace the best individual in the current population; after every quarter of the total number of iterations, the proportion of rolling dung beetles is reduced, and the proportions of breeding dung beetles and small dung beetles are increased; In the formula, r Percent When the population is initialized, the proportion of the population occupied by the rolling ball dung beetle, v1, v2, v3 are weight constants related to the number of iterations; when the number of iterations t is equal to MAX it Will As the optimal solution.

Images