A UAV 3D Path Planning Method Based on Enhanced Slime Mold Algorithm

Abstract

This invention relates to the field of UAV path planning technology and discloses a UAV 3D path planning method based on an enhanced slime mold algorithm. First, a multi-objective path planning mathematical model is constructed, comprehensively considering path length, threat avoidance, altitude constraints, and path smoothness. Based on real digital elevation model data, the optimal flight path is obtained by solving the above model using a chaotic enhanced leader slime mold algorithm. This chaotic enhanced leader slime mold algorithm optimizes the population initialization, weight adjustment, and position update processes by introducing a triple leader mechanism and a Logistic chaotic mapping perturbation strategy. It solves the technical problems of path planning in complex 3D environments easily getting trapped in local optima and having low convergence accuracy, thus planning a safe, smooth, and less costly feasible trajectory. It exhibits excellent performance in terms of range, threat avoidance capability, and flight stability, and is robust, making it suitable for autonomous UAV flight missions in complex scenarios such as agricultural monitoring and disaster patrol.

Description

Technical Field

[0001] This invention relates to the field of UAV path planning technology, specifically to a UAV 3D path planning method based on a chaotic enhanced leader slime mold algorithm. Background Technology

[0002] With their superior environmental adaptability and flexible deployment capabilities, unmanned aerial vehicles (UAVs) have been widely used in various fields such as agricultural monitoring, communication relay, disaster investigation, and urban governance. Path planning is the core technology for achieving fully autonomous flight of UAVs, directly determining their safety, efficiency, and reliability in performing tasks in complex three-dimensional environments. An ideal flight path must meet the physical and dynamic constraints of the UAV itself, avoid terrain obstacles and threat areas, and simultaneously optimize multiple objectives such as path length, flight energy consumption, and mission time.

[0003] However, the three-dimensional path planning problem is a typical NP-hard problem, characterized by high dimensionality, nonlinearity, and strong constraints, which limits the effectiveness of traditional path planning methods. For example, the grid-based A* search algorithm can guarantee global optimality in structured environments, but its computational complexity and memory consumption increase significantly with the scale of the environment, making it difficult to apply to large-scale scenarios; the artificial potential field method is computationally efficient but prone to getting trapped in local optima; and the Rapid Extended Random Tree (RRT) sampling method has good exploration capabilities in continuous high-dimensional spaces, but it is difficult to balance path quality and convergence efficiency.

[0004] In recent years, metaheuristic algorithms have become a research hotspot in UAV path planning due to their powerful global search capabilities and low dependence on problem structure. The Slime Mucin Algorithm (SMA), as an emerging metaheuristic algorithm, has advantages such as few parameters, simple structure, and fast convergence speed, and has been applied in multiple engineering fields. However, the standard SMA still suffers from problems such as being prone to getting trapped in local optima, insufficient convergence accuracy, and an imbalance between the exploration and development stages when dealing with complex high-dimensional optimization problems. Existing improved methods still have room for improvement in overall performance when dealing with complex UAV 3D path planning problems.

[0005] Therefore, there is an urgent need for a UAV path planning method that can effectively balance global exploration and local development capabilities, improve convergence accuracy and robustness, and is suitable for complex 3D environments. Summary of the Invention

[0006] One of the technical problems to be solved by this application is to overcome the defects of the above-mentioned related technologies and provide a UAV 3D path planning method based on the enhanced slime mold algorithm, so as to solve the technical problem that UAV path planning in complex 3D environments is prone to getting trapped in local optima and has low convergence accuracy.

[0007] The technical solution adopted by this invention to solve the technical problem is: a three-dimensional path planning method for unmanned aerial vehicles based on an enhanced slime mold algorithm, which specifically includes the following steps: S1 constructs a three-dimensional path planning mathematical model for the UAV, which includes a multi-objective cost function consisting of distance cost, safety cost, altitude cost, and smoothing cost; S2 constructs a three-dimensional flight environment that includes terrain obstacles and threat areas based on a real digital elevation model, and sets the flight constraint parameters for the UAV. S3 uses a chaotic enhanced leader slime mold algorithm to optimize and solve the three-dimensional path planning mathematical model, and obtains the optimal flight path that satisfies all flight constraint parameters; S4 outputs the coordinate sequence of the optimal path and the comprehensive cost evaluation results; The chaotic enhanced leader slime mold algorithm integrates a triple leader mechanism and a Logistic chaotic mapping perturbation strategy, specifically including: A1 uses Logistic chaotic mapping for population initialization and generates chaotic sequences to enhance population diversity. A2 introduces a triple leader mechanism, selecting the three individuals with the best fitness in the population as leaders to collaboratively guide the population's search direction; A3 is designed with an adaptive weight coefficient adjustment strategy based on Logistic chaotic mapping to balance the algorithm’s global exploration and local development capabilities. A4 utilizes a chaotic perturbation strategy to optimize the individual position update process, driving the population out of local optima.

[0008] Compared with related technologies, the present invention has the following advantages: 1. The proposed chaotic enhanced leader slime mold algorithm is deeply linked with the multi-objective cost function. Through chaotic initialization, triple leader mechanism, chaotic adaptive weight adjustment and chaotic perturbation strategy, it effectively balances global exploration and local development capabilities, significantly improves convergence accuracy, convergence speed and robustness, alleviates the problem of easily getting trapped in local optima and losing diversity in high-dimensional complex optimization problems, and can accurately search for the global optimal solution of the multi-objective cost function; 2. The constructed three-dimensional path planning mathematical model includes a multi-objective cost function, which comprehensively considers path length, safety, height constraints and smoothness. Each cost item is flexibly integrated through weight coefficients, which can adapt to the needs of different task scenarios and provide a scientific and reasonable objective guide for algorithm optimization. 3. The verification scenario built based on the real digital elevation model is close to the actual application. Simulation experiments show that the method of the present invention can plan a feasible flight path with shorter path, better safety performance and more stable flight. It shows excellent comprehensive performance in a variety of typical scenarios, with fewer approach times to threat areas and lower altitude deviation. 4. The computational complexity of the Chaotic Enhanced Leader Slime Algorithm is comparable to that of the original slime mold algorithm, showing a linear growth trend. It can efficiently complete path planning tasks under limited computing resources and is suitable for various autonomous flight scenarios of UAVs, such as agricultural monitoring, disaster inspection, and urban governance.

[0009] Preferably, the process of constructing the multi-objective cost function in step S1 is as follows: S11 Constructs the distance cost function: Calculates the sum of the Euclidean distances between all adjacent path points in the path to obtain the distance cost. ; S12 constructs a security cost function: Obstacles and threat areas are modeled as cylinders, and the security cost is calculated based on the shortest distance between the path segment and the threat area. ; S13 constructs the altitude cost function: Calculates the cost based on the deviation between the drone's flight altitude and the preset altitude range, thus obtaining the altitude cost. ; S14 Constructs a smoothing cost function: Calculates the path smoothing cost based on the horizontal steering angle and climb angle, obtaining the smoothing cost. ; S15 Constructs the comprehensive cost function: By integrating various cost items through weighting coefficients, the expression is as follows: ; in, w 1. w 2. w 3. w 4 represents the weighting coefficients of each cost item, and satisfies the following conditions: w 1+ w 2+ w 3+ w 4 = 1.

[0010] Preferably, the expression for the Logistic chaotic mapping is: ; in, It is a random function with a range of values ​​(0, 1) and is not in the set {0.25, 0.5, 0.75}; t This represents the current iteration number.

[0011] Preferably, the individual position update formula for the triple leader mechanism is: ; in, For the first i The individual The position of +1 iteration, , and The first The positions of the first three best individuals in the next iteration. These are the chaotic adaptive weighting coefficients. The numbers are uniformly random numbers in the range (0, 1). , The locations of two individuals are randomly selected.

[0012] Preferably, the chaotic adaptive weighting coefficients The expression is: ; in, For the first The Logistic chaotic mapping value of the next iteration takes the value in the range (0, 1), such that... It is dynamically adjusted within the range of (0.1, 0.6).

[0013] As a preferred approach, based on a real digital elevation model to recreate the actual terrain undulations, threat sources related to radar, artillery, and buildings are modeled as cylindrical no-fly zones with different radii and heights, and the starting point coordinates and ending point coordinates of the UAVs are set.

[0014] Preferably, the flight constraint parameters of the UAV include: a flight altitude limit of 100m to 200m, a UAV diameter of 1m, and a UAV's own safety radius. =0.5m, safe distance from obstacles =10m, the maximum heading angle and maximum pitch angle of the adjacent path segment are both 45°.

[0015] Preferably, the chaotic enhanced leader slime mold algorithm for optimizing and solving the three-dimensional path planning mathematical model specifically includes the following steps: S31 Path Encoding: The UAV path is encoded into an algorithm-recognizable individual vector. Each individual corresponds to a candidate path, the vector dimension is 3n, and each node contains radial distance. Pitch angle Azimuth The three parameters are used to obtain the three-dimensional coordinates of the path points through the transformation relationship between the spherical coordinate system and the Cartesian coordinate system; S32 Chaotic Population Initialization: Generation Based on Logistic Chaotic Mapping There are 1 initial individual, and the parameter values ​​of each individual are mapped to the search range of the corresponding variable through a chaotic sequence; S33 Fitness Evaluation: Substitute the path coordinates corresponding to each initial individual into the multi-objective cost function to calculate the fitness value F( ) for each individual. A smaller fitness value indicates better overall path performance; S34 Leader screening: sort the population according to the fitness value, and select the first three individuals with the optimal fitness values as leaders , and , which respectively correspond to the paths with the lowest, second lowest and third lowest comprehensive costs; S35 Update adaptive weight coefficient based on Logistic chaotic map , in the initial iteration stage, the randomness of the chaotic sequence makes maintain a large value to enhance the global exploration capability of the algorithm; in the later iteration stage, gradually decreases to 0.1 to 0.2 to strengthen the local exploitation capability of the algorithm; S36 Position update and chaotic perturbation: combine the triple-leader mechanism and the chaotic weight coefficient to update individual positions, and introduce a chaotic perturbation strategy to avoid search stagnation, with the specific update rules as follows: When the random number T 2<Z, the leader-guided update is adopted , When the random number T 2>Z, the chaotic perturbation update is adopted , wherein, is the current global optimal individual position, is a control parameter, , is a uniform random number in the (0, 1) interval, is the maximum number of iterations; S37 Boundary processing: perform boundary check on the updated individual parameters, and if the parameters exceed the preset range, regenerate the parameter values through the chaotic map to ensure all path points meet the UAV flight constraints; S38 Fitness update: substitute the updated individual coordinates into the multi-objective cost function again to calculate a new fitness value, and if the new fitness value is better than the historical optimal value, update the individual historical optimal position and the global optimal position ; S39 Iteration termination: repeat steps S34 to S38 until the maximum number of iterations Itermax is reached, and the path corresponding to the global optimal position at this time is the optimal flight path that meets the multi-objective optimization requirements.

[0016] As a preference, the computational complexity of the chaotic enhanced leader slime mold algorithm is: ; wherein, is the population size, is the maximum number of iterations, From the perspective of the problem dimension, This represents the number of obstacles. Attached Figure Description

[0017] Figure 1 This is a 3D environment example diagram of the UAV path planning in this application.

[0018] Figure 2 This is a diagram showing the coordinate system transformation relationship of this application.

[0019] Figure 3 This is a schematic diagram of the path security cost of this application.

[0020] Figure 4 This is a schematic diagram of the path height cost of this application.

[0021] Figure 5 This is a schematic diagram of the path smoothing cost of this application.

[0022] Figure 6 This is a flowchart of the chaotic enhanced leader slime mold algorithm of this application. Detailed Implementation

[0023] First, those skilled in the art should understand that these embodiments are merely used to explain the technical principles of the embodiments of this application and are not intended to limit the scope of protection of the embodiments of this application. Those skilled in the art can make adjustments as needed to adapt to specific application scenarios.

[0024] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0025] This invention relates to a three-dimensional path planning method for unmanned aerial vehicles (UAVs) based on an enhanced slime mold algorithm, specifically including the following steps: S1 constructs a three-dimensional path planning mathematical model for the UAV, which includes a multi-objective cost function consisting of distance cost, safety cost, altitude cost, and smoothing cost; S2 constructs a three-dimensional flight environment that includes terrain obstacles and threat areas based on a real digital elevation model, and sets the flight constraint parameters for the UAV. S3 uses the Chaotic Enhanced Leader Slimebacterium (CELSMA) algorithm to optimize and solve the three-dimensional path planning mathematical model, and obtains the optimal flight path that satisfies all flight constraint parameters; S4 outputs the coordinate sequence of the optimal path and the comprehensive cost evaluation results; In this embodiment, the chaotic enhanced leader slime mold algorithm integrates a triple leader mechanism and a Logistic chaotic mapping perturbation strategy, specifically including: A1 uses Logistic chaotic mapping for population initialization and generates chaotic sequences to enhance population diversity. A2 introduces a triple leader mechanism, selecting the three individuals with the best fitness in the population as leaders to collaboratively guide the population's search direction; A3 is designed with an adaptive weight coefficient adjustment strategy based on Logistic chaotic mapping to balance the algorithm’s global exploration and local development capabilities. A4 utilizes a chaotic perturbation strategy to optimize the individual position update process, driving the population out of local optima.

[0026] In this embodiment, the process of constructing the multi-objective cost function in step S1 is as follows: S11 constructs the distance cost function: calculates the sum of the Euclidean distances between all adjacent path points in the path, expressed as: ; in, For distance cost, and The first The and the first +1 path point's 3D coordinates This represents the number of nodes excluding the starting and ending points. S12 constructs a security cost function: Obstacles and threat areas are modeled as cylinders, and the security cost is calculated based on the shortest distance between the path segment and the threat area. The expression is: ; in, For safety costs, For the path segment to the first m The shortest distance from the center of each obstacle. For the drone's own safety radius, For the first m The radius of the obstacle, For minimum safe distance, The number of obstacles, This is a safety penalty coefficient; S13 constructs the altitude cost function: Calculates the cost based on the deviation between the drone's flight altitude and the preset altitude range. The expression is: ; in, Due to high costs, path point Flight altitude and These are the minimum and maximum flight altitudes of the drone, respectively. A high penalty coefficient; S14 Constructs a smoothing cost function: Calculates the path smoothing cost based on the horizontal steering angle and climb angle, with the following expression: ; in, To smooth out costs, and These are the penalty coefficients for the horizontal steering angle and the climb angle, respectively. The horizontal steering angle, For the climb angle, , ; S15 Constructs the comprehensive cost function: By integrating various cost items through weighting coefficients, the expression is as follows: ; in, w 1. w 2. w 3. w 4 represents the weighting coefficients of each cost item, and satisfies the following conditions: w 1+ w 2+ w 3+ w 4 = 1.

[0027] In this embodiment, the expression for the Logistic chaotic mapping is: ; in, It is a random function with a range of values ​​(0, 1) and is not in the set {0.25, 0.5, 0.75}; This represents the current iteration number.

[0028] In this embodiment, the individual position update formula of the triple leader mechanism is: ; in, For the first The individual The position of +1 iteration, , and The first The positions of the first three best individuals in the next iteration. These are the chaotic adaptive weighting coefficients. The numbers are uniformly random numbers in the range (0, 1). , The locations of two individuals are randomly selected.

[0029] In this embodiment, the chaotic adaptive weight coefficient The expression is: ; in, For the first The Logistic chaotic mapping value of the next iteration takes the value in the range (0, 1), such that... It is dynamically adjusted within the range of (0.1, 0.6).

[0030] In this embodiment, the real terrain undulations are restored based on the real digital elevation model, and the threat sources related to radar, artillery, and buildings are modeled as cylindrical no-fly zones with different radii and heights, and the starting point coordinates and ending point coordinates of the UAV are set.

[0031] In this embodiment, the flight constraint parameters of the UAV include: a flight altitude limit of 100m to 200m, a UAV diameter of 1m, and the UAV's own safety radius. =0.5m, safe distance from obstacles =10m, the maximum heading angle and maximum pitch angle of the adjacent path segment are both 45°.

[0032] In step S3 of this embodiment, the chaotic enhanced leader slime mold algorithm optimizes and solves the three-dimensional path planning mathematical model, specifically including the following steps: S31 Path Encoding: The UAV path is encoded into an algorithm-recognizable individual vector. Each individual corresponds to a candidate path, the vector dimension is 3n, and each node contains radial distance. Pitch angle Azimuth Three parameters, such as Figure 2 As shown, the three-dimensional coordinates of the path points are obtained through the transformation relationship between spherical coordinates and Cartesian coordinates: ; in, and The first and second parts of the UAV in the Cartesian coordinate system are respectively The and the first -1 path point coordinates , , ; S32 Chaotic Population Initialization: Generation Based on Logistic Chaotic Mapping There are 1 initial individual, and the parameter values ​​of each individual are mapped to the search range of the corresponding variable through a chaotic sequence, specifically: ; ; ; in, , These are the minimum and maximum values ​​of the radial distance. , are the minimum value and the maximum value of the pitch angle, , are the minimum value and the maximum value of the azimuth angle; S33 Fitness evaluation: Substituting the path coordinate corresponding to each initial individual into the described multi-objective cost function, and calculating to obtain the fitness value F( ) of each individual, where a smaller fitness value indicates better comprehensive performance of the path; S34 Leader screening: Sorting the population according to the fitness values, and selecting the first three individuals with the optimal fitness as leaders , and , which respectively correspond to the paths with the lowest comprehensive cost, the second lowest comprehensive cost and the third lowest comprehensive cost; S35 Updating the adaptive weight coefficient based on the Logistic chaotic mapping , in the initial iteration stage ( < / 3, is the maximum number of iterations), the randomness of the chaotic sequence makes maintain a large value (0.3 to 0.6), so as to enhance the global exploration capability of the algorithm; in the later iteration stage ( >2 / 3), gradually decreases to 0.1 to 0.2, so as to strengthen the local exploitation capability of the algorithm; S36 Position updating and chaotic perturbation: Updating the individual position by combining the triple-leader mechanism and the chaotic weight coefficient, and introducing a chaotic perturbation strategy to avoid search stagnation at the same time, and the specific updating rule is as follows: When the random number T 2<Z, the leader-guided updating is adopted , When the random number T 2>Z, the chaotic perturbation updating is adopted , wherein, T 2 is a uniformly distributed random number in the interval of (0, 1), is the position of the th individual in the th iteration, is the current global optimal individual position, is a control parameter, , is a uniformly distributed random number in the interval of (0, 1), is the maximum number of iterations; S37 Boundary Processing: Perform boundary checks on the updated individual parameters. If a parameter exceeds the preset range, regenerate the parameter value through chaotic mapping to ensure that all path points meet the UAV flight constraints. S38 Fitness Update: Substitute the updated individual coordinates back into the multi-objective cost function to calculate a new fitness value. If the new fitness value is better than the historical best value, then update the individual's historical best position and global best position. ; S39 Iteration Termination: Repeat steps S34-S38 until the maximum number of iterations (Itermax) is reached, at which point the global optimum is found. The corresponding path is the optimal flight path that satisfies the requirements of multi-objective optimization.

[0033] In this embodiment, the computational complexity of the chaotic enhanced leader slime mold algorithm is: ; in, For population size, The maximum number of iterations, The problem dimension is (3n). This represents the number of obstacles.

[0034] Verification Example Step 1: Construct as follows Figure 1 The mathematical model of UAV three-dimensional path planning shown 1.1. Distance Cost Function: Calculates the sum of distances between adjacent path nodes using Euclidean distance, where the number of path nodes... =12 (excluding the starting point and the ending point (target point); 1.2. Safety Cost Function: (e.g.) Figure 3 As shown, the obstacle is modeled as a cylinder, and the drone's own safe radius is... =1m, obstacle radius Minimum safe distance based on scenario settings (60m~80m) =10m, when the path segment Shortest distance to the center of the obstacle ≤ + When a collision is detected, the cost is set to infinity; when > + + No punishment at times; when + < ≤ + + Apply gradient penalty at the time; 1.3. High-cost function: such as Figure 4 As shown, the minimum flight altitude of the drone is set. =100m, maximum flight altitude =200m, ideal height =150m, calculate the cost based on the deviation between the path point height and the height range; 1.4. Smoothing Cost Function: Setting the Steering Angle Penalty Coefficient =1, Climb angle penalty coefficient =1, such as Figure 5 As shown, path segments By horizontal steering angle With climb angle Calculate smooth costs to avoid abrupt shifts and high changes; They are respectively Projection on the XY plane for At The projection onto a plane that is parallel to the XY plane.

[0035] 1.5. Overall Cost Function: Setting Weighting Coefficients w 1 = 0.4 (distance cost weight) w 2 = 0.3 (safety cost weight) w 3 = 0.2 (height cost weight) w 4 = 0.1 (smoothing cost weight), integrating various cost items to highlight the importance of height constraints and distance optimization.

[0036] Step 2: CELSMA Algorithm Parameter Settings 2.1. Population size =500, maximum number of iterations =200, Question Dimensions =36 (12 path nodes, 3 parameters per node); 2.2. Custom parameter Z=0.03, control parameter Adjust dynamically based on fitness value ,in A random number in the interval (0,1); 2.3. Initial values ​​of Logistic chaotic mapping Randomly select values ​​in the interval (0,1) that are not equal to {0.25,0.5,0.75}, and generate a dynamic sequence through chaotic mapping; 2.4. Chaotic Adaptive Weight Coefficients =0.5 +0.1, dynamically adjusted as the iteration progresses.

[0037] Step 3: Construction of 3D Flight Environment and Setting of Constraint Parameters 3.1. Based on real DEM data of Christmas Island, Australia, the terrain undulations were reconstructed, and four scenarios with increasing complexity were constructed. Scenario 1-4 each contained four obstacles, with obstacle radii ranging from 60m to 80m and heights from 100m to 150m. The obstacle distribution coordinates in Scenario 1 were: (450m, 300m, 120m), (550m, 500m, 130m), (650m, 200m, 140m), and (400m, 600m, 125m). The obstacle distribution coordinates in Scenario 2-4 were adjusted according to the principle of increasing complexity to ensure a reasonable gradient of difficulty. 3.2. The starting point coordinates of the UAV are (350m, 100m, 150m), and the ending point coordinates are (750m, 800m, 150m). 3.3. Flight constraint parameters: flight altitude 100m~200m, UAV diameter 1m, safe distance from obstacles 10m, maximum heading angle and pitch angle of adjacent path segments are both 45°.

[0038] Step 4: Optimization of CELSMA algorithm and multi-objective cost function 4.1. Path Encoding: Each individual vector contains the radial distance of 12 nodes. (50m~100m), pitch angle (-45°~45°), azimuth angle (0°~360°), Cartesian coordinates are obtained through spherical coordinate system transformation; 4.2. Chaotic Population Initialization: 500 initial individuals are generated based on Logistic chaotic mapping. Each parameter value is mapped to the corresponding range through a chaotic sequence to ensure uniform distribution of the initial path; 4.3. Fitness Evaluation: Substitute the path coordinates of each initial individual into the multi-objective cost function to calculate the fitness value F(…). The three individuals with the best fitness were selected as leaders. , , ; 4.4. Iterative Updates: (a1) Iterations 1-50 (exploration phase): Chaotic adaptive weight coefficients Maintaining the value between 0.3 and 0.6 enhances global exploration capabilities and allows for traversing the search space of multi-objective cost functions; (a2) Iterations 51-150 (transition phase): Gradually reduce to 0.1~0.3 to balance exploration and development capabilities; (a3) Iterations 151-200 (Development Phase): Stabilize at 0.1~0.2, strengthen local development, and finely optimize the optimal solution of the multi-objective cost function; (a4) Introduce chaotic perturbation after each iteration to avoid the population getting trapped in local optima and regenerate parameters that exceed the constraint range; 4.5. Iteration Termination: After 200 iterations, output the path corresponding to the globally optimal individual, which is the optimal solution of the multi-objective cost function.

[0039] like Figure 6 As shown, For chaotic variables, They are respectively The vector of changing trends; The search space is respectively The upper and lower bounds of a dimension. Leaders , , The corresponding number Next iteration position; , For the t-th iteration The, the Solution vectors , This represents two randomly selected slime mold locations.

[0040] Step 5: Result Evaluation Output the 3D coordinate sequence of the optimal path, calculated as follows: (b1) Maximum range: 801.515m for scenario 1, 849.313m for scenario 2, 824.906m for scenario 3, and 816.942m for scenario 4, all of which are better than the planning results of the standard SMA algorithm, A* algorithm and RRT algorithm; (b2) Number of times the threat area was approached: 0 times in scenario 2 and 1 time in all other scenarios, which is significantly lower than the 2-3 times of the comparison algorithm; (b3) Altitude deviation: The range is 521.617m~570.032m, and the flight altitude stability is better than the comparison algorithm (deviation 600m~750m).

[0041] The UAV 3D path planning method proposed in this invention, based on the chaotic enhanced leader slime mold algorithm, effectively solves the problems of local optima and convergence accuracy in path planning in complex 3D environments through the deep linkage of the CELSMA algorithm and multi-objective cost function. The planned path has shorter flight range, higher safety and better stability, and is suitable for a variety of UAV autonomous flight missions, with broad application prospects.

[0042] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A UAV 3D path planning method based on enhanced slime mold algorithm, characterized in that, Specifically, the steps include the following: S1 constructs a three-dimensional path planning mathematical model for the UAV, which includes a multi-objective cost function consisting of distance cost, safety cost, altitude cost, and smoothing cost; S2 constructs a three-dimensional flight environment that includes terrain obstacles and threat areas based on a real digital elevation model, and sets the flight constraint parameters for the UAV. S3 uses a chaotic enhanced leader slime mold algorithm to optimize and solve the three-dimensional path planning mathematical model, and obtains the optimal flight path that satisfies all flight constraint parameters; S4 outputs the coordinate sequence of the optimal path and the comprehensive cost evaluation results; The chaotic enhanced leader slime mold algorithm integrates a triple leader mechanism and a Logistic chaotic mapping perturbation strategy, specifically including: A1 uses Logistic chaotic mapping for population initialization and generates chaotic sequences to enhance population diversity. A2 introduces a triple leader mechanism, selecting the three individuals with the best fitness in the population as leaders to collaboratively guide the population's search direction; A3 is designed with an adaptive weight coefficient adjustment strategy based on Logistic chaotic mapping to balance the algorithm’s global exploration and local development capabilities. A4 utilizes a chaotic perturbation strategy to optimize the individual position update process, driving the population out of local optima.

2. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 1, characterized in that, The construction process of the multi-objective cost function in step S1 is as follows: S11 Constructs the distance cost function: Calculates the sum of the Euclidean distances between all adjacent path points in the path to obtain the distance cost. ; S12 constructs a security cost function: Obstacles and threat areas are modeled as cylinders, and the security cost is calculated based on the shortest distance between the path segment and the threat area. ; S13 constructs the altitude cost function: Calculates the cost based on the deviation between the drone's flight altitude and the preset altitude range, thus obtaining the altitude cost. ; S14 Constructs a smoothing cost function: Calculates the path smoothing cost based on the horizontal steering angle and climb angle, obtaining the smoothing cost. ; S15 Constructs the comprehensive cost function: By integrating various cost items through weighting coefficients, the expression is as follows: ； in, w 1. w 2. w 3. w 4 represents the weighting coefficients of each cost item, and satisfies the following conditions: w 1+ w 2+ w 3+ w 4 = 1.

3. The UAV 3D path planning method based on enhanced slime mold algorithm according to claim 1, characterized in that, The expression for the Logistic chaotic mapping is: ； in, It is a random function with a range of values ​​(0, 1) and is not in the set {0.25, 0.5, 0.75}; t This represents the current iteration number.

4. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 1, characterized in that, The individual position update formula for the triple leader mechanism is as follows: ； in, For the first i The individual t The position of +1 iteration, , and The first t The positions of the first three best individuals in the next iteration. These are the chaotic adaptive weighting coefficients. The numbers are uniformly random numbers in the range (0, 1). , The locations of two individuals are randomly selected.

5. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 4, characterized in that, The chaotic adaptive weight coefficient The expression is: ； in, For the first t The Logistic chaotic mapping value of the next iteration takes the value in the range (0, 1), such that... It is dynamically adjusted within the range of (0.1, 0.6).

6. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 1, characterized in that, Based on a real digital elevation model to recreate the actual terrain undulations, threat sources related to radar, artillery, and buildings are modeled as cylindrical no-fly zones with different radii and heights, and the starting point coordinates and ending point coordinates of UAVs are set.

7. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 1, characterized in that, The flight constraint parameters of the UAV include: flight altitude limited to 100m-200m, UAV diameter of 1m, and UAV's own safety radius. =0.5m, safe distance from obstacles =10m, the maximum heading angle and maximum pitch angle of the adjacent path segment are both 45°.

8. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 1, characterized in that, The chaotic enhanced leader slime mold algorithm optimizes and solves the three-dimensional path planning mathematical model by including the following steps: S31 Path Encoding: The UAV path is encoded into an algorithm-recognizable individual vector. Each individual corresponds to a candidate path, the vector dimension is 3n, and each node contains radial distance. Pitch angle Azimuth The three parameters are used to obtain the three-dimensional coordinates of the path points through the transformation relationship between the spherical coordinate system and the Cartesian coordinate system; S32 Chaotic Population Initialization: Generation Based on Logistic Chaotic Mapping There are 1 initial individual, and the parameter values ​​of each individual are mapped to the search range of the corresponding variable through a chaotic sequence; S33 Fitness Evaluation: Substitute the path coordinates corresponding to each initial individual into the multi-objective cost function to calculate the fitness value F( ) for each individual. A smaller fitness value indicates better overall path performance; S34 Leader Selection: The population is sorted according to fitness values, and the top three individuals with the best fitness are selected as leaders. , and These correspond to the paths with the lowest, second lowest, and third lowest overall costs, respectively. S35 updates adaptive weight coefficients based on Logistic chaotic mapping. In the early stages of iteration, the randomness of the chaotic sequence makes... Maintaining a larger value enhances the algorithm's global exploration capability; in the later stages of iteration, Gradually reduce it to 0.1–0.2 to enhance the algorithm's local development capabilities; S36 Position Update and Chaotic Perturbation: Individual position updates are performed by combining a triple leader mechanism and a chaotic weight coefficient. A chaotic perturbation strategy is also introduced to prevent search stagnation. The specific update rules are as follows: When the random number T 2<Z, the leader-guided update is adopted ， When random number T When 2>Z, chaotic perturbation is used for updating. ， in, This represents the current globally optimal individual position. For control parameters, , The numbers are uniformly random numbers in the interval (0, 1). This represents the maximum number of iterations. S37 Boundary Processing: Perform boundary checks on the updated individual parameters. If a parameter exceeds the preset range, regenerate the parameter value through chaotic mapping to ensure that all path points meet the UAV flight constraints. S38 Fitness Update: Substitute the updated individual coordinates back into the multi-objective cost function to calculate a new fitness value. If the new fitness value is better than the historical best value, then update the individual's historical best position and global best position. ; S39 Iteration Termination: Repeat steps S34-S38 until the maximum number of iterations (Itermax) is reached, at which point the global optimum is found. The corresponding path is the optimal flight path that satisfies the requirements of multi-objective optimization.

9. The UAV three-dimensional path planning method based on the enhanced slime mold algorithm according to claim 1, characterized in that, The computational complexity of the chaotic enhanced leader slime mold algorithm is: ； in, For population size, The maximum number of iterations, From the perspective of the problem dimension, This represents the number of obstacles.

Images