Unmanned aerial vehicle path planning method based on Chebyshev chaotic mapping sparrow optimization

Abstract

The invention relates to the technical field of unmanned aerial vehicle path planning in a complex environment, and provides a complex environment unmanned aerial vehicle path planning method based on Chebyshev chaotic mapping sparrow optimization. Under the constraint of flight distance and height, a total cost function is constructed, Chebyshev chaotic mapping is utilized to initialize a sparrow population, population diversity is improved, global search and local development capabilities in an exploration rate balance position updating process are introduced, a producer and follower step length adjustment strategy based on the exploration rate is adopted, and a sparrow swarm optimization algorithm is established. According to the method, the self-adaptability of the step length in the position updating process of the producer and the follower is improved, the minimization of the total cost function is realized in the process from the early global search of the minimization of the total cost function to the later focusing local search, the optimal solution of the total cost function is sought, and the flight path which is short in distance, low in energy consumption and capable of avoiding the threat source is planned for the unmanned aerial vehicle.

Description

Technical Field

[0001] This invention relates to the field of UAV path planning technology in complex environments, and more specifically, to a UAV path planning method based on Chebyshev chaotic mapping sparrow optimization. Background Technology

[0002] With technological advancements and the expansion of the power industry, traditional power is evolving into a more powerful, data-driven smart grid. Smart grids deploy numerous intelligent sensing devices, such as smart meters, video surveillance equipment, and IoT nodes, to monitor and collect data in real time on various operational scenarios, including video surveillance of transmission line corridors, local business aggregation at substations, and distribution control operations. These intelligent sensing devices are typically fixed in specific locations, and when performing tasks such as video surveillance, they may not be able to cover all areas requiring monitoring, especially inaccessible or dangerous locations. Thanks to their high mobility and ease of deployment, unmanned aerial vehicles (UAVs) can carry lightweight dual-light pods within the smart grid communication network, rapidly reaching inaccessible or dangerous locations to collect monitoring data. This provides continuous coverage for areas requiring monitoring and can improve the coverage of video surveillance of transmission line corridors.

[0003] The path planning problem for unmanned aerial vehicle (UAV) flights has become a research hotspot. In complex three-dimensional environments, considering multiple constraints such as terrain, obstacle avoidance, radar, no-fly zones, energy consumption, and flight time, making reasonable path planning for autonomous UAV flight is a challenging problem. Reference 1 (Phung, MD; Ha, QP Safety-enhanced UAVpath planning with spherical vector-based particle swarm optimization. Applied Soft Computing 2021, 107, 107376.) designs a spherical vector-based particle swarm optimization (SPSO) algorithm to solve the global path planning problem for UAVs facing multiple threats. By leveraging the correspondence between particle positions and the UAV's velocity, turning angle, and climb / dive angle, it efficiently searches the UAV's configuration space to find the optimal path that minimizes the cost function. Reference 2 (Ait-Saadi, A.; Meraihi, Y.; Soukane, A.; Ramdane-Cherif, A.; Gabis, AB A novelhybrid Chaotic Aquila Optimization algorithm with Simulated Annealing for Unmanned Aerial Vehicles path planning. Computers and Electrical Engineering 2022, 104B, 108461. ISSN 0045-7906.) proposes a hybrid optimization scheme based on Chaotic Aquila Optimization and Simulated Annealing to solve the UAV path planning problem. Although these methods perform well in solving path planning problems, they often perform poorly in terms of global search capabilities, and research on improving methods with strong global search capabilities remains limited. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a UAV path planning method based on Chebyshev chaotic mapping sparrow optimization for UAV path planning in complex environments. This invention utilizes Chebyshev chaotic mapping to initialize sparrow populations, thereby improving population diversity, and designs a producer step size adjustment strategy based on the exploration rate. During the iteration process, it shifts from focusing on global search in the early stage to focusing on local search in the later stage, providing high-quality flight routes for UAVs.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for UAV path planning in complex environments based on Chebyshev chaotic mapping and sparrow optimization is proposed. In a three-dimensional space with multiple threat sources, under the constraints of UAV flight distance and altitude, a total cost function is constructed, including flight distance cost, threat source cost, energy consumption cost, and angle cost. During the minimization of the total cost function, a sparrow population is initialized using Chebyshev chaotic mapping to improve population diversity. A producer and follower step size adjustment strategy based on the exploration rate is used to improve the adaptability of the step size during producer and follower position updates. The sparrow serves as the solution for minimizing the total cost function. From the initial global search to the later focused local search, the method minimizes the total cost function, thus completing the UAV flight path planning. The specific steps include: Step 1. Construct the total cost function during the flight of the UAV. C : in, Indicates the cost of flight distance. Indicates the cost of threats. This indicates the cost of energy consumption. This represents the cost of the angle between the horizontal positions of two consecutive time slots. λ 1. λ 2. λ 3 and λ 4 represents the weighting factor corresponding to the cost, and ; Step 2. Using the total cost function obtained in Step 1 as the fitness function, perform population initialization based on Chebyshev chaotic mapping to improve population diversity: Represents the mapped individual, where These represent the minimum and maximum values ​​of the actual population variables, respectively; the fitness values ​​of all sparrows are represented by a vector: F XThe value in each row is the fitness value of the individual; Step 3. Introduce the exploration rate Balancing global search and local development capabilities during the position update process. The system updates the positions of producers and followers, adjusts the producer step size based on the exploration rate, improves the adaptability of the step size during the producer and follower position update process, minimizes the total cost function, and completes the UAV flight path planning.

[0006] Furthermore, in step 1, the entire flight time of the UAV is divided into N time slots, i.e. , δ n For each individual time slot, 1≤ n ≤N, in the nth time slot, the horizontal position of the UAV is denoted as The height is recorded as Flight distance cost for ; There are M threat regions in the three-dimensional space of multiple threat sources. Each threat region is modeled as a cylindrical or spherical region, and the threat cost is... Indicates the proximity of the UAV flight path to the threat area: , To prevent constants with a denominator of zero, Indicates the current position Distance to the m-th threat center It is an indicator function, representing the horizontal position of the UAV (Unmanned Aerial Vehicle). If the target area is entered, the value is 1; otherwise, it is 0. Energy consumption cost Calculation formula: , and Factors for adjusting the energy consumption weights of climb and level flight; Angle Cost Calculation formula: , Indicates the current position Horizontal position relative to the previous time slot The angle between them.

[0007] Furthermore, in step 2, when generating the initial population, the population size is P, and each individual in the population has a D-dimensional variable. For each dimension d=1,2,…D, a random initial value is selected. Sequences are generated iteratively using Chebyshev chaotic mapping. This forms P points, constituting an initial value sequence; these are then mapped to the actual range of variable values ​​to obtain the mapped individuals. Forming the initial population: .

[0008] Furthermore, in step 3, the producer position update formula is as follows: in, t This is the current iteration number. j =1,2… d , It is in the t During the nth iteration i Only sparrow j Dimension value, iter max It is the upper limit of the number of iterations. It is a random number. R 2 and ST These represent the alarm value and the safety threshold, respectively. R 2∈[0,1], ST∈[0.5,1]; L It is a 1× d The matrix.

[0009] Furthermore, in step 3, the follower's position update formula is as follows: in, X P It is the optimal position for producers; X worst It is currently the worst position globally; A It is a 1× where each element is randomly 1 or -1. d Matrix; L It is a 1× d The matrix.

[0010] In summary, the invention has the following beneficial effects: Unmanned aerial vehicles (UAVs) perform missions in three-dimensional space, flying over complex areas with multiple threat sources. Under the condition of meeting flight safety and constraints, this invention plans short-distance, low-energy flight paths for UAVs that avoid threat sources by finding the optimal solution to the total cost function. The optimal solution to the total cost function is found using the Chebyshev Chaotic Map SSA (CCMSSA) algorithm. The Chebyshev chaotic map is used to initialize the sparrow population, improving population diversity and introducing an exploration rate. By balancing global search and local development capabilities during the position update process, and employing a producer step size adjustment strategy based on the exploration rate, the adaptability of the step size during the producer position update process is improved, providing high-quality flight paths for unmanned aerial vehicles (UAVs). Attached Figure Description

[0011] Figure 1 This is a flowchart of the present invention; Figure 2 and Figure 3 This is a graph comparing the total cost of the present invention with different methods in complex scenario 1 and complex scenario 2. Detailed Implementation

[0012] The present invention will now be described in further detail with reference to the accompanying drawings.

[0013] It should be noted that, for ease of description, the descriptions of direction in the following text are consistent with the directions in the accompanying drawings, but they do not limit the structure of the present invention.

[0014] like Figures 1-3 As shown, this invention discloses a path planning method for unmanned aerial vehicles (UAVs) in complex environments based on Chebyshev chaotic mapping sparrow optimization. In a three-dimensional space with multiple threat sources, under the constraints of flight distance and altitude, a total cost function is constructed, including flight distance cost, threat source cost, energy consumption cost, and angle cost. During the minimization of the total cost function, a sparrow population is initialized using Chebyshev chaotic mapping to improve population diversity. Producers and followers are the two objects in the sparrow algorithm during the optimization process of minimizing the total cost function. A producer and follower step size adjustment strategy based on the exploration rate is used to improve the adaptability of the step size during the producer and follower position update process. The sparrow, as the solution for minimizing the total cost function, achieves the minimization of the total cost function through the process from global search in the early stage to focused local search in the later stage. The UAV performs its mission in three-dimensional space, flying over complex areas with multiple threat sources. Under the conditions of flight safety and constraints, this invention plans a short-distance, low-energy flight path for the UAV that avoids threat sources by finding the optimal solution that minimizes the total cost function. Specifically, the method includes the following steps: Step 1. Construct the total cost function during the flight of the UAV. C : λ 1. λ 2. λ 3 and λ 4 represents the weighting factor corresponding to the cost, and .

[0015] The entire flight time of a UAV is divided into N time slots, i.e. , δ n For each individual time slot, 1≤ n ≤N, in the nth time slot, the horizontal position of the UAV is denoted as The height is recorded as Flight distance cost for .

[0016] There are M threat regions in the three-dimensional space of multiple threat sources. Each threat region is modeled as a cylindrical or spherical region, and the threat cost is... Indicates the proximity of the UAV flight path to the threat area: , To prevent constants with a denominator of zero, Indicates the current position Distance to the m-th threat center It is an indicator function, representing the horizontal position of the UAV (Unmanned Aerial Vehicle). If the target area is entered, the value is 1; otherwise, it is 0.

[0017] Energy consumption cost Calculation formula: , and A factor used to adjust the energy consumption weights for climb and level flight.

[0018] Angle Cost Calculation formula: , Indicates the current position Horizontal position relative to the previous time slot The angle between them.

[0019] Step 2. Using the total cost function obtained in Step 1 as the fitness function, perform population initialization based on the Chebyshev chaotic mapping. The population, as a solution in the optimization process, represents the solution that minimizes the total cost function, thereby increasing population diversity and expanding the solution set of the total cost function. The optimization objective of this invention is... Using the total cost function as the fitness function, the sparrow algorithm based on Chebyshev chaotic mapping is used to complete the 3D obstacle avoidance path planning for the UAV: Sparrow Search Algorithm (SSA) is a social optimization method inspired by the collective intelligence, foraging, and anti-predation behaviors of sparrows. Sparrows are mainly divided into two types: producers and followers. Producers actively seek food sources, while followers obtain food by following producers. The position (solution) of the producer serves as a reference point or guiding direction for other individuals' searches. Followers are analogous to those individuals who have not yet found the optimal solution; they update their own positions based on the producer's position in hopes of finding a better solution. During the execution of the Sparrow Search Algorithm, individuals update their positions based on the currently found optimal solution. This process involves information sharing and cooperation between producers and followers to increase the probability of the entire group finding the global optimum.

[0020] Producers with better fitness values ​​have priority in obtaining food during the search process. Furthermore, producers are responsible for finding food and guiding the actions of the entire population. Therefore, producers can search for food in a wide area. In each iteration, the producer's position is updated as follows: (1) in t This is the current iteration number. . It is in the t During the nth iteration i Only sparrow j The value of the dimension. iter max It is the upper limit of the number of iterations. It is a random number. and These represent alarm values ​​and safety thresholds, respectively. Q It is a random number that follows a normal distribution. L It is a 1× d The matrix. When R 2 < ST This means there are no predators around, and producers enter an extensive search mode. If R 2 ≥ ST This means that some sparrows have already spotted the predator, and all the sparrows need to quickly fly to other safe areas. The follower's position update formula is as follows: (2) in, X P It is the optimal position for producers. X worst It is currently the worst position globally. A It is a 1× where each element is randomly 1 or -1. d The matrix. When i > n / At time 2, it indicates that the first i Followers with poor fitness are most likely to starve.

[0021] When generating the initial population, the population size is P, and each individual in the population has D-dimensional variables. For each dimension d=1,2,…D, a random initial value is chosen. Sequences are generated iteratively using Chebyshev chaotic mapping. This forms P points, constituting an initial value sequence; these are mapped to the actual range of variable values: Represents the mapped individual, where These are the minimum and maximum values ​​of the actual population variable; obtaining the mapped individuals Forming the initial population: (3) The fitness values ​​of all sparrows are represented by a vector: (4) F X The value in each row is the fitness value of the individual.

[0022] Step 3. Introduce the exploration rate Balancing global search and local development capabilities during the position update process. The producer position is updated. When updating the producer position, the random number Q in formula (1) and formula (2) is adjusted, Q = 1 / (1 + ... t The producer position update formula is as follows: (5) The formula for updating the position of followers is as follows: (6) in, t This is the current iteration number. j =1,2… d , It is in the t During the nth iteration i Only sparrow j Dimension value, iter max It is the upper limit of the number of iterations. It is a random number. and These represent the alarm value and the safety threshold, respectively. L It is a 1× d The matrix.

[0023] Adjusting the step size of producers and followers based on the exploration rate enhances the adaptability of the step size during producer position updates. In the optimization process of minimizing the total cost function, a larger step size in the early stage is conducive to finding the global optimal solution, while a smaller step size in the later stage is conducive to rapid convergence of the optimization, thus minimizing the total cost function. The positions where the total cost function of each time slot is minimized are connected to form the UAV flight path, thus completing the UAV flight path planning.

[0024] UAV from the starting point Departure, and eventual arrival at the destination. The threat area is modeled as a sphere and defined as follows: All units are meters. Indicates the radius of the threat area (obstacles, radar, no-fly zone). Indicates the altitude of the threat area (obstacles, radar, no-fly zone). Figure 2 In the complex scene 1, the parameters are: S;T (200,100,150); (800,800,150) are the starting and ending coordinates, respectively. O1 (300,200,150,100,250); O2 (500,150,150,150,250); O3 (650,550,150,100,10); O4 (100,170,150,70,100); O5 (480,720,150,180,510); O6 (690,800,150,60,100); O7 (910,690,150,110,110); O8 (500,380,150,100,480) are the three-dimensional coordinates, radius, and height of the threat area. Population size: 500.

[0025] Figure 3 Parameters for medium-complex scenario 2: S;T: (220,100,150); (820,780,150) are the three-dimensional coordinates of the starting point and the ending point, respectively; O1(280,210,160,110,230); O2(510,490,150,130,200); O3(660,560,150,110,150); O4(100,190,150,90,100); O5(520,720,150,180,400); O6(710,790,150,85,120); O7(900,670,150,115,100); O8(480,330,150,95,450) are the three-dimensional coordinates, radius, and height of the threat area. Population size: 500.

[0026] Figure 2 , Figure 3 The total cost of UAV path generation was simulated under two complex scenarios, and the total cost was compared with that of SPSO in Reference 1, SSA in Reference 3, and CCMSSA of this invention. The two complex scenarios have different start points, end points, and threat sources. The simulation results show that the total cost of CCMSSA of this invention is lower than that of Reference 1 and Reference 3 under the complex scenario.

[0027] Figure 3In the initial iteration phase, the total cost of this invention is slightly higher than that of SPSO in the first literature. However, as the iteration progresses, this invention demonstrates its advantage in finding the optimal path. Due to the introduction of Chebyshev chaotic mapping, this invention has good traversal properties, improves population diversity, and is beneficial for global search. Combined with the producer's exploration-based position update strategy, this invention can achieve a lower total cost when finding the lowest cost path.

[0028] This invention constructs a total cost function comprising flight distance cost, threat cost, energy consumption cost, and angle cost. Under constraints of flight distance and altitude, it establishes a path planning problem for UAVs with the objective of minimizing this total cost function. This invention solves the path planning problem using the Chebyshev Chaotic Map SSA (CCMSSA) algorithm. It initializes the sparrow population using Chebyshev chaotic mapping to improve population diversity and designs a producer step size adjustment strategy based on the exploration rate to enhance the adaptability of the step size during producer position updates. As iterations progress, this invention shifts from focusing on global search in the early stages to focusing on local search in the later stages. This invention achieves a lower total cost and provides high-quality flight paths for UAVs.

[0029] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A path planning method for unmanned aerial vehicles (UAVs) in complex environments based on Chebyshev chaotic mapping sparrow optimization, characterized in that: In a three-dimensional space with multiple threat sources, under the constraints of flight distance and altitude of a UAV, a total cost function is constructed, including flight distance cost, threat source cost, energy consumption cost, and angle cost. During the minimization of the total cost function, a Chebyshev chaotic mapping is used to initialize a sparrow population, improving population diversity. A producer and follower step size adjustment strategy based on the exploration rate is used to improve the adaptability of the step size during producer and follower position updates. The sparrow serves as a solution for minimizing the total cost function. From the initial global search to the later focused local search, the total cost function is minimized, completing the UAV flight path planning. Specifically, the following steps are included: Step 1. Construct the total cost function during the flight of the UAV. C : in, Indicates the cost of flight distance. Indicates the cost of threats. This indicates the cost of energy consumption. This represents the cost of the angle between the horizontal positions of two consecutive time slots. λ 1. λ 2. λ 3 and λ 4 represents the weighting factor corresponding to the cost, and ; Step 2. Using the total cost function obtained in Step 1 as the fitness function, perform population initialization based on Chebyshev chaotic mapping to improve population diversity: Represents the mapped individual, where These represent the minimum and maximum values ​​of the actual population variables, respectively; the fitness values ​​of all sparrows are represented by a vector: F X The value in each row is the fitness value of the individual; Step 3. Introduce the exploration rate Balancing global search and local development capabilities during the position update process. The system updates the positions of producers and followers, adjusts the producer step size based on the exploration rate, improves the adaptability of the step size during the producer and follower position update process, minimizes the total cost function, and completes the UAV flight path planning.

2. The UAV path planning method for complex environments according to claim 1, characterized in that: In step 1, the entire flight time of the UAV is divided into N time slots. , δ n For each individual time slot, 1≤ n ≤N, in the nth time slot, the horizontal position of the UAV is denoted as The height is recorded as Flight distance cost for ; There are M threat regions in the three-dimensional space of multiple threat sources. Each threat region is modeled as a cylindrical or spherical region, and the threat cost is... Indicates the proximity of the UAV flight path to the threat area: , To prevent constants with a denominator of zero, Indicates the current position Distance to the m-th threat center It is an indicator function, representing the horizontal position of the UAV (Unmanned Aerial Vehicle). If the target area is entered, the value is 1; otherwise, it is 0. Energy consumption cost Calculation formula: , and Factors for adjusting the energy consumption weights of climb and level flight; Angle Cost Calculation formula: , Indicates the current position Horizontal position relative to the previous time slot The angle between them.

3. The UAV path planning method for complex environments according to claim 1, characterized in that: In step 2, when generating the initial population, the population size is P, and each individual in the population has D-dimensional variables. For each dimension d=1,2,…D, a random initial value is selected. Sequences are generated iteratively using Chebyshev chaotic mapping. This forms P points, constituting an initial value sequence; Mapping to the actual range of values ​​of the variable, we obtain the mapped individual. Forming the initial population: 。 4. The UAV path planning method for complex environments according to claim 1, characterized in that: In step 3, the producer position update formula is as follows: in, t This is the current iteration number. j =1,2… d , It is in the t During the nth iteration i Only sparrow j Dimension value, iter max It is the upper limit of the number of iterations. It is a random number. R 2 and ST These represent the alarm value and the safety threshold, respectively. R 2∈[0,1], ST∈[0.5,1]; L It is a 1× d The matrix.

5. The UAV path planning method for complex environments according to claim 1, characterized in that: In step 3, the position update formula for the follower is as follows: in, X P It is the optimal position for producers; X worst It is currently the worst position globally; A It is a 1× where each element is randomly 1 or -1. d Matrix; L It is a 1× d The matrix.

Images