UAV trajectory planning method and system based on radio map in wind field environment

Abstract

The present invention discloses a method and system for planning UAV trajectories based on a radio map in a wind field environment. The method comprises: obtaining radio map initialization parameters and UAV initialization parameters to generate a radio map and a wind vector model; solving a weighted directed graph based on the radio map and the wind field vector model, wherein the weighted directed graph is used to store the energy consumption of each path; initializing the UAV access sequence based on the weighted directed graph, and optimizing the initial UAV access sequence using a genetic algorithm to obtain the optimal UAV access sequence. The present invention initializes the UAV access sequence based on the weighted directed graph, and optimizes the initial UAV access sequence using a genetic algorithm to obtain the optimal UAV access sequence. The energy consumption of UAV tasks in a wind field environment can be obtained through the radio map and the wind field vector model, and then the distribution sequence and UAV trajectory can be reasonably set to improve the distribution efficiency under the premise that the UAV communication quality meets the requirements.

Description

Technical Field

[0001] The present invention belongs to the field of path planning, and in particular relates to a method and system for planning the trajectory of an unmanned aerial vehicle (UAV) based on a radio map in a wind field environment. Background Art

[0002] As a key component of the low-altitude economy, drones have achieved technological penetration in multiple fields thanks to their high maneuverability and flexible deployment capabilities. In recent years, with the surge in express delivery services driven by e-commerce, urban terminal logistics networks are facing the dual challenges of optimizing energy efficiency and coordinating spatiotemporal resources. Compared to traditional manual delivery models, drone systems offer comparative advantages such as low space occupancy, controllable unit transportation costs, and the flexibility of vertical takeoff and landing. Compared to ground-based unmanned vehicles, their airspace penetration capabilities can circumvent the topological constraints of road traffic and enable vertical crossing of complex terrain (such as waterways and mountains), providing a new technical path for building logistics networks. Therefore, the application of drones in the logistics industry holds great promise.

[0003] However, when a drone is forced to adjust its flight altitude to maintain communication quality (e.g., crossing a signal attenuation zone), it may simultaneously enter an unfavorable wind field (e.g., a low-altitude turbulent layer), creating a conflict between communication assurance and energy efficiency. Traditional methods, such as geometric planning based on lines or circles, and planning based on the Transit Planner (TSP), are unable to resolve this conflict. This leads to inefficient communication resource allocation and trajectory optimization, excessive energy loss for the drone, and potentially flight accidents. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a method and system for UAV trajectory planning based on radio maps in a wind farm environment. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0005] A UAV trajectory planning method based on a radio map in a wind environment, comprising:

[0006] Obtain radio map initialization parameters and drone initialization parameters to generate radio maps and wind vector models;

[0007] solving a weighted directed graph according to the radio map and the wind field vector model, wherein the weighted directed graph is used to store energy consumption of each path;

[0008] The UAV access sequence is initialized based on the weighted directed graph, and the initial UAV access sequence is optimized using a genetic algorithm to obtain the optimal UAV access sequence.

[0009] In one specific embodiment, the drone initialization parameters include: hangar location, source user node set, destination user node set, drone flight altitude range, drone flight horizontal range, and drone flight speed;

[0010] Correspondingly, the wind vector model is: = + ,in, is the ground speed of the UAV, is the relative air speed of the drone, is the wind speed, , Wind speed The absolute value of Wind speed The angle with the positive horizontal axis.

[0011] In a specific embodiment, solving a weighted directed graph based on the radio map and the wind vector model includes:

[0012] Obtaining a vertex set and an edge set of a weighted directed graph to obtain a weight expression for a UAV flying between any two points, wherein the weight is determined by a flight path and energy consumption of the UAV corresponding to the flight path;

[0013] Converting the solution of the weight into the solution of a mobility cost function; wherein the mobility cost function is determined by the drone energy consumption and the drone signal strength, the drone energy consumption and the drone signal strength are related to the flight time, and the flight time is related to the wind field vector model;

[0014] Traverse each source user node set to obtain the solution of the mobility cost function between any two points that meets the preset conditions to obtain the minimum energy consumption path;

[0015] Each of the minimum energy consumption paths is stored to obtain a weighted directed graph.

[0016] In a specific embodiment, the movement cost function is: ;

[0017] in,

[0018]

[0019] Among them, E(v,t) represents the instantaneous energy consumption of the UAV, Indicates the airspeed of the drone, Indicates the flight time of the drone, represents the signal-to-noise ratio at the drone’s location, Indicates whether the SNR value at a certain location reaches the threshold that allows drones to pass through. , indicating that the mission allows the drone to pass, otherwise the drone is not allowed to pass.

[0020] In a specific embodiment, initializing the drone access sequence based on the weighted directed graph includes:

[0021] Obtain a source user node set and a destination user node set to obtain a plurality of drone user groups, wherein each drone user group includes a source user node and a destination user node;

[0022] An initial strategy is set to process each drone user group to obtain an original drone access sequence, wherein the original drone access sequence is the source user node and the destination user node in sequence;

[0023] Determine whether the remaining payload margin of the drone satisfies the requirement for continuous access to the next source user node. If so, calculate the full-segment energy consumption when continuously accessing the next source user node in sequence according to the weighted directed graph, and when the energy consumption when continuously accessing the next source user node is greater than the current energy consumption, use the drone access sequence corresponding to the current energy consumption as the initial drone access sequence.

[0024] In one embodiment, a genetic algorithm is used to optimize the initial UAV access sequence to obtain the optimal UAV access sequence, including:

[0025] Generate the initial drone chromosome according to the initial drone visit sequence;

[0026] The tournament selection method is used to perform crossover, mutation and replication operations on the initial drone chromosomes to obtain offspring individuals;

[0027] The fitness of each parent individual and offspring individual is iteratively calculated so that the sequence order corresponding to the drone chromosome with the optimal fitness function is used as the optimal drone access order, wherein the fitness function is determined by the path energy consumption of the drone flying in the wind field.

[0028] The present invention also discloses a UAV trajectory planning system based on a radio map in a wind field environment, comprising:

[0029] An initialization module, used to obtain radio map initialization parameters and UAV initialization parameters to generate radio maps and wind vector models;

[0030] a weighted directed graph generation module, configured to solve a weighted directed graph according to the radio map and the wind field vector model, wherein the weighted directed graph is used to store the energy consumption of each path;

[0031] The UAV access sequence optimization module initializes the UAV access sequence based on the weighted directed graph, and uses a genetic algorithm to optimize the initial UAV access sequence to obtain the optimal UAV access sequence.

[0032] In a specific embodiment, the weighted directed graph generation module includes:

[0033] a weight expression calculation unit, configured to obtain a vertex set and an edge set of a weighted directed graph to obtain a weight expression for a UAV flying between any two points, wherein the weight is determined by a flight path and energy consumption of the UAV corresponding to the flight path;

[0034] a mobility cost function solving unit, configured to convert the weight solution into a mobility cost function solution; wherein the mobility cost function is determined by drone energy consumption and drone signal strength, wherein the drone energy consumption and drone signal strength are related to flight time, and the flight time is related to a wind field vector model;

[0035] A minimum energy consumption path calculation unit is used to traverse each source user node set to obtain a solution of a movement cost function between any two points that meets preset conditions to obtain a minimum energy consumption path;

[0036] The weighted directed graph generating unit is used to store each of the minimum energy consumption paths to obtain a weighted directed graph.

[0037] In one specific embodiment, the drone access sequence optimization module specifically includes:

[0038] A drone user group generating unit is configured to obtain a source user node set and a destination user node set to obtain a plurality of drone user groups, wherein each drone user group includes a source user node and a destination user node;

[0039] An original UAV access sequence processing unit is used to set an initial strategy to process each UAV user group to obtain an original UAV access sequence, wherein the original UAV access sequence is the source user node and the destination user node in sequence;

[0040] The initial UAV access sequence calculation unit is used to determine whether the remaining payload margin of the UAV satisfies the requirement for continuous access to the next source user node. If so, the full-segment energy consumption when continuously accessing the next source user node is calculated in sequence according to the weighted directed graph, and when the energy consumption when continuously accessing the next source user node is greater than the current energy consumption, the UAV access sequence corresponding to the current energy consumption is used as the initial UAV access sequence.

[0041] In a specific embodiment, the drone access sequence optimization module further includes:

[0042] A chromosome initialization unit, used to generate an initial drone chromosome according to the initial drone visit sequence;

[0043] The offspring individual generation unit is used to perform crossover, mutation and replication operations on the initial drone chromosomes using the tournament selection method to obtain offspring individuals;

[0044] The optimal drone access sequence optimization unit is used to iteratively calculate the fitness of each parent individual and offspring individual so that the sequence order corresponding to the drone chromosome with the optimal fitness function is used as the optimal drone access sequence, wherein the fitness function is determined by the path energy consumption of the drone flying in the wind field.

[0045] Beneficial effects of the present invention:

[0046] The present invention's radio map-based drone trajectory planning method for a wind farm environment includes: obtaining radio map initialization parameters and drone initialization parameters to generate a radio map and a wind vector model; solving a weighted directed graph based on the radio map and wind vector model, wherein the weighted directed graph is used to store the energy consumption of each path; initializing the drone access sequence based on the weighted directed graph, and optimizing the initial drone access sequence using a genetic algorithm to obtain the optimal drone access sequence. In this way, the radio map and wind vector model can be used to determine the drone mission energy consumption in a wind farm environment, thereby rationally setting the delivery sequence and drone trajectory to improve delivery efficiency while ensuring that drone communication quality meets requirements.

[0047] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flow chart of a method for UAV trajectory planning based on a radio map in a wind farm environment provided by an embodiment of the present invention;

[0049] Figure 2 This is a module block diagram of a radio map-based drone trajectory planning system in a wind farm environment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The present invention will be further described in detail below with reference to specific examples, but the embodiments of the present invention are not limited thereto.

[0051] Example 1

[0052] See Figure 1 , Figure 1 The present invention provides a flow chart of a method for planning a UAV trajectory based on a radio map in a wind farm environment, including:

[0053] S1. Obtain radio map initialization parameters and UAV initialization parameters to generate a radio map and a wind vector model;

[0054] The radio map initialization parameters mainly include discrete precision , number of base stations m, measured data ; UAV initialization parameters mainly include: hangar location , source user node set , destination user node set , UAV flight altitude range , the UAV's horizontal flight range W×W and the UAV's flight speed ;

[0055] Correspondingly, the wind vector model is: = + ,in, is the ground speed of the UAV, is the relative air speed of the drone, is the wind speed, , Wind speed The absolute value of Wind speed The angle with the positive horizontal axis.

[0056] S2. Solving a weighted directed graph according to the radio map and the wind field vector model, wherein the weighted directed graph is used to store the energy consumption of each path;

[0057] In a specific embodiment, solving a weighted directed graph based on the radio map and the wind vector model includes:

[0058] S21. Obtaining a vertex set and an edge set of a weighted directed graph to obtain a weight expression for a UAV flying between any two points, wherein the weight is determined by a flight path and energy consumption of the UAV corresponding to the flight path;

[0059] It should be noted that a weighted directed graph is a mathematical model used to describe relationships with directionality and weight. It consists of a set of vertices (nodes) and a set of directed edges (arcs), each of which has a specific directionality, pointing from one vertex to another. In addition, each edge is assigned a numerical weight, which is typically used to represent a quantitative relationship between two vertices, such as time, cost, or distance. In this embodiment, the vertex set of the weighted directed graph is represented as:

[0060]

[0061] in and Indicates the starting point and end point respectively.

[0062] The edge set can be represented as:

[0063]

[0064] The weight of each edge can be given by:

[0065] ,

[0066] in express and The distance between Indicates that the drone follows the edge [ ]The energy consumed per unit time during flight, the above formula shows that the drone flies along the edge [ ]The weight of the flight and the The weights are not equal when flying.

[0067] S22. Converting the solution of the weight into the solution of a mobility cost function; wherein the mobility cost function is determined by drone energy consumption and drone signal strength, wherein the drone energy consumption and drone signal strength are related to flight time, and the flight time is related to a wind field vector model;

[0068] In order to filter out flight paths that meet the communication requirements, this embodiment defines a function:

[0069]

[0070] In this embodiment, the moving cost is defined as:

[0071] ,

[0072] Among them, E(v,t) represents the instantaneous energy consumption of the UAV, Indicates the airspeed of the drone, Indicates the flight time of the drone, represents the signal-to-noise ratio at the drone’s location, Indicates whether the SNR value at a certain location reaches the threshold that allows drones to pass through. , indicating that the mission allows the drone to pass, otherwise the drone is not allowed to pass.

[0073] This means that at time t, if the signal strength at the drone’s location is less than the threshold When , the movement cost of this location is considered infinite, which means that the drone is not allowed to pass through this location.

[0074] And under the influence of the wind field, there are:

[0075]

[0076] The distance between any two points is known, so .

[0077] To ensure that the drone can Departure, traverse all cruise points and finally fly to the end point , and will not traverse itself, the following constraints are also required:

[0078] .

[0079] S23, traversing each source user node set to obtain a solution of a mobility cost function between any two points that satisfies a preset condition to obtain a minimum energy consumption path;

[0080] S24. Storing each of the minimum energy consumption paths to obtain a weighted directed graph.

[0081] S3. Initialize the drone access sequence based on the weighted directed graph, and use a genetic algorithm to optimize the initial drone access sequence to obtain the optimal drone access sequence.

[0082] In a specific embodiment, initializing the drone access sequence based on the weighted directed graph includes:

[0083] S31. Obtain a source user node set and a destination user node set to obtain a plurality of drone user groups, wherein each drone user group includes a source user node and a destination user node;

[0084] In this embodiment, the source user node refers to the drone pickup node, and the destination user node refers to the drone delivery node. The source and destination user nodes appear in pairs. In some examples, the drone hangar, which is the initial node before the drone picks up the goods, and the final node after the drone has delivered all the goods, may also be included.

[0085] S32. Setting an initial strategy to process each drone user group to obtain an original drone access sequence, wherein the original drone access sequence is the source user node and the destination user node in order;

[0086] Specifically, the initial strategy might be to first block the source user node for each group of drone users. Then, randomly select a node from among the unblocked destination user nodes as the next access node and block it. If the selected node is the destination user node, its paired source user node is also unblocked. Repeat this process until all destination user nodes are blocked, completing the construction of the drone access sequence. Finally, the drone's feasible sequence is prepended with the initial node before the drone picks up the goods and the final node after the drone has delivered all the goods. These sequences are then concatenated to form the drone's PDP (Pickup-Delivery-Pickup) chromosome.

[0087] S33. Determine whether the remaining payload margin of the drone satisfies the requirement for continuous access to the next source user node. If so, calculate the full-segment energy consumption when continuously accessing the next source user node in sequence according to the weighted directed graph, and when the energy consumption when continuously accessing the next source user node is greater than the current energy consumption, use the drone access sequence corresponding to the current energy consumption as the initial drone access sequence.

[0088] It's important to note that in drone logistics mission planning, the energy efficiency paradox caused by payload constraints warrants special attention. After completing its initial pickup, the drone's remaining payload capacity only allows for a single additional pickup, necessitating immediate delivery to avoid exceeding the maximum payload threshold. However, if sufficient payload capacity remains, it can opt for sequential pickups followed by continuous deliveries. Of course, due to payload capacity constraints, the number of sequential pickups or continuous deliveries may vary, and it's best to incorporate at least three steps of decision-making. If a continuous "pick-up-delivery" operation is adopted, a PPD can be achieved. However, whether the continuous delivery approach is optimal still requires energy consumption assessment. If the continuous operation is determined to be suboptimal, the original solution remains. Specifically, it also includes a load limit detection step. For a certain drone access sequence, the real-time load of the drone at each access point is calculated. At the same time, considering that the actual drone has a certain degree of redundancy in cargo capacity when it is designed, it is stipulated that when the drone picks up the cargo, if the cargo carried by the drone exceeds the set cargo limit, the delivery point paired with the pickup point will be advanced to after the pickup point.

[0089] In one embodiment, a genetic algorithm is used to optimize the initial UAV access sequence to obtain the optimal UAV access sequence, including:

[0090] S34. Generate an initial drone chromosome according to the initial drone access sequence; that is, similar to the aforementioned PDP....PDP.

[0091] A tournament selection algorithm is used to perform crossover, mutation, and replication on the initial drone chromosomes to generate offspring individuals. Specifically, an appropriate number of individuals are selected from the population and subjected to crossover, mutation, and replication to generate offspring individuals. Regarding the crossover operation, since a chromosome contains the access sequences of two drones, crossover operations can be divided into intra-segment crossover and inter-segment crossover. Intra-segment crossover can be understood as exchanging access points between the two drones. Due to the presence of a pairing relationship, during the crossover operation, each drone randomly selects a pair of pickup and delivery points to exchange, while the rest remain unchanged. Subsequently, for inter-segment crossover, an order-based crossover (OBX) is used. To avoid conflicts, the position of the separator is first determined. Next, one of the two parents selected by the tournament selection operator is randomly selected as parent individual 1. The delivery node and all nodes corresponding to the separator position are masked. Multiple crossover points are randomly selected from the unmasked pickup nodes to obtain the locations of their paired nodes. The pickup and delivery pairs that require crossover are placed in offspring individual 1 in the order of parent individual 1. After removing the crosslinked genes, the remaining genes are inserted in the order of parent individual 2. Swap parent individuals 1 and 2, and repeat this operation to obtain offspring individual 2. The mutation operation simulates the genetic mutation process in the biological world. Its purpose is to increase perturbations by generating non-existent chromosomes, preventing the genetic algorithm from prematurely falling into a local optimum during the optimization process. This refers to the concept of crossover within a market segment. For each chromosome, two pairs of take-and-send pairs are randomly selected for exchange, regardless of whether they are in the same drone's access sequence. Furthermore, this embodiment uses a neighborhood search method to further avoid local convergence. For the "elite" individuals with the highest fitness in the population, their chromosomes are replicated during the replication operation to produce offspring individuals, embodying the principle of "survival of the fittest" in evolutionary theory. This operation ensures that the genes of individuals with excellent traits in the population are passed on. During the iterative process of the genetic algorithm, high-quality genes are continuously passed on to the next generation, maintaining high genetic diversity in the population. By directly replicating the chromosomes of elite individuals, offspring individuals have a relatively high foundation of adaptability from the outset, helping the entire population evolve towards a more optimal solution. At the same time, this simple and direct replication method, while ensuring the efficiency of the algorithm, also strengthens the screening and retention mechanism of dominant genes in the genetic algorithm, providing high-quality genetic material for crossover and mutation operations, so that new individuals subsequently generated through crossover and mutation can be explored and optimized from a higher starting point, promoting the population to continuously approach the global optimal solution in the search space.

[0092] S25. Iteratively calculate the fitness of each parent and offspring individual to make the sequence order corresponding to the drone chromosome with the optimal fitness function as the optimal drone access order, wherein the fitness function is determined by the energy consumption of the drone's flight path in the wind field. For example, by merging the offspring population and the parent population, calculating the fitness, and eliminating the last 50% of individuals to control the population size. At the same time, after performing the above operations, calculate the fitness of the offspring individual. When its fitness is lower than the worst individual in the parent population, it will be marked as an invalid solution, and then repeat the above operations until its fitness is higher than the worst individual. At this point, a genetic process is completed. Repeat the above operations until the iteration is completed to obtain the multi-drone access order and initialization trajectory.

[0093] The radio map-based drone trajectory planning method for a wind farm environment in this embodiment includes: obtaining radio map initialization parameters and drone initialization parameters to generate a radio map and a wind vector model; solving a weighted directed graph based on the radio map and wind vector model, wherein the weighted directed graph is used to store the energy consumption of each path; initializing the drone access sequence based on the weighted directed graph, and optimizing the initial drone access sequence using a genetic algorithm to obtain the optimal drone access sequence. In this way, the energy consumption of drone missions in a wind farm environment can be determined using the radio map and wind vector model, and the delivery sequence and drone trajectory can be reasonably set to improve delivery efficiency while ensuring that the drone communication quality meets the requirements.

[0094] Please continue to see Figure 2 The present invention also discloses a UAV trajectory planning system based on a radio map in a wind field environment, comprising:

[0095] An initialization module, used to obtain radio map initialization parameters and UAV initialization parameters to generate radio maps and wind vector models;

[0096] a weighted directed graph generation module, configured to solve a weighted directed graph according to the radio map and the wind field vector model, wherein the weighted directed graph is used to store the energy consumption of each path;

[0097] The UAV access sequence optimization module initializes the UAV access sequence based on the weighted directed graph, and uses a genetic algorithm to optimize the initial UAV access sequence to obtain the optimal UAV access sequence.

[0098] In a specific embodiment, the weighted directed graph generation module includes:

[0099] a weight expression calculation unit, configured to obtain a vertex set and an edge set of a weighted directed graph to obtain a weight expression for a UAV flying between any two points, wherein the weight is determined by a flight path and energy consumption of the UAV corresponding to the flight path;

[0100] a mobility cost function solving unit, configured to convert the weight solution into a mobility cost function solution; wherein the mobility cost function is determined by drone energy consumption and drone signal strength, wherein the drone energy consumption and drone signal strength are related to flight time, and the flight time is related to a wind field vector model;

[0101] A minimum energy consumption path calculation unit is used to traverse each source user node set to obtain a solution of a movement cost function between any two points that meets preset conditions to obtain a minimum energy consumption path;

[0102] The weighted directed graph generating unit is used to store each of the minimum energy consumption paths to obtain a weighted directed graph.

[0103] In one specific embodiment, the drone access sequence optimization module specifically includes:

[0104] A drone user group generating unit is configured to obtain a source user node set and a destination user node set to obtain a plurality of drone user groups, wherein each drone user group includes a source user node and a destination user node;

[0105] An original UAV access sequence processing unit is used to set an initial strategy to process each UAV user group to obtain an original UAV access sequence, wherein the original UAV access sequence is the source user node and the destination user node in sequence;

[0106] The initial UAV access sequence calculation unit is used to determine whether the remaining payload margin of the UAV satisfies the requirement for continuous access to the next source user node. If so, the full-segment energy consumption when continuously accessing the next source user node is calculated in sequence according to the weighted directed graph, and when the energy consumption when continuously accessing the next source user node is greater than the current energy consumption, the UAV access sequence corresponding to the current energy consumption is used as the initial UAV access sequence.

[0107] In a specific embodiment, the drone access sequence optimization module further includes:

[0108] A chromosome initialization unit, used to generate an initial drone chromosome according to the initial drone access sequence;

[0109] The offspring individual generation unit is used to perform crossover, mutation and replication operations on the initial drone chromosomes using the tournament selection method to obtain offspring individuals;

[0110] The optimal drone access sequence optimization unit is used to iteratively calculate the fitness of each parent individual and offspring individual so that the sequence order corresponding to the drone chromosome with the optimal fitness function is used as the optimal drone access sequence, wherein the fitness function is determined by the path energy consumption of the drone flying in the wind field.

[0111] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification.

[0112] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art can understand and implement other changes to the disclosed embodiments by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple situations. A single processor or other unit can implement several functions listed in the claims. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0113] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. For those skilled in the art of the present invention, without departing from the concept of the present invention, several simple deductions or substitutions can be made, which should be considered to fall within the scope of protection of the present invention.

Claims

1. A UAV trajectory planning method based on radio maps in a wind environment, characterized in that: include: Obtain radio map initialization parameters and drone initialization parameters to generate radio maps and wind vector models; solving a weighted directed graph according to the radio map and the wind field vector model, wherein the weighted directed graph is used to store energy consumption of each path; Initializing the UAV access sequence based on the weighted directed graph, and optimizing the initial UAV access sequence using a genetic algorithm to obtain an optimal UAV access sequence; Solving a weighted directed graph according to the radio map and the wind field vector model, comprising: Obtaining a vertex set and an edge set of a weighted directed graph to obtain a weight expression for a UAV flying between any two points, wherein the weight is determined by a flight path and energy consumption of the UAV corresponding to the flight path; Converting the solution of the weight into the solution of a mobility cost function; wherein the mobility cost function is determined by the drone energy consumption and the drone signal strength, the drone energy consumption and the drone signal strength are related to the flight time, and the flight time is related to the wind field vector model; Traverse each source user node set to obtain the solution of the mobility cost function between any two points that meets the preset conditions to obtain the minimum energy consumption path; Storing each of the minimum energy consumption paths to obtain a weighted directed graph; Initializing the drone access sequence based on the weighted directed graph includes: Obtain a source user node set and a destination user node set to obtain a plurality of drone user groups, wherein each drone user group includes a source user node and a destination user node; An initial strategy is set to process each drone user group to obtain an original drone access sequence, wherein the original drone access sequence is the source user node and the destination user node in sequence; Determine whether the remaining payload margin of the drone satisfies the requirement for continuous access to the next source user node. If so, calculate the full-segment energy consumption when continuously accessing the next source user node in sequence according to the weighted directed graph, and when the energy consumption when continuously accessing the next source user node is greater than the current energy consumption, use the drone access sequence corresponding to the current energy consumption as the initial drone access sequence.

2. The UAV trajectory planning method based on radio map in wind farm environment according to claim 1 is characterized in that: The initialization parameters of the UAV include: hangar location, source user node set, destination user node set, UAV flight altitude range, UAV flight horizontal range, and UAV flight speed; Correspondingly, the wind vector model is: = + ,in, is the ground speed of the UAV, is the relative air speed of the drone, is the wind speed, , Wind speed The absolute value of Wind speed The angle with the positive horizontal axis.

3. The method for UAV trajectory planning based on radio map in wind farm environment according to claim 1, characterized in that: The movement cost function is: ; in, Among them, E(v,t) represents the instantaneous energy consumption of the UAV, Indicates the airspeed of the drone, Indicates the flight time of the drone, represents the signal-to-noise ratio at the drone’s location, Indicates whether the SNR value at a certain location reaches the threshold that allows drones to pass through. , indicating that the mission allows the drone to pass, otherwise the drone is not allowed to pass.

4. The method for UAV trajectory planning based on radio map in wind farm environment according to claim 1, characterized in that: The genetic algorithm is used to optimize the initial UAV access sequence to obtain the optimal UAV access sequence, including: Generate the initial drone chromosome according to the initial drone visit sequence; The tournament selection method is used to perform crossover, mutation and replication operations on the initial drone chromosomes to obtain offspring individuals; The fitness of each parent individual and offspring individual is iteratively calculated so that the sequence order corresponding to the drone chromosome with the optimal fitness function is used as the optimal drone access order, wherein the fitness function is determined by the path energy consumption of the drone flying in the wind field.

5. A UAV trajectory planning system based on radio maps in a wind farm environment, characterized by: include: An initialization module, used to obtain radio map initialization parameters and UAV initialization parameters to generate radio maps and wind vector models; a weighted directed graph generation module, configured to solve a weighted directed graph according to the radio map and the wind field vector model, wherein the weighted directed graph is used to store the energy consumption of each path; A UAV access sequence optimization module initializes the UAV access sequence based on the weighted directed graph and uses a genetic algorithm to optimize the initial UAV access sequence to obtain an optimal UAV access sequence; The weighted directed graph generation module includes: a weight expression calculation unit, configured to obtain a vertex set and an edge set of a weighted directed graph to obtain a weight expression for a UAV flying between any two points, wherein the weight is determined by a flight path and energy consumption of the UAV corresponding to the flight path; a mobility cost function solving unit, configured to convert the weight solution into a mobility cost function solution; wherein the mobility cost function is determined by drone energy consumption and drone signal strength, wherein the drone energy consumption and drone signal strength are related to flight time, and the flight time is related to a wind field vector model; A minimum energy consumption path calculation unit is used to traverse each source user node set to obtain a solution of a movement cost function between any two points that meets preset conditions to obtain a minimum energy consumption path; A weighted directed graph generating unit, configured to store each of the minimum energy consumption paths to obtain a weighted directed graph; The drone access sequence optimization module specifically includes: A drone user group generating unit is configured to obtain a source user node set and a destination user node set to obtain a plurality of drone user groups, wherein each drone user group includes a source user node and a destination user node; An original UAV access sequence processing unit is used to set an initial strategy to process each UAV user group to obtain an original UAV access sequence, wherein the original UAV access sequence is the source user node and the destination user node in sequence; The initial UAV access sequence calculation unit is used to determine whether the remaining payload margin of the UAV satisfies the requirement for continuous access to the next source user node. If so, the full-segment energy consumption when continuously accessing the next source user node is calculated in sequence according to the weighted directed graph, and when the energy consumption when continuously accessing the next source user node is greater than the current energy consumption, the UAV access sequence corresponding to the current energy consumption is used as the initial UAV access sequence.

6. The UAV trajectory planning system based on radio map in wind farm environment according to claim 5, characterized in that: The drone access sequence optimization module also includes: A chromosome initialization unit, used to generate an initial drone chromosome according to the initial drone access sequence; The offspring individual generation unit is used to perform crossover, mutation and replication operations on the initial drone chromosomes using the tournament selection method to obtain offspring individuals; The optimal drone access sequence optimization unit is used to iteratively calculate the fitness of each parent individual and offspring individual so that the sequence order corresponding to the drone chromosome with the optimal fitness function is used as the optimal drone access sequence, wherein the fitness function is determined by the path energy consumption of the drone flying in the wind field.

Images