A method, apparatus, and system for joint optimization of UAV flight trajectory and user scheduling
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-14
- Publication Date
- 2026-08-14
Smart Images
Figure CN122569501A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of wireless communication technology, and in particular to a method, apparatus and system for joint optimization of UAV flight trajectory and user scheduling. Background Technology
[0002] Drone data collection has revolutionized traditional information acquisition methods, enabling large-scale, high-efficiency surface monitoring from a high-altitude perspective, significantly reducing labor costs and safety risks. Its flexible and mobile nature allows for widespread application in fields such as agricultural plant protection, geological exploration, emergency rescue, and urban planning, providing real-time and reliable spatiotemporal data support for precise decision-making and driving the intelligent and digital transformation of various industries.
[0003] In UAV data acquisition systems, UAV flight trajectory planning and resource management are core components for improving overall efficiency. Scientific trajectory planning, by optimizing flight paths, ensures the integrity and spatial coverage of collected data; resource management encompasses user scheduling and payload allocation, enabling dynamic task adjustments. Their synergistic effect allows UAVs to maintain high-quality communication in complex environments, improving system performance. The UAV flight trajectory planning and resource management problem can be solved using swarm intelligence optimization algorithms. Among them, the Harris Eagle Optimization (HHO) algorithm, with its advantages of fast convergence speed, strong global search capability, and simple parameter settings, can effectively balance the exploration and development process, making it suitable for such complex optimization scenarios.
[0004] However, existing research on using the HHO algorithm to solve UAV trajectory planning and resource management problems is mostly limited to continuous variable optimization, with little attention paid to discrete variable processing. This limits the existing algorithms to the original HHO framework, making them prone to getting stuck in local optima when planning UAV trajectories, and unable to simultaneously consider convergence speed and improve the minimum energy efficiency for all users. Summary of the Invention
[0005] Based on this, the purpose of the present invention is to provide a method, apparatus and system for joint optimization of UAV flight trajectory and user scheduling.
[0006] A method for joint optimization of UAV flight trajectory and user scheduling includes the following steps: S10. Obtain the user scheduling policy for each data source and measure the signal-to-noise ratio of each data source at a reference distance of 1 meter. ; S20. Initialize the UAV flight trajectory and obtain the candidate flight trajectory set Q; S30. Construct a set of eagles based on the candidate flight trajectory set Q and the user scheduling policies of each data source. The set of eagles is calculated by combining the energy efficiency and penalty function of the data source. The fitness value of each eagle is determined, and the eagle with the highest fitness value is selected as the optimal solution. ; S40, Collection of Eagles For each eagle, Harris Eagle optimization and constraint processing are performed to obtain the optimal solution. ; S50, control the drone according to the optimized optimal solution To carry out flight missions.
[0007] The UAV flight trajectory and user scheduling joint optimization method described in this invention constructs a hybrid variable by combining continuous variables representing flight trajectories and discrete variables representing user scheduling strategies, and then performs HHO optimization updates on the hybrid variable. This allows the obtained optimal solution to simultaneously consider the convergence speed of the flight path and the energy efficiency of the user scheduling strategy, significantly improving the minimum energy efficiency among all users while ensuring the convergence rate.
[0008] Further, step S30 includes: S31. Construct a set of eagles based on the candidate flight trajectory set and the user scheduling policy sent by the data source. ; S32. Calculate the energy efficiency of each data source; S33. Calculate the penalty function of the candidate flight trajectory set and the user scheduling policy sent by the data source. ; S34. Calculate the fitness value of each eagle and select the eagle with the highest fitness value as the optimal solution. .
[0009] Further, step S40 includes: S41. Initialize the iteration index ; S42. Initialize the eagle's index. and constraint processing flags And generate a random number representing the escape probability. and a random number representing the prey's initial energy ; S43. Calculate energy E, and based on the calculation results, use the corresponding path to target the eagle. Perform optimization updates to the Harris Eagle to obtain the updated eagle. and update the constraint handling flags. ; S44, Constraint Processing Flags The updated eagle Perform constraint processing; S45. Update the eagle's index. If at this time If the result is positive, return to step S43; otherwise, output the updated set of eagles. ; S46. Calculate the updated set of eagles. All The eagle with the highest fitness is taken as the updated optimal solution. .
[0010] S47, if or Then the optimal solution will be output. Otherwise, update the iterative index. Return to step S42.
[0011] Furthermore, the process of step S43 includes: Calculate energy E using the following formula:
[0012] In the m-th iteration, the... An eagle can be represented as ; like Execute the first path; First path: Perform a global exploration, and calculate the first path according to the following formula. A hawk Update:
[0013] In the formula, Denotes the (m+1)th iteration. A single eagle; This represents the Hadamard product operation; All are related to eagles A random matrix of the same dimension, whose elements are random numbers in the interval [0,1]; LB and UB are the lower and upper bounds of the variable variation; A gathering of eagles The average position of all eagles in the region; A gathering of eagles One eagle is randomly selected from the list; A gathering of eagles The optimal solution in; like and Execute the second path; Second approach: Conduct a soft siege, based on the following formula... A hawk Update:
[0014] In the formula, J is a random number in the interval [0,2]. like and Execute the third path; The third path: Perform a gradual, rapid, soft encirclement, and calculate the second variable Y and the second variable Z according to the following formula;
[0015] In the formula, J is a random number in the interval [0,2].
[0016] In the formula, This represents the Hadamard product operation; It is a random tuple, where each element is randomly distributed in [0,1]. It is a random tuple, and the value of each element is determined by the Levy flight function; The Levy flight function is defined as:
[0017] In the formula, and All obey Gaussian distribution; It is the Gamma function; It is a constant, with a value of 1.5; Apply constraints to the second variable Y and the second variable Z, and update the constraint flags. And according to the following formula, the first A hawk Update:
[0018] In the formula, Indicates constraint handling; The fitness function; like and 0.5, execute the fourth path; Fourth path: Conduct a hard siege, based on the following formula for the... A hawk Update:
[0019] like and 0.5, execute the fifth path; Fifth path: Execute a progressive, rapid dive hard encirclement, and calculate the second variable Y and the second variable Z according to the following formula;
[0020] In the formula, J is a random number in the interval [0,2].
[0021] Apply constraints to the second variable Y and the second variable Z, and update the constraint flags. And according to the following formula, the first A hawk Update:
[0022] In the formula, Indicates constraint handling; This is the fitness function.
[0023] Furthermore, the process of step S44 includes: If constraint processing flag This indicates the updated eagle. Without constraint processing, the following constraint calculations are required: The updated eagle It can be represented as:
[0024] Where m+1 represents the number of iterations; For continuous variables If its coordinates exceed the flight area, it needs to be pulled back to the boundary, i.e., a continuous variable. The following conditions must be met:
[0025] like Then let ;like Then let ; ordinate Similarly; For discrete variables When performing HHO updates, the Sigmoid function is typically used for discrete transformation to obtain...
[0026] in,
[0027] like Then choose The maximum value of is used to associate the corresponding user with 1, that is:
[0028]
[0029] like Then establish a set , and select
[0030] In the formula, Before retrieving a certain set The index of the maximum value; Then there is
[0031] Finally, let Complete the discrete variables Update.
[0032] Further, step S20 includes: S21. Initialize the eagle's index. ; S22. Initialize the time slot index ; S23. Based on the UAV coordinates in the current time slot n Calculate the UAV coordinates corresponding to the next time slot n+1. ; S24. Update the slot index n = n + 1. If n ≥ 1 at this time... If the condition is met, the drone's flight trajectory will be output; otherwise, step S23 will be executed. S25. Update the eagle's index. If at this time Output the candidate flight trajectory set Q; otherwise, execute step S22.
[0033] Further, step S23 includes: S231. Based on the UAV coordinates in time slot n, calculate the maximum distance the UAV can fly with the remaining time. Distance of the drone from the destination and the angle of the drone ; S232, Compare the maximum distance the drone can fly with the remaining time. Distance of the drone from the finish line Based on the comparison results, the corresponding strategy is selected to calculate the flight angle of the UAV from the current time slot n to the next time slot n+1. and flight stride ; S233. Calculate the UAV coordinates corresponding to the next time slot n+1. .
[0034] Furthermore, the process of step S232 includes: Compare and The size, if Execute the first strategy; First strategy: If the drone cannot reach the destination on time, the flight path is not feasible, and the flight path should be abandoned. like Execute the second strategy; Second strategy: Randomly generate flight angles for Randomly generate flight step size ; like Execute the third strategy; Third strategy: Calculate the maximum deflection angle using the following formula. :
[0035] Randomly generated angles ; Solve the equation and take the location The solution of the interval is used as the lower bound of the step size. ; Then the flight angle for ; Randomly generate flight step size ; like Execute the fourth strategy; Fourth strategy: Calculate the maximum deflection angle using the following formula. :
[0036] make ; Randomly generate flight angle ; Randomly generate flight step size .
[0037] The present invention also provides a device for joint optimization of UAV flight trajectory and user scheduling, comprising: The acquisition module is used to obtain the user scheduling policies of each data source and measure the signal-to-noise ratio of each data source at a reference distance of 1 meter. ; The flight trajectory initialization module is used to initialize the UAV's flight trajectory and obtain a candidate flight trajectory set Q; The fitness calculation module is used to construct a set of eagles based on the candidate flight trajectory set Q and the user scheduling policies of each data source. The set of eagles is calculated by combining the energy efficiency and penalty function of the data source. The fitness value of each eagle is determined, and the eagle with the highest fitness value is selected as the optimal solution. ; Optimize and update the module for collecting eagles. For each eagle, Harris Eagle optimization and constraint processing are performed to obtain the optimal solution. ; The execution module is used to control the UAV according to the optimized optimal solution. To carry out flight missions.
[0038] The present invention also provides a joint optimization system for UAV flight trajectory and user scheduling, comprising: a data source, a UAV, and the aforementioned joint optimization device for UAV flight trajectory and user scheduling; The user sends a user scheduling policy to the drone through the data source. The drone flight trajectory and user scheduling joint optimization device obtains the user scheduling policy from the data source, measures the base station received signal-to-noise ratio at a reference distance of 1 meter, and then combines the drone's altitude information and the scheduling information. Based on the HHO framework, it calculates the optimal drone flight trajectory and the optimal user scheduling policy that maximizes the minimum energy efficiency of all users, and controls the drone to perform flight missions accordingly.
[0039] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the module of the UAV flight trajectory and user scheduling joint optimization device provided by the present invention; Figure 2 The flowchart illustrates the execution of the UAV flight trajectory and user scheduling joint optimization method provided by this invention. Figure 3 The execution flowchart of step S20 provided by the present invention; Figure 4 The execution flowchart of step S22 provided by the present invention; Figure 5 The execution flowchart of step S30 provided by the present invention; Figure 6 The execution flowchart of step S40 provided by the present invention; Figure 7 A comparison of the convergence performance of the UAV flight trajectory and user scheduling joint optimization method and the random initialization method provided by this invention; Figure 8 The diagram shows the flight trajectory of the UAV flight trajectory and user scheduling joint optimization method provided by this invention and three other schemes when five data sources are set. Detailed Implementation
[0041] To address the shortcomings of existing UAV flight trajectory planning methods that fail to simultaneously balance convergence speed and minimum energy efficiency for all users, the applicant proposes a joint optimization method for UAV flight trajectory and user scheduling. This method constructs a hybrid variable by combining continuous variables representing flight trajectories and discrete variables representing user scheduling strategies. A novel constraint method is designed for the HHO algorithm optimization update of this hybrid variable, effectively improving search efficiency and constraint satisfaction accuracy. This significantly enhances minimum energy efficiency for all users while maintaining convergence speed. Furthermore, during UAV flight trajectory initialization, multiple strategies are employed to calculate the UAV's flight trajectory, ensuring the feasibility of each trajectory and introducing randomness. This provides a better search starting point for subsequent HHO optimization updates, effectively improving convergence efficiency.
[0042] The following is a detailed description of the UAV flight trajectory and user scheduling joint optimization method of the present invention: This invention provides a joint optimization system for UAV flight trajectory and user scheduling, including a data source, a UAV, and a device for joint optimization of UAV flight trajectory and user scheduling.
[0043] The user sends a user scheduling policy to the drone through the data source. The drone flight trajectory and user scheduling joint optimization device obtains the user scheduling policy from the data source, measures the base station received signal-to-noise ratio at a reference distance of 1 meter, and then combines the drone's altitude information and the scheduling information. Based on the HHO framework, it calculates the optimal drone flight trajectory and the optimal user scheduling policy that maximizes the minimum energy efficiency of all users, and controls the drone to perform flight missions accordingly.
[0044] Please see Figure 1 and Figure 2 The UAV flight trajectory and user scheduling joint optimization device includes a data acquisition module 10, a flight trajectory initialization module 20, a fitness calculation module 30, an optimization update module 40, and an execution module 50.
[0045] In this embodiment, it is assumed that the drone is in a size of The flight mission operates within a rectangular monitoring area, flying at a constant altitude H, collecting environmental data from K data sources, with a total flight time of [missing information]. , The total number of flight time slots, The length of a time slot.
[0046] The flight time of the drone can be expressed as: .
[0047] The acquisition module 10 is used to perform step S10: acquiring the user scheduling policy of each data source and measuring the signal-to-noise ratio of each data source at a reference distance of 1 meter. .
[0048] Specifically, the user scheduling strategy for each data source includes the location information of each data source. and transmission power Additionally, the signal-to-noise ratio (SNR) of each data source at a reference distance of 1 meter can be measured using the pilot signal method. Taking data source k as an example, its signal-to-noise ratio... It can be calculated using the following formula:
[0049] In the formula, The transmitting power of data source k; The channel gain is defined with a reference distance of 1 meter. This represents noise power.
[0050] The flight trajectory initialization module 20 is used to execute step S20: initialize the UAV flight trajectory and obtain a candidate flight trajectory set Q. Specifically, step S20 includes: Specifically, the candidate flight trajectory set Q can be represented as:
[0051] In the formula, For the eagle's index, , The population size for the HHO algorithm is [value].
[0052] Please see Figure 3 Step S20 includes: S21. Initialize the eagle's index. .
[0053] Specifically, this application uses the HHO algorithm for UAV flight trajectory planning. When using it, the eagle's index is first initialized. .
[0054] S22. Initialize the time slot index ; After the Eagle's index initialization is complete, initialize the time slot index. Based on this, the drone flight trajectory for each eagle in terms of time slot is calculated.
[0055] S23. Based on the UAV coordinates in the current time slot n Calculate the UAV coordinates corresponding to the next time slot n+1. ; Please see Figure 4 Step S23 includes: S231. Based on the UAV coordinates in time slot n, calculate the maximum distance the UAV can fly with the remaining time. Distance of the drone from the destination and the angle of the drone ; Specifically, the starting point of the drone's flight is defined as the origin. The drone's flight destination is .
[0056] The maximum distance the drone can fly with the remaining time. for:
[0057] In the formula, This represents the total number of flight time slots. The length of a time slot; This is the maximum flight speed of the drone.
[0058] The distance between the drone and the destination is :
[0059] In the formula, This indicates a search for the norm.
[0060] The current coordinates of the drone can be represented as:
[0061] The drone's deflection angle for:
[0062] S232. Compare the maximum distance the drone can fly with the remaining time. Distance of the drone from the finish line Based on the comparison results, the corresponding strategy is selected to calculate the flight angle of the UAV from the current time slot n to the next time slot n+1. and flight stride .
[0063] Specifically, if Execute the first strategy; First strategy: If the drone cannot reach the destination on time, the flight path is not feasible, and the flight path should be abandoned. like Execute the second strategy; Second strategy: Randomly generate flight angles for Randomly generate flight step size ; like Execute the third strategy; Third strategy: Calculate the maximum deflection angle using the following formula. :
[0064] Randomly generated angles ; Solve the equation and take the location The solution of the interval is used as the lower bound of the step size. ; Then the flight angle for ; Randomly generate flight stride ; like Execute the fourth strategy; Fourth strategy: Calculate the maximum deflection angle using the following formula. :
[0065] make ; Randomly generate flight angle ; Randomly generate flight stride .
[0066] S233. Calculate the UAV coordinates corresponding to the next time slot n+1. .
[0067] Specifically, the UAV coordinates corresponding to the next time slot n+1 It can be represented as:
[0068] Its coordinates are calculated using the following formula:
[0069]
[0070] S24. Update the slot index n = n + 1. If n ≥ 1 at this time... If the condition is met, the drone's flight trajectory will be output; otherwise, step S23 will be executed.
[0071] S25. Update the eagle's index. If at this time Output the candidate flight trajectory set Q; otherwise, execute step S22.
[0072] Specifically, from the index of the eagle Start by calculating the drone's flight trajectory, and then iterate until the conditions are met. ,get The flight paths of the drones are combined to form a candidate flight path set Q.
[0073] The fitness calculation module 30 is used to perform step S30: constructing a set of eagles based on the candidate flight trajectory set Q and the user scheduling strategies of each data source. The set of eagles is calculated by combining the energy efficiency and penalty function of the data source. The fitness value of each eagle is determined, and the eagle with the highest fitness value is selected as the optimal solution. ; Please see Figure 5 Step S30 includes: S31. Construct a set of eagles based on the candidate flight trajectory set and the user scheduling policy sent by the data source. .
[0074] Specifically, the candidate flight trajectory set can be represented as
[0075] The user scheduling strategy can be expressed as:
[0076] in, ;and , indicating the first Only Eagle (the first) In the candidate solutions, the connection status between the data source k in time slot n and the UAV is represented by 0 indicating no connection and 1 indicating connection.
[0077] A gathering of eagles The combination of candidate flight trajectory sets and user scheduling strategies, i.e.
[0078] S32. Calculate the energy efficiency of each data source.
[0079] Taking data source k as an example, for data source k, the channel gain of the UAV in time slot n is...
[0080] In the formula, This represents the channel gain at a reference distance of 1 meter. Indicates the norm; This represents the ground projection coordinates of the UAV in time slot n within the j-th flight trajectory; This indicates the location information of the data source k.
[0081] Within a time slot, the change in the communication rate between data source k and the UAV is approximately linear, and it is calculated according to the following formula:
[0082] in,
[0083] Then, during the drone's flight, the total amount of data collected from data source k is:
[0084] In time slot n, the flight energy consumption of the UAV is...
[0085] In the formula, This indicates the flight power of the drone.
[0086] The energy consumption brought by data source k is
[0087] In the formula, The transmitting power of data source k; This refers to the receiving power loss of the drone.
[0088] Then the energy efficiency of data source k is:
[0089] S33. Calculate the penalty function of the candidate flight trajectory set and the user scheduling policy sent by the data source. .
[0090] Specifically, the penalty function It can be calculated using the following formula:
[0091] in,
[0092]
[0093] In the formula, and Indicates the weighted terms; This indicates the minimum communication rate between the data source and the drone.
[0094] S34. Calculate the fitness value of each eagle and select the eagle with the highest fitness value as the optimal solution. .
[0095] Specifically, the fitness value of each eagle is calculated using a fitness function, i.e.
[0096] in,
[0097] but
[0098] The optimization and update module 40 is used to perform step S40: collecting the eagles. For each eagle, Harris Eagle optimization and constraint processing are performed to obtain the optimal solution. .
[0099] In this embodiment, the collection of eagles Perform m Harris Eagle optimization updates until the maximum number of iterations is reached or the iteration converges. The value of m ranges from 1 to 1. ,in This represents the maximum number of iterations. Through Harris Eagle iterative optimization, the drone flight trajectory and user scheduling strategy are jointly optimized and updated, thereby minimizing the efficiency of all data sources. Please refer to [link / reference]. Figure 6 Step S40 includes: S41. Initialize the iteration index .
[0100] Specifically, during the Harris Hawking update, the iteration index is first initialized. 0.
[0101] S42. Initialize the eagle's index. and constraint processing flags And generate a random number representing the escape probability. and a random number representing the prey's initial energy .
[0102] Specifically, after initializing the iterative index, the eagle's index is then initialized. Initialize constraint processing flags This indicates that no constraint processing has been performed.
[0103] In addition, the random number representing the escape probability and a random number representing the prey's initial energy The following conditions must be met:
[0104]
[0105] S43. Calculate energy E, and based on the calculation results, use the corresponding path to target the eagle. Perform optimization updates to the Harris Eagle to obtain the updated eagle. and update the constraint handling flags. .
[0106] Specifically, the energy E is calculated according to the following formula:
[0107] In the m-th iteration, the... An eagle can be represented as ; like Execute the first path; First path: Perform a global exploration, and calculate the first path according to the following formula. A hawk Update:
[0108] In the formula, Denotes the (m+1)th iteration. A single eagle; This represents the Hadamard product operation; All are related to eagles A random matrix of the same dimension, whose elements are random numbers in the interval [0,1]; LB and UB are the lower and upper bounds of the variable variation; A gathering of eagles The average position of all eagles in the region; A gathering of eagles One eagle is randomly selected from the list; A gathering of eagles The optimal solution in; like and Execute the second path; Second approach: Conduct a soft siege, based on the following formula... A hawk Update:
[0109] In the formula, J is a random number in the interval [0,2]. like and Execute the third path; The third path: Perform a gradual, rapid, soft encirclement, and calculate the second variable Y and the second variable Z according to the following formula;
[0110] In the formula, J is a random number in the interval [0,2].
[0111] In the formula, This represents the Hadamard product operation; It is a random tuple, where each element is randomly distributed in [0,1]. It is a random tuple, and the value of each element is determined by the Levy flight function; The Levy flight function is defined as:
[0112] In the formula, and All obey Gaussian distribution; It is the Gamma function; It is a constant, with a value of 1.5; Apply constraints to the second variable Y and the second variable Z, and update the constraint flags. And according to the following formula, the first A hawk Update:
[0113] In the formula, Indicates constraint handling; The fitness function; like and 0.5, execute the fourth path; Fourth path: Conduct a hard siege, based on the following formula for the... A hawk Update:
[0114] like and 0.5, execute the fifth path; Fifth path: Execute a progressive, rapid dive hard encirclement, and calculate the second variable Y and the second variable Z according to the following formula;
[0115] In the formula, J is a random number in the interval [0,2].
[0116] Apply constraints to the second variable Y and the second variable Z, and update the constraint flags. And according to the following formula, the first A hawk Update:
[0117] In the formula, Indicates constraint handling; This is the fitness function.
[0118] S44, constraint processing flag The updated eagle Perform constraint processing.
[0119] Specifically, if constraint processing flag This indicates the updated eagle. Without constraint processing, the following constraint calculations are required: The updated eagle It can be represented as:
[0120] Where m+1 represents the number of iterations; For continuous variables If its coordinates exceed the flight area, it needs to be pulled back to the boundary, i.e., a continuous variable. The following conditions must be met:
[0121] like Then let ;like Then let ; ordinate Similarly; For discrete variables When performing HHO updates, the Sigmoid function is typically used for discrete transformation to obtain...
[0122] in,
[0123] like Then choose The maximum value of is used to associate the corresponding user with 1, that is:
[0124]
[0125] like Then establish a set , and select
[0126] In the formula, Before retrieving a certain set The index of the maximum value; Then there is
[0127] Finally, let Complete the discrete variables Update.
[0128] Those skilled in the art will understand that step S43 represents the constraint processing of the second variable Y and the second variable Z. The symbols, because the sets Y and Z to be processed are the same as the set in step S44. Since there is a linear relationship, the calculation process is the same as above, and will not be repeated here.
[0129] S45. Update the eagle's index. If at this time If the result is positive, return to step S43; otherwise, output the updated set of eagles. .
[0130] Specifically, the index of eagles Begin with the eagle Perform Harris Eagle optimization updates until the conditions are met. , indicating to The eagle has completed the Harris Eagle optimization update and outputs a set of updated eagles. .
[0131] S46. Calculate the updated set of eagles. All inside The eagle with the highest fitness is taken as the updated optimal solution. .
[0132] Specifically, after performing Harris Eagle optimization updates on all eagles, the fitness of the eagle corresponding to the current iteration is recalculated. The eagle with the highest fitness is then selected as the updated optimal solution. .
[0133] S47, if or Then the optimal solution will be output. Otherwise, update the iterative index. Return to step S42.
[0134] Specifically, the iteration index m=1 begins with Harris Eagle optimization updates on the set of all eagles, and then iterates until either the termination convergence condition or the iteration termination condition is met. The two conditions are as follows:
[0135] In the formula, It is a preset positive integer; For the convergence threshold value This condition indicates that after... After several iterations, the optimal fitness value did not increase significantly, indicating that the iteration has met the convergence condition.
[0136]
[0137] In the formula, This represents the maximum number of iterations. This condition indicates that the Harris Eagle optimization iterations have reached the maximum number, at which point the Harris Eagle optimization process terminates regardless of whether the iteration condition is still met.
[0138] By applying the two conditions mentioned above, the final optimal solution can be obtained. This solution is the one that minimizes the performance of all users.
[0139] The execution module 50 is used to execute step S50: controlling the drone according to the optimized optimal solution. To carry out flight missions.
[0140] Specifically, the optimized optimal solution It can be represented as
[0141] in, This represents the optimal flight path for the drone; This represents the optimal user scheduling strategy. The execution module controls the UAV to perform flight missions according to the optimal UAV flight path and the optimal user scheduling strategy.
[0142] Please see Figure 7 , Figure 7 A comparison of the convergence performance of the UAV flight trajectory and user scheduling joint optimization method and the random initialization method provided by this invention.
[0143] The specific implementation of the random initialization method is as follows: the drone starts from the starting point and flies in a direction roughly towards the destination, with a heading angle of... The system randomly perturbs the drone's trajectory within a given range, with each flight distance randomly generated between 0 and 50 meters. Except for the method of generating the initial drone trajectory, all other processing steps in the random initialization method are consistent with the method of this invention.
[0144] As shown in the figure, the fitness of the method of the present invention converges to 2.98 Mbit / J after approximately 3000 iterations, while the random initialization method can only achieve 2.68 Mbit / J. Therefore, the method of the present invention has superior convergence performance and can achieve higher energy efficiency.
[0145] Please see Figure 8 , Figure 8 The diagram illustrates the flight trajectories of the UAV flight trajectory and user scheduling joint optimization method provided by this invention, along with three other schemes, when five data sources are set. Wherein: Option 1 adopts the nearest neighbor strategy, that is, the drone flies to the nearest data source each time, collects data while flying, until the minimum data volume requirement is met, and immediately turns to the next data source after the collection is completed.
[0146] Option 2 adopts a center-of-gravity convergence strategy, in which the drone first flies to the center of gravity of each data source, and then flies to the destination, randomly selecting 3 data sources for communication in each time slot.
[0147] Scheme 3 adopts a diagonal uniform speed strategy, that is, the UAV flies at a constant speed along the diagonal (shortest path) and randomly selects 3 data sources for communication in each time slot.
[0148] The table below compares the energy efficiency of the method of this invention with three other solutions:
[0149] As shown in the table, the minimum energy efficiency achieved by the method of this invention is much higher than that of the other three schemes, proving that the joint optimization method of UAV flight trajectory and user scheduling provided by this invention can effectively improve the minimum energy efficiency of all users.
[0150] Compared with existing technologies, the joint optimization method for UAV flight trajectory and user scheduling provided in this application has the following advantages: (1) When initializing the UAV flight trajectory, multiple strategies were used to calculate the UAV flight trajectory to ensure the feasibility of each trajectory and reduce the difficulty of subsequent optimization. On this basis, randomness was also introduced to ensure that the obtained UAV flight trajectory set has both a certain degree of diversity and an overall trend toward the destination, thereby providing a better search starting point for subsequent HHO optimization and updates, and effectively improving the convergence efficiency.
[0151] (2) When using the HHO framework for optimization, a mixed variable is first constructed by combining the continuous variable representing the flight trajectory and the discrete variable representing the user scheduling strategy. A new constraint method is designed for the HHO algorithm optimization update of the mixed variable. The discrete variable processing mechanism is introduced into the constraint method so that the HHO algorithm can effectively solve the mixed integer programming problem. The refined constraint processing effectively improves the search efficiency and constraint satisfaction accuracy, thereby significantly improving the minimum energy efficiency among all users while ensuring the convergence rate.
[0152] Based on the same inventive concept described above, this invention also provides an electronic device, which can be a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). This device includes one or more processors and a memory, wherein the processor is used to execute a program to implement the aforementioned method for jointly optimizing the UAV flight trajectory and user scheduling; the memory is used to store computer programs executable by the processor.
[0153] Based on the same inventive concept, the present invention also provides a computer-readable storage medium corresponding to the aforementioned embodiments of the joint optimization method for UAV flight trajectory and user scheduling. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps described in any of the above embodiments.
[0154] This invention can take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to: phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.
[0155] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and the present invention also intends to include these modifications and variations.
Claims
1. A method for joint optimization of UAV flight trajectory and user scheduling, characterized in that, Including the following steps: S10. Obtain the user scheduling policy for each data source and measure the signal-to-noise ratio of each data source at a reference distance of 1 meter. ; S20. Initialize the UAV flight trajectory and obtain the candidate flight trajectory set Q; S30. Construct a set of eagles based on the candidate flight trajectory set Q and the user scheduling policies of each data source. The set of eagles is calculated by combining the energy efficiency and penalty function of the data source. The fitness value of each eagle is determined, and the eagle with the highest fitness value is selected as the optimal solution. ; S40, Collection of Eagles For each eagle, Harris Eagle optimization and constraint processing are performed to obtain the optimal solution. ; S50, control the drone according to the optimized optimal solution To carry out flight missions.
2. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 1, characterized in that, Step S30 includes: S31. Construct a set of eagles based on the candidate flight trajectory set and the user scheduling policy sent by the data source. ; S32. Calculate the energy efficiency of each data source; S33. Calculate the penalty function of the candidate flight trajectory set and the user scheduling policy sent by the data source. ; S34. Calculate the fitness value of each eagle and select the eagle with the highest fitness value as the optimal solution. .
3. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 1, characterized in that, Step S40 includes: S41. Initialize the iteration index ; S42. Initialize the eagle's index. and constraint processing flags And generate a random number representing the escape probability. and a random number representing the prey's initial energy ; S43. Calculate energy E, and based on the calculation results, use the corresponding path to target the eagle. Perform optimization updates to the Harris Eagle to obtain the updated eagle. and update the constraint handling flags. ; S44, Constraint Processing Flags The updated eagle Perform constraint processing; S45. Update the eagle's index. If at this time If the result is positive, return to step S43; otherwise, output the updated set of eagles. ; S46. Calculate the updated set of eagles. All inside The eagle with the highest fitness is taken as the updated optimal solution. ; S47, if or Then the optimal solution will be output. Otherwise, update the iterative index. Return to step S42.
4. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 3, characterized in that, The process of step S43 includes: Calculate energy E using the following formula: In the m-th iteration, the... An eagle can be represented as ; like Execute the first path; First path: Perform a global exploration, and calculate the first path according to the following formula. A hawk Update: In the formula, Denotes the (m+1)th iteration. A single eagle; This represents the Hadamard product operation; All are related to eagles A random matrix of the same dimension, whose elements are random numbers in the interval [0,1]; LB and UB are the lower and upper bounds of the variable variation; A gathering of eagles The average position of all eagles in the region; A gathering of eagles One eagle is randomly selected from the list; A gathering of eagles The optimal solution in; like and Execute the second path; Second approach: Conduct a soft siege, based on the following formula... A hawk Update: In the formula, J is a random number in the interval [0,2]. like and Execute the third path; The third path: Perform a gradual, rapid, soft encirclement, and calculate the second variable Y and the second variable Z according to the following formula; In the formula, J is a random number in the interval [0,2]. In the formula, This represents the Hadamard product operation; It is a random tuple, where each element is randomly distributed in [0,1]. It is a random tuple, and the value of each element is determined by the Levy flight function; The Levy flight function is defined as: In the formula, and All obey Gaussian distribution; It is the Gamma function; It is a constant, with a value of 1.5; Apply constraints to the second variable Y and the second variable Z, and update the constraint flags. And according to the following formula, the first A hawk Update: In the formula, Indicates constraint handling; The fitness function; like and 0.5, execute the fourth path; Fourth path: Conduct a hard siege, based on the following formula for the... A hawk Update: like and 0.5, execute the fifth path; Fifth path: Execute a progressive, rapid dive hard encirclement, and calculate the second variable Y and the second variable Z according to the following formula; In the formula, J is a random number in the interval [0,2]. Apply constraints to the second variable Y and the second variable Z, and update the constraint flags. And according to the following formula, the first A hawk Update: In the formula, Indicates constraint handling; This is the fitness function.
5. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 4, characterized in that, The process of step S44 includes: If constraint processing flag This indicates the updated eagle. Without constraint processing, the following constraint calculations are required: The updated eagle It can be represented as: Where m+1 represents the number of iterations; For continuous variables If its coordinates exceed the flight area, it needs to be pulled back to the boundary, i.e., a continuous variable. The following conditions must be met: like Then let ;like Then let ; ordinate Similarly; For discrete variables When performing HHO updates, the Sigmoid function is typically used for discrete transformation to obtain... in, like Then choose The maximum value of is used to associate the corresponding user with 1, that is: like Then establish a set , and select In the formula, Before retrieving a certain set The index of the maximum value; Then there is Finally, let Complete the discrete variables Update.
6. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 1, characterized in that, Step S20 includes: S21. Initialize the eagle's index. ; S22. Initialize the time slot index ; S23. Based on the UAV coordinates in the current time slot n Calculate the UAV coordinates corresponding to the next time slot n+1. ; S24. Update the slot index n = n + 1. If n ≥ 1 at this time... If the condition is met, the drone's flight trajectory will be output; otherwise, step S23 will be executed. S25. Update the eagle's index. If at this time Output the candidate flight trajectory set Q; otherwise, execute step S22.
7. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 6, characterized in that, Step S23 includes: S231. Based on the UAV coordinates in time slot n, calculate the maximum distance the UAV can fly with the remaining time. Distance of the drone from the destination and the angle of the drone ; S232. Compare the maximum distance the drone can fly with the remaining time. Distance of the drone from the finish line Based on the comparison results, the corresponding strategy is selected to calculate the flight angle of the UAV from the current time slot n to the next time slot n+1. and flight stride ; S233. Calculate the UAV coordinates corresponding to the next time slot n+1. .
8. The method for joint optimization of UAV flight trajectory and user scheduling according to claim 7, characterized in that, The process of step S232 includes: Compare and The size, if Execute the first strategy; First strategy: If the drone cannot reach the destination on time, the flight path is not feasible, and the flight path should be abandoned. like Execute the second strategy; Second strategy: Randomly generate flight angles for Randomly generate flight step size ; like Execute the third strategy; Third strategy: Calculate the maximum deflection angle using the following formula. : Randomly generated angles ; Solve the equation and take the location The solution of the interval is used as the lower bound of the step size. ; Then the flight angle for ; Randomly generate flight stride ; like Execute the fourth strategy; Fourth strategy: Calculate the maximum deflection angle using the following formula. : make ; Randomly generate flight angle ; Randomly generate flight stride .
9. A device for joint optimization of UAV flight trajectory and user scheduling, characterized in that, include: The acquisition module is used to obtain the user scheduling policies of each data source and measure the signal-to-noise ratio of each data source at a reference distance of 1 meter. ; The flight trajectory initialization module is used to initialize the UAV's flight trajectory and obtain a candidate flight trajectory set Q; The fitness calculation module is used to construct a set of eagles based on the candidate flight trajectory set Q and the user scheduling policies of each data source. The set of eagles is calculated by combining the energy efficiency and penalty function of the data source. The fitness value of each eagle is determined, and the eagle with the highest fitness value is selected as the optimal solution. ; Optimize and update the module for collecting eagles. For each eagle, Harris Eagle optimization and constraint processing are performed to obtain the optimal solution. ; The execution module is used to control the UAV according to the optimized optimal solution. To carry out flight missions.
10. A joint optimization system for UAV flight trajectory and user scheduling, characterized in that, include: Data source, drone, and the drone flight trajectory and user scheduling joint optimization device as described in claim 9; The user sends a user scheduling policy to the drone through the data source. The drone flight trajectory and user scheduling joint optimization device obtains the user scheduling policy from the data source, measures the base station received signal-to-noise ratio at a reference distance of 1 meter, and then combines the drone's altitude information and the scheduling information. Based on the HHO framework, it calculates the optimal drone flight trajectory and the optimal user scheduling policy that maximizes the minimum energy efficiency of all users, and controls the drone to perform flight missions accordingly.