A terminal scheduling method and system of a uav swarm assisted data acquisition system

Abstract

The application relates to the technical field of unmanned aerial vehicle scheduling, and discloses a terminal scheduling method and system of an unmanned aerial vehicle group assisted data acquisition system, which comprises the following steps: S1. constructing an unmanned aerial vehicle group assisted data acquisition system model; S2. constructing an optimization problem of terminal scheduling of the unmanned aerial vehicle group assisted data acquisition system; S3. calculating the time cost and energy consumption cost of unmanned aerial vehicle group flight and hovering; S4. considering the energy consumption of unmanned aerial vehicle group flight and hovering through TSPE, and updating the optimization problem; S5. constructing an influence factor of each terminal; according to the influence factor of the terminal, the priority of each terminal to each unmanned aerial vehicle is calculated through a random optimization algorithm; S6. further updating the optimization problem; and S7. solving the further updated optimization problem by iteratively optimizing the unmanned aerial vehicle group. The application solves the problem of mutual interference among unmanned aerial vehicles in the scheduling technology of the unmanned aerial vehicle group assisted data acquisition system, and has the characteristics of simple calculation and actuality.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) scheduling technology, and more specifically, to a terminal scheduling method and system for an UAV swarm-assisted data acquisition system. Background Technology

[0002] Drones, with their high flexibility, low cost, and wide coverage, play a crucial role in wireless communication applications. They can meet the computing and communication resource needs of users in areas without infrastructure or during large-scale disasters. Corresponding application scenarios include aerial base stations, data acquisition / transmission, environmental and traffic monitoring, surveillance search and rescue, and more. On the other hand, with the rapid development of wireless communication technology, the application of the Internet of Things (IoT) in daily life is gradually increasing. However, most IoT devices are significantly constrained in terms of energy, computing, and storage capabilities due to size limitations. When IoT terminals have large amounts of data to process, timely data acquisition becomes crucial. Utilizing drones to assist IoT data acquisition is an effective solution; therefore, researching drone-assisted wireless data acquisition systems has practical significance.

[0003] Many scholars have conducted research on the combined use of wireless communication and drones. For example, drones are used to collect information from a group of mobile ground devices. In this system, total energy consumption is optimized by optimizing the drone trajectory while meeting service requirements. Alternatively, the mission time of drones assisting in data collection tasks is minimized by associating with the Traveling Salesman Problem (TSP). Another approach involves minimizing flight time when drones need to return to a warehouse to recharge when their onboard batteries are depleted, by optimizing flight speed, transmission power, and flight trajectory during this process.

[0004] Existing research often fails to consider the endurance of drones. Due to their limited battery capacity, drones need to make multiple trips to charging stations to replenish their energy. Furthermore, in practical applications, there are often numerous terminals requesting services. Data on these terminals has a lifespan; once it expires, it becomes invalid. Given the large number of requesting terminals and time-sensitive data, the key challenge is how to select which terminals to serve. Existing research is largely inapplicable to this situation. In addition, for multi-drone applications, how to rationally schedule drone swarms to maximize the number of terminals served is also a frequently overlooked issue.

[0005] To address this issue, a multi-terminal, multi-UAV hierarchical scheduling method for assisted edge computing resource allocation is proposed. This method aims to minimize the number of UAVs and optimize UAV trajectories and task offloading strategies. A hierarchical UAV scheduling model is established, comprehensively considering task transmission latency, task offloading decisions, and the minimum energy consumption required to complete tasks. Parameters such as trajectory allocation, flight speed, data transmission rate, and flight path of each UAV node are jointly optimized to reduce the energy consumption of the UAV-assisted edge computing system in completing tasks, thereby obtaining the optimal system resource allocation strategy. The position of each UAV is incorporated into a unit, and the variable length optimization problem is transformed into a two-dimensional optimization problem. The weighting factor for UAV hovering energy consumption is adjusted to reduce the complexity of the optimization problem.

[0006] However, the scheduling technology of existing UAV swarm-assisted data acquisition systems suffers from the problem of mutual interference between UAVs. How to invent a terminal scheduling method for UAV swarm-assisted data acquisition systems that takes into account mutual interference between UAVs is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] To address the problem of mutual interference between drones in the scheduling technology of drone swarm-assisted data acquisition systems, this invention provides a terminal scheduling method and system for drone swarm-assisted data acquisition systems, which is characterized by simple calculation and practicality.

[0008] To achieve the above-mentioned objectives of this invention, the technical solution adopted is as follows: A terminal scheduling method for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system includes the following steps: S1. Construct a model for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system; S2. Assume that the channel between the UAV swarm and each terminal it serves in the UAV swarm-assisted data acquisition system model follows a line-of-sight fading model. Construct an optimization problem for terminal scheduling in the UAV swarm-assisted data acquisition system. S3. Using a flight-hover communication protocol, the mission of the UAV swarm is decomposed into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are calculated. S4. By applying the Traveling Salesman Problem (TSPE), consider the energy consumption of drone swarm flight and hovering, and update the optimization problem; S5. Combining K-means clustering, feature selection, and stochastic optimization, construct the influence factor for each terminal; based on the terminal's influence factor, calculate the priority of each terminal for each UAV using a stochastic optimization algorithm; S6. Optimize the power, speed, and trajectory of the drone swarm, and further update the optimization issues; S7. By iteratively optimizing the UAV swarm, we solve the further updated optimization problem to obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system.

[0009] Preferably, in step S1, constructing the UAV swarm-assisted data acquisition system model specifically involves: Let U represent the number of drones, and K represent the number of terminals requesting services; the coordinates of terminal i are... , The amount of data to be collected is Drones from charging stations Departure, maintaining a constant altitude during data collection. Flying above the terminal; the lifespan of data on the terminal is... Assume the drone u departed from the charging station in total. Next, with variables , , The total time taken by the drone for the nth flight mission is represented by the variable. Indicate the position of drone u in the nth flight mission, where This represents the two-dimensional coordinates of the UAV's projection onto the horizontal plane.

[0010] Furthermore, in step S2, the optimization problem of terminal scheduling in constructing the UAV swarm-assisted data acquisition system involves the following steps: S201. The binary variable indicating whether the drone u serves terminal i during the nth flight is: ,in, This represents the total time taken by drone u on the nth flight mission. The internal system can complete the data acquisition task of terminal i, and transmit the data between the UAV u at time t during the nth flight and terminal i. Represented as: (1) Where B represents the communication bandwidth, This indicates the transmission power between the drone and terminal i. This represents the reference channel power gain at a distance of 1 m. Power indicating channel noise; S202. The amount of data collected by UAV u from terminal i is represented as: (2) in, This represents the time required for drone u to travel to service terminal i on the nth trip; S203. The energy consumption due to communication during the nth flight mission of the UAV. Represented as: (3) S204. Express the power consumption of the UAV as follows:

[0011] in, and These represent the profile power and induced power of the blade in the hovering state, respectively; S205. Define the flight speed of the UAV u at time t as: The total energy consumption of each flight mission Represented as:

[0012] in, This represents the total time taken by drone u during the nth flight mission; S206. Let the variable Indicates the scheduling status of the terminal, variables The variable represents the position of drone u at time t during its nth flight. The variable represents the transmission power of the communication between the UAV u and the terminal. The variable represents the time taken by drone u for its nth mission. express The number of flight missions during the period; the optimization problem of terminal scheduling of the UAV swarm-assisted data acquisition system is expressed as:

[0013]

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] Equation (6b) indicates that the total flight service time of each UAV does not exceed the data lifetime. Equation (6c) indicates that the time U serves terminal i during the nth flight does not exceed the total flight time, and Equation (6d) indicates that the energy consumption of each UAV per flight does not exceed its onboard battery capacity. Equation (6e) indicates that the departure point and return destination of each drone are charging stations; Equation (6f) indicates that each terminal is served by at most one drone; Equation (6g) indicates the binary constraint of the terminal scheduling variable; and Equation (6h) indicates that the transmission power of the drone and the terminal does not exceed the maximum power. .

[0021] Furthermore, in step S3, a flight-hover communication protocol is used to decompose the task of the UAV swarm into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are calculated. The specific steps are as follows: S301. Let... Let be the horizontal coordinates of the drone, representing the hovering position of drone u when it collects data from terminal i; let the initial hovering position of drone u be... In the initial state, the drone collects data directly above the terminal; the transmission rate between the drone and terminal i is expressed as:

[0022] S302. The hovering time of drone u serving terminal i is represented as:

[0023] in, and Let U and U represent the departure and arrival times of U at service terminal i, respectively. This represents the amount of data transmitted by terminal i; S303. The energy consumption during the hovering period consists of the energy consumption of both transmission and hovering, expressed as: (9) in, Indicates the hovering power of the drone; S304. During the flight of the UAV to the hovering point, let the set of sequence pairs representing the order of all terminals be . Sequence pairs This represents the state from terminal i to terminal j; when the start and end points are fixed, the optimal solution for the UAV is to fly at a constant speed; let... This represents the flight speed of the drone from terminal i to terminal j. This represents the flight power of the drone; the time consumed by drone u flying from terminal i to terminal j is expressed as: (10) S305. Flight energy consumption is simplified as follows:

[0024]

[0025]

[0026] (11) in, , and These represent constants related to drones; S306. Divide the time cost of each terminal into hovering virtual time and flight virtual time, and refer to hovering virtual time and flight virtual time as hovering VTC and flight VTC, respectively; wherein, each VTC segment includes the actual duration and the energy consumption charging time of that stage; let The charging power of the drone is represented by the VTC during the hovering phase of communication between drone u and terminal i, as follows: (12) S307. The flight VTC from terminal i to terminal j is represented as:

[0027]

[0028] (13) in, and These represent constants related to drones; S308. Consider the charging pile as terminal 0, and the set of terminals including the charging pile is... Since the charging time has already been calculated in VTC, the hovering time and energy consumption of each drone during the charging phase are set to 0, and the scheduling variable... A value of 1 indicates that the charging station has been scheduled, and the charging station status is set as follows: (14).

[0029] Furthermore, in step S4, the energy consumption of drone swarm flight and hovering is considered through the Traveling Salesman Problem (TSPE), and the optimization problem is updated. The specific steps are as follows: S401. Let the terminal scheduling variable be... This indicates a rechargeable drone. Is it possible to The data acquisition of terminal i is completed within the internal system; let the binary variable be... Indicates the order of access to the terminal, if This indicates that there is a route from terminal i to terminal j during the service process; otherwise... ;set up This indicates changes in energy consumption; this stage of drone operation... The energy consumption changes, where the previous data collection point of terminal j is terminal i; the constraint equation (6c) and equation (6f) on the flight trajectory of UAV u in optimization problem P1 are equivalent to: (15a) (15b) (15c) (15d) Equation (15a) indicates that the number of times the drone departs from the charging station is equal to the number of times it returns to the charging station, ensuring that the drone... The starting and ending points of each sub-trip are charging stations; Equations (15b) and (15c) represent Terminal scheduling variables The constraints; Equation (15d) represents It is a binary variable; S402. Transform the energy consumption constraint (6d) in optimization problem P1 into: using TSPE. (16a) (16b) (16c) Wherein, equation (16a) represents when hour, drones during the time period The energy consumption used by terminal i is equal to the sum of the energy consumption during hovering and flight; Equation (16b) indicates that when the UAV... The battery is fully charged when the drone departs from the charging station; Equation (16c) represents the drone. The energy consumed does not exceed ; S403. Transform the time constraint equation (6b) in optimization problem P1 into equation (17), which represents the sum of VTC during all hovering and flight phases over the data lifetime. Within, that is: (17) S404. Transform optimization problem P1 into optimization problem P2:

[0030] (14) (15a)~Equation (15d), Equation (16a)~Equation (16c), Equation (17)

[0031]

[0032]

[0033] in, (18b) ensures that the terminal scheduling variable is a binary variable, and equations (18c) and (18d) ensure that the UAV... The transmission power and flight speed are within the maximum constraints. Indicates drone Maximum flight speed.

[0034] Furthermore, in step S5, an influence factor for each terminal is constructed by combining K-means clustering, feature selection, and stochastic optimization. Based on the influence factor of the terminal, the priority of each terminal to each UAV is calculated using a stochastic optimization algorithm. The specific steps are as follows: S501. Assume the amount of data to be transmitted by the terminal. Feature one: the distance between the terminal and the charging station Feature 2; Feature 3; Feature 4; Feature 5; Feature 6; Feature 7; Feature 8; Feature 9; Feature 1; Feature 2; Feature 2; Feature 3; Feature 4; Feature 5; Feature 6; Feature 7; Feature 8; Feature 9; Feature 1; Feature 2; Feature 3; Feature 4; Feature 5; Feature 6 ...4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6 S502. Normalize the above 6 features for each terminal to construct the terminal's influence factor; S503. Calculate the priority of each terminal for each drone using a random selection algorithm.

[0035] Furthermore, the random selection algorithm described in step S503 is specifically as follows: S5101. Based on six features, six influencing factors were obtained after normalization. ; S5102. Randomly generate N sets of influence factor weight reorganizations Calculate the maximum number of terminals served by different weighted reconfigurations under the current scenario conditions, and record the current maximum number of services. The optimal corresponding weight reorganization ; S5103. If optimal weight reorganization If the weight values ​​of the next N groups are updated, repeat step 5102; if the optimal weight reorganization... The maximum number of services among the weight values ​​of the next N groups No longer changing, let For this scenario weight value Execute step 5104; S5104. Solve for the weighted sum of the influence factors in step 5101, and set it as the priority of the terminal. .

[0036] Furthermore, in step S6, the power, speed, and trajectory of the drone swarm are optimized, and the optimization problem is further updated, specifically as follows: The UAV problem is simplified into multiple fixed... Optimization problem P3:

[0037] st Equations (14), (15a) to (15d), (16a) to (16c), (18c) to (18d).

[0038] Furthermore, in step S7, by iteratively optimizing the UAV swarm, a further updated optimization problem is solved to obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system. The specific steps are as follows: S701. Let... Indicates whether the drones are fully scheduled. This indicates that the shortest service time for drones did not exceed [time limit]. , Indicates the lifespan of data on the terminal; divides variables into... and The two groups were preprocessed separately to obtain the initial flight speed of the drone swarm. Transmission power and initial drone hovering position , Let i be the coordinates of terminal i; S702. Assume that the initial values ​​of the slack variables for the transmission rate are all... H represents the constant altitude at which the UAV flies above the terminal during data acquisition. ,set up ; S703. If The random selection algorithm described in step S5 is used to determine the current scheduling of the terminal on the UAV. The priority is determined by the terminal priority; otherwise, the drone will add a service terminal. ; S704. Solve the optimization problem P3 according to equation (19), and calculate the total time required for this flight mission. ; S705. Let the number of iterations be... threshold The terminal flight velocity was solved using the block coordinate descent method. Flight trajectory Transmission power ; S706. Order And obtain drones Total time ; S707. Save the shortest service time. Terminal flight speed Flight trajectory Transmission power ; S708. Repeat steps S705 to S707 until... ; S709. ,if Save the current drone scheduling plan; otherwise... ; S710. If Repeat steps S703 to S709; otherwise, end the loop and obtain the scheduling scheme for all drones.

[0039] A terminal scheduling system for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system includes a model building module, an initialization optimization problem module, a flight hovering cost module, a TPSE module, a random selection module, an optimization update module, and a solution output module. The model building module is used to build a model of an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system. The initialization optimization problem module is used to construct an optimization problem for terminal scheduling of the UAV swarm assisted data acquisition system model, assuming that the channel between the UAV swarm and each terminal it serves follows a line-of-sight fading model. The aforementioned flight hovering cost module is used to decompose the mission of the UAV swarm into two stages, flight and hovering, using a flight-hovering communication protocol, and to calculate the time and energy costs of the UAV swarm's flight and hovering. The TPSE module is used to consider the energy consumption of drone swarm flight and hovering through the Traveling Salesman Problem (TSPE) and to update and optimize the problem. The random selection module is used to combine K-means clustering, feature selection and random optimization to construct the influence factor of each terminal; based on the influence factor of the terminal, the priority of each terminal to each UAV is calculated by the random selection algorithm. The optimization and update module is used to optimize the power, speed, and trajectory of the drone swarm, and further update and optimize the problem; The solution output module is used to iteratively optimize the UAV swarm, solve the further updated optimization problem, and obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system.

[0040] The beneficial effects of this invention are as follows: This invention discloses a terminal scheduling method for a drone swarm-assisted data acquisition system. It considers a more practical solution for drone swarms with latency requirements, proposing a random optimization algorithm to significantly reduce computational load while minimizing interference between drones, thus better suited for practical applications. This invention employs a multi-pass optimization approach for the drone swarm, combining K-means clustering and feature selection to construct an influence factor that measures the impact of selected terminals on unselected terminals. Through random optimization, the scheduling of other drones is fixed during the optimization process, and the terminal with the highest priority among the currently optimized drones is determined for scheduling. This ensures that terminals served by different drones are as far apart as possible to reduce mutual interference, maximizing the total number of terminals served by the drone swarm through rational terminal scheduling. Simultaneously, it skips redundant time-cost calculations and optimizations, solving the problem of mutual interference between drones in drone swarm-assisted data acquisition system scheduling technology. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the terminal scheduling method of a UAV-assisted data acquisition method according to the present invention. Figure 2 This is a model example diagram of an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system according to the present invention; Figure 3 This is an experimental dual-UAV trajectory comparison diagram of an UAV-assisted data acquisition method according to the present invention; Figure 4 This is a comparison diagram of experimental algorithms for a UAV-assisted data acquisition method according to the present invention. Detailed Implementation

[0042] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0043] Example 1 like Figure 1 As shown, a terminal scheduling method for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system includes the following steps: S1. Construct a model for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system; S2. Assume that the channel between the UAV swarm and each terminal it serves in the UAV swarm-assisted data acquisition system model follows a line-of-sight fading model. Construct an optimization problem for terminal scheduling in the UAV swarm-assisted data acquisition system. S3. Using a flight-hover communication protocol, the mission of the UAV swarm is decomposed into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are calculated. S4. By applying the Traveling Salesman Problem (TSPE), consider the energy consumption of drone swarm flight and hovering, and update the optimization problem; S5. Combining K-means clustering, feature selection, and stochastic optimization, construct the influence factor for each terminal; based on the terminal's influence factor, calculate the priority of each terminal for each UAV using a stochastic optimization algorithm; S6. Optimize the power, speed, and trajectory of the drone swarm, and further update the optimization issues; S7. By iteratively optimizing the UAV swarm, we solve the further updated optimization problem to obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system.

[0044] Example 2 like Figure 1 As shown, a terminal scheduling method for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system includes the following steps: S1. Construct a model for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system; S2. Assume that the channel between the UAV swarm and each terminal it serves in the UAV swarm-assisted data acquisition system model follows a line-of-sight fading model. Construct an optimization problem for terminal scheduling in the UAV swarm-assisted data acquisition system. S3. Using a flight-hover communication protocol, the mission of the UAV swarm is decomposed into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are calculated. S4. By applying the Traveling Salesman Problem (TSPE), consider the energy consumption of drone swarm flight and hovering, and update the optimization problem; S5. Combining K-means clustering, feature selection, and stochastic optimization, construct the influence factor for each terminal; based on the terminal's influence factor, calculate the priority of each terminal for each UAV using a stochastic optimization algorithm; S6. Optimize the power, speed, and trajectory of the drone swarm, and further update the optimization issues; S7. By iteratively optimizing the UAV swarm, we solve the further updated optimization problem to obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system.

[0045] like Figure 2As shown, in a specific embodiment, step S1, constructing the UAV swarm-assisted data acquisition system model, specifically involves: Let U represent the number of drones, and K represent the number of terminals requesting services; the coordinates of terminal i are... , The amount of data to be collected is Drones from charging stations Departure, maintaining a constant altitude during data collection. Flying above the terminal; the lifespan of data on the terminal is... Assume the drone u departed from the charging station in total. Next, with variables , , The total time taken by the drone for the nth flight mission is represented by the variable. Indicate the position of drone u in the nth flight mission, where This represents the two-dimensional coordinates of the UAV's projection onto the horizontal plane.

[0046] In one specific embodiment, step S2, which addresses the optimization problem of terminal scheduling in a UAV swarm-assisted data acquisition system, includes the following steps: S201. The binary variable indicating whether the drone u serves terminal i during the nth flight is: ,in, This represents the total time taken by drone u on the nth flight mission. The internal system can complete the data acquisition task of terminal i, and transmit the data between the UAV u at time t during the nth flight and terminal i. Represented as: (1) Where B represents the communication bandwidth, This indicates the transmission power between the drone and terminal i. This represents the reference channel power gain at a distance of 1 m. Power indicating channel noise; S202. The amount of data collected by UAV u from terminal i is represented as: (2) in, This represents the time required for drone u to travel to service terminal i on the nth trip; S203. The energy consumption due to communication during the nth flight mission of the UAV. Represented as: (3) S204. Express the power consumption of the UAV as follows:

[0047] in, and These represent the profile power and induced power of the blade in the hovering state, respectively; S205. Define the flight speed of the UAV u at time t as: The total energy consumption of each flight mission Represented as:

[0048] in, This represents the total time taken by drone u during the nth flight mission; S206. Let the variable Indicates the scheduling status of the terminal, variables The variable represents the position of drone u at time t during its nth flight. The variable represents the transmission power of the communication between the UAV u and the terminal. The variable represents the time taken by the drone u for its nth mission. express The number of flight missions during the period; the optimization problem of terminal scheduling of the UAV swarm-assisted data acquisition system is expressed as:

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] Equation (6b) indicates that the total flight service time of each UAV does not exceed the data lifetime. Equation (6c) indicates that the time U serves terminal i during the nth flight does not exceed the total flight time, and Equation (6d) indicates that the energy consumption of each UAV per flight does not exceed its onboard battery capacity. Equation (6e) indicates that the departure point and return destination of each drone are charging stations; Equation (6f) indicates that each terminal is served by at most one drone; Equation (6g) indicates the binary constraint of the terminal scheduling variable; and Equation (6h) indicates that the transmission power of the drone and the terminal does not exceed the maximum power. .

[0057] In one specific embodiment, step S3 involves using a flight-hover communication protocol to decompose the task of the UAV swarm into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are then calculated. The specific steps are as follows: S301. Let... Let be the horizontal coordinates of the drone, representing the hovering position of drone u when it collects data from terminal i; let the initial hovering position of drone u be... In the initial state, the drone collects data directly above the terminal; the transmission rate between the drone and terminal i is expressed as:

[0058] S302. The hovering time of drone u serving terminal i is represented as:

[0059] in, and Let U and U represent the departure and arrival times of U at service terminal i, respectively. This represents the amount of data transmitted by terminal i; S303. The energy consumption during the hovering period consists of the energy consumption of both transmission and hovering, expressed as: (9) in, Indicates the hovering power of the drone; S304. During the flight of the UAV to the hovering point, let the set of sequence pairs representing the order of all terminals be . Sequence pairs This represents the state from terminal i to terminal j; when the start and end points are fixed, the optimal solution for the UAV is to fly at a constant speed; let... This represents the flight speed of the drone from terminal i to terminal j. This represents the flight power of the drone; the time consumed by drone u flying from terminal i to terminal j is expressed as: (10) S305. Flight energy consumption is simplified as follows:

[0060]

[0061]

[0062] (11) in, , and These represent constants related to drones; S306. Divide the time cost of each terminal into hovering virtual time and flight virtual time, and refer to hovering virtual time and flight virtual time as hovering VTC and flight VTC, respectively; wherein, each VTC segment includes the actual duration and the energy consumption charging time of that stage; let The charging power of the drone is represented by the VTC during the hovering phase of communication between drone u and terminal i, as follows: (12) S307. The flight VTC from terminal i to terminal j is represented as:

[0063]

[0064] (13) in, and These represent constants related to drones; S308. Consider the charging pile as terminal 0, and the set of terminals including the charging pile is... Since the charging time has already been calculated in VTC, the hovering time and energy consumption of each drone during the charging phase are set to 0, and the scheduling variable... A value of 1 indicates that the charging station has been scheduled, and the charging station status is set as follows: (14)

[0065] In one specific embodiment, in step S4, the energy consumption of drone swarm flight and hovering is considered through the Traveling Salesman Problem (TSPE), and the optimization problem is updated. The specific steps are as follows: S401. Let the terminal scheduling variable be... This indicates a rechargeable drone. Is it possible to The data acquisition of terminal i is completed within the internal system; let the binary variable be... Indicates the order of access to the terminal, if This indicates that there is a route from terminal i to terminal j during the service process; otherwise... ;set up This indicates changes in energy consumption; this stage of drone operation... The energy consumption changes, where the previous data collection point of terminal j is terminal i; the constraint equation (6c) and equation (6f) on the flight trajectory of UAV u in optimization problem P1 are equivalent to: (15a) (15b) (15c) (15d) Equation (15a) indicates that the number of times the drone departs from the charging station is equal to the number of times it returns to the charging station, ensuring that the drone... The starting and ending points of each sub-trip are charging stations; Equations (15b) and (15c) represent Terminal scheduling variables The constraints; Equation (15d) represents It is a binary variable; S402. Transform the energy consumption constraint (6d) in optimization problem P1 into: using TSPE. (16a) (16b) (16c) Wherein, equation (16a) represents when hour, drones during the time period The energy consumption used by terminal i is equal to the sum of the energy consumption during hovering and flight; Equation (16b) indicates that when the UAV... The battery is fully charged when the drone departs from the charging station; Equation (16c) represents the drone. The energy consumed does not exceed ; S403. Transform the time constraint equation (6b) in optimization problem P1 into equation (17), which represents the sum of VTC during all hovering and flight phases over the data lifetime. Within, that is: (17) S404. Transform optimization problem P1 into optimization problem P2:

[0066] (14) (15a)~Equation (15d), Equation (16a)~Equation (16c), Equation (17)

[0067]

[0068]

[0069] in, (18b) ensures that the terminal scheduling variable is a binary variable, and equations (18c) and (18d) ensure that the UAV... The transmission power and flight speed are within the maximum constraints. Indicates drone Maximum flight speed.

[0070] In one specific embodiment, step S5 involves combining K-means clustering, feature selection, and stochastic optimization to construct an influence factor for each terminal. Based on the terminal's influence factor, a stochastic optimization algorithm is used to calculate the priority of each terminal relative to each drone. The specific steps are as follows: S501. Assume the amount of data to be transmitted by the terminal. Feature one: the distance between the terminal and the charging station Feature 2; Feature 3; Feature 4; Feature 5; Feature 6; Feature 7; Feature 8; Feature 9; Feature 1; Feature 2; Feature 2; Feature 3; Feature 4; Feature 5; Feature 6; Feature 7; Feature 8; Feature 9; Feature 1; Feature 2; Feature 3; Feature 4; Feature 5; Feature 6 ...4; Feature 5; Feature 6; Feature 2; Feature 3; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6; Feature 4; Feature 5; Feature 6 S502. Normalize the above 6 features for each terminal to construct the terminal's influence factor; S503. Calculate the priority of each terminal for each drone using a random selection algorithm.

[0071] In one specific embodiment, the random selection algorithm described in step S503 is as follows: S5101. Based on six features, six influencing factors were obtained after normalization. ; S5102. Randomly generate N sets of influence factor weight reorganizations Calculate the maximum number of terminals served by different weighted reconfigurations under the current scenario conditions, and record the current maximum number of services. The optimal corresponding weight reorganization ; S5103. If optimal weight reorganization If the weight values ​​of the next N groups are updated, repeat step 5102; if the optimal weight reorganization... The maximum number of services among the weight values ​​of the next N groups No longer changing, let For this scenario weight value Execute step 5104; S5104. Solve for the weighted sum of the influence factors in step 5101, and set it as the priority of the terminal. .

[0072] In one specific embodiment, step S6 involves optimizing the power, speed, and trajectory of the drone swarm, further updating the optimization problem, specifically as follows: The UAV problem is simplified into multiple fixed... Optimization problem P3:

[0073] st Equations (14), (15a) to (15d), (16a) to (16c), (18c) to (18d).

[0074] In one specific embodiment, in step S7, by iteratively optimizing the UAV swarm, a further updated optimization problem is solved to obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system. The specific steps are as follows: S701. Let... Indicates whether the drones are fully scheduled. This indicates that the shortest service time for drones did not exceed [time limit]. , Indicates the lifespan of data on the terminal; divides variables into... and The two groups were preprocessed separately to obtain the initial flight speed of the drone swarm. Transmission power and initial drone hovering position , Let i be the coordinates of terminal i; S702. Assume that the initial values ​​of the slack variables for the transmission rate are all... H represents the constant altitude at which the UAV flies above the terminal during data acquisition. ,set up ; S703. If The random selection algorithm described in step S5 is used to determine the current scheduling of the terminal on the UAV. The priority is determined by the terminal priority; otherwise, the drone will add a service terminal. ; S704. Solve the optimization problem P3 according to equation (19), and calculate the total time required for this flight mission. ; S705. Let the number of iterations be... threshold The terminal flight velocity was solved using the block coordinate descent method. Flight trajectory Transmission power ; S706. Order And obtain drones Total time ; S707. Save the shortest service time. Terminal flight speed Flight trajectory Transmission power ; S708. Repeat steps S705 to S707 until... ; S709. ,if Save the current drone scheduling plan; otherwise... ; S710. If Repeat steps S703 to S709; otherwise, end the loop and obtain the scheduling scheme for all drones.

[0075] like Figure 3 As shown, the two trajectories represent the flight paths of two drones. It can be seen that the algorithm selects relatively densely distributed terminals for the drones. The fact that closer terminals are discarded from the trajectories indicates that the algorithm comprehensively considers distance and data transmission volume on each terminal when calculating terminal priority. Furthermore, the service ranges of the two drones form an approximately obtuse angle, demonstrating that the algorithm can reduce communication interference between drones by keeping them away from the service range of other drones.

[0076] like Figure 4 As shown, most optimization algorithms for UAV swarms fall into two main categories: one simplifies the problem into a solvable one for optimization, which often generates a large amount of computation; the other optimizes each UAV individually to obtain the optimization result. By comparing the approach of optimizing each UAV individually with a distance-based service terminal selection method, the following three comparison schemes are derived: Opt: Random selection + iterative optimization of UAV terminal scheduling (the algorithm in this paper); Benchmark 1: Random selection + individual drone-based terminal scheduling optimization; Benchmark2: Terminal selection based on distance + optimization of terminal scheduling for individual drones; Benchmark 3: Distance-based terminal selection service + iterative optimization of drone terminal scheduling.

[0077] Depend on Figure 4 As can be seen, the algorithm presented in this paper has significant advantages in scenarios involving a single drone, two drones, or three drones, with even the worst performance result being on par with the best result of the comparative scheme. Furthermore, the advantages of the algorithm become increasingly apparent as the data lifetime increases.

[0078] In this embodiment, in order to enable the UAV swarm to serve as many terminals as possible among numerous requesting service terminals containing a large amount of data to be transmitted, this patent studies a wireless communication framework: the framework assigns priority to terminals based on random optimization and feature selection, so that the algorithm can skip a large amount of computation to obtain a terminal scheduling scheme under the dual constraints of data lifetime and battery capacity.

[0079] In this embodiment, compared to the individual drone optimization scheme in the traditional multi-drone optimization algorithm, this patent, based on the block coordinate descent method, first fixes the scheduling scheme of the remaining drones, then updates the priority of the unscheduled terminal for the currently optimized drone, then adds a service terminal to the drone, and repeats the above iterative process until all drones are at full capacity and can no longer serve more terminals. This optimization scheme can make maximum use of feature six in the random selection algorithm, making the service terminal range of a single drone more concentrated, far away from the service range between drones, and minimizing interference between drones, which is more in line with the actual application of drone swarms in wireless communication scenarios.

[0080] In this embodiment, the UAV swarm-assisted data acquisition system model proposed in this invention is applicable to UAV swarms and data acquisition systems with lifecycle characteristics. It can schedule terminals when the amount of data to be transmitted is large and the number of requesting service terminals is large, thus making the model more widely applicable.

[0081] This invention discloses a terminal scheduling method for a drone swarm-assisted data acquisition system. It considers a more practical solution for drone swarms with latency requirements, proposing a random optimization algorithm to significantly reduce computational load while minimizing interference between drones, thus better suited for practical applications. This invention employs a multi-pass optimization approach for the drone swarm, combining K-means clustering, feature selection, and random optimization. During optimization, the scheduling of other drones is fixed, and the terminal with the highest priority among the currently optimized drones is selected for scheduling. This ensures that terminals served by different drones are as far apart as possible to reduce mutual interference, rationally scheduling terminals to maximize the total number of terminals served by the drone swarm, while skipping redundant time-cost calculations and optimizations. This solves the problem of mutual interference between drones in drone swarm-assisted data acquisition system scheduling technology.

[0082] Example 3 A terminal scheduling system for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system includes a model building module, an initialization optimization problem module, a flight hovering cost module, a TPSE module, a random selection module, an optimization update module, and a solution output module. The model building module is used to build a model of an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system. The initialization optimization problem module is used to construct an optimization problem for terminal scheduling of the UAV swarm assisted data acquisition system model, assuming that the channel between the UAV swarm and each terminal it serves follows a line-of-sight fading model. The aforementioned flight hovering cost module is used to decompose the mission of the UAV swarm into two stages, flight and hovering, using a flight-hovering communication protocol, and to calculate the time and energy costs of the UAV swarm's flight and hovering. The TPSE module is used to consider the energy consumption of drone swarm flight and hovering through the Traveling Salesman Problem (TSPE) and to update and optimize the problem. The random selection module is used to combine K-means clustering, feature selection, and random optimization to construct the influence factor of each terminal; based on the influence factor of the terminal, the priority of each terminal to each UAV is calculated by the random selection algorithm. The optimization and update module is used to optimize the power, speed, and trajectory of the drone swarm, and further update and optimize the problem; The solution output module is used to iteratively optimize the UAV swarm, solve the further updated optimization problem, and obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system.

[0083] Compared to previous research, this patent considers a more realistic and complex system model scenario. Unlike studies that propose multi-objective optimization or simply minimize task time, this patent focuses on how to rationally schedule multiple rechargeable drones in situations with a large number of service requesting terminals and data with a limited lifespan. This allows the drone swarm to dynamically and rationally select and adjust its position to serve as many requesting terminals as possible within a limited timeframe, while minimizing mutual interference. To address this, this patent proposes a framework for rapidly selecting and scheduling terminals in multi-drone scenarios, addressing the issue of how drones can collaboratively complete data collection tasks. This framework introduces influence factors based on feature selection to replace computational time costs. A stochastic optimization algorithm is designed to obtain influence factor weights, and the terminal priority is determined by weighted summation of these influence factors. This enables the drone swarm to schedule a large number of requesting terminals, and then a scheduling scheme is calculated based on these priorities. This scheduling method achieves high-quality scheduling schemes while maintaining low computational complexity. Furthermore, by measuring the service range of different drones through influence factors, it minimizes interference between drones, thereby maximizing the total number of serviceable terminals and scheduling them more rationally.

[0084] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A terminal scheduling method for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system, characterized in that: Includes the following steps: S1. Construct a model for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system; S2. Assume that the channel between the UAV swarm and each terminal it serves in the UAV swarm-assisted data acquisition system model follows a line-of-sight fading model. Construct an optimization problem for terminal scheduling in the UAV swarm-assisted data acquisition system. S3. Using a flight-hover communication protocol, the mission of the UAV swarm is decomposed into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are calculated. S4. By applying the Traveling Salesman Problem (TSPE), consider the energy consumption of drone swarm flight and hovering, and update the optimization problem; S5. Combining K-means clustering, feature selection, and stochastic optimization, construct the influence factor for each terminal; based on the terminal's influence factor, calculate the priority of each terminal for each UAV using a stochastic optimization algorithm; S6. Optimize the power, speed, and trajectory of the drone swarm, and further update the optimization problem; specifically: optimize the drone... u The problem is simplified into multiple fixed Optimization problem P3: ； in, Indicates whether the drone u can be in T max The data acquisition of terminal i is completed within the internal system; Indicates changes in energy consumption; Indicates drone u With terminal i Virtual timeout (VTC) during the hovering phase in communication; Indicates from the terminal i Flying to the Terminal j Virtual flight time (VTC); S7. By iteratively optimizing the UAV swarm, solve the further updated optimization problem to obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system; the specific steps are as follows: S701. Let... Indicates whether the drones are fully scheduled. This indicates that the shortest service time for drones did not exceed [time limit]. , Indicates the lifespan of data on the terminal; Divide the variables and The two groups were preprocessed separately to obtain the initial flight speed of the drone swarm. Transmission power and initial drone hovering position , Let i be the coordinates of terminal i; S702. Assume that the initial values ​​of the slack variables for the transmission rate are all... H represents the constant altitude at which the UAV flies above the terminal during data acquisition. ,set up ; S703. If The random selection algorithm described in step S5 is used to determine the current scheduling of the terminal on the UAV. The priority is determined by the terminal priority; otherwise, the drone will add a service terminal. ; S704. Solve the optimization problem P3 according to equation (19) and calculate the total time required for the current flight mission. ; S705. Let the number of iterations be... threshold The terminal flight velocity was solved using the block coordinate descent method. Flight trajectory Transmission power ; S706. Order And obtain drones Total time ; S707. Minimum-capacity drone. Total time Terminal flight speed Flight trajectory Transmission power ; S708. Repeat steps S705 to S707 until... ; S709. ,if Save the current drone scheduling plan; otherwise... ; S710. If Repeat steps S703 to S709; otherwise, end the loop and obtain the scheduling scheme for all drones.

2. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 1, characterized in that: In step S1, the model for the UAV swarm-assisted data acquisition system is constructed, specifically as follows: set up U K represents the number of drones, and K represents the number of terminals requesting services; terminals i The coordinates are , The amount of data to be collected is Drones from charging stations Departure, maintaining a constant altitude during data collection. Flying above the terminal; the lifespan of data on the terminal is... ; Set up drones u We started from the charging station. Next, with variables Indicates the first n The total time taken by the drone for a flight mission , , with variables Indicates drone u In the n Location during the flight mission ,in Indicates drone u Two-dimensional coordinates projected onto the horizontal plane.

3. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 2, characterized in that: In step S2, the optimization problem of terminal scheduling in the UAV swarm-assisted data acquisition system is constructed, and the specific steps are as follows: S201. Unmanned Aerial Vehicle (UAV) u In the n Is it the terminal during the flight? i The binary variable of the service is ,in, This represents the total time taken by drone u on the nth flight mission. Internal terminal can complete i The task of data collection will involve drones u In the n A flight t Time and Terminal i transmission rate Represented as: (1) in, B Indicates communication bandwidth. Indicates drones and terminals i Transmission power, This represents the reference channel power gain at a distance of 1 m. Power indicating channel noise; S202. The drone u From the terminal i The amount of data collected is expressed as follows: (2) in, Indicates that the drone u is in the n Service Terminal i Time required; S203. The drone u No. n Energy consumption due to communication during a flight mission Represented as: (3) S204. Express the power consumption of the UAV as follows: in, and These represent the profile power and induced power of the blade in the hovering state, respectively; S205. Definition of Unmanned Aerial Vehicles u exist t The flight speed at that moment is The total energy consumption of each flight mission Represented as: in, This represents the total time taken by drone u during the nth flight mission; S206. Let the variable Indicates the scheduling status of the terminal, variables The variable represents the position of drone u at time t during its nth flight. The variable represents the transmission power of the communication between the UAV u and the terminal. The variable represents the time taken by drone u for its nth mission. express The number of flight missions during the period; the optimization problem of terminal scheduling of the UAV swarm-assisted data acquisition system is expressed as: Equation (6b) indicates that the total flight service time of each UAV does not exceed the data lifetime. Equation (6c) represents the unmanned aerial vehicle (UAV). u In the n During the flight, the terminal i The service duration shall not exceed the total flight duration, and formula (6d) indicates that the energy consumption of each drone per flight shall not exceed its onboard battery capacity. Equation (6e) indicates that the departure point and return destination of each drone are charging stations; Equation (6f) indicates that each terminal is served by at most one drone; Equation (6g) indicates the binary constraint of the terminal scheduling variable; and Equation (6h) indicates that the transmission power of the drone and the terminal does not exceed the maximum power. .

4. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 3, characterized in that: In step S3, a flight-hover communication protocol is used to decompose the task of the UAV swarm into two phases: flight and hovering. The time and energy costs of the UAV swarm's flight and hovering are calculated. The specific steps are as follows: S301. Let... The horizontal coordinates of the drone represent the drone's position. u Give to the terminal i Hover position during data acquisition; set drones u The initial hovering position is In the initial state, the drone collects data directly above the terminal; the drone and the terminal are then positioned... i The transmission rate between them is expressed as: S302. The drone u For the terminal i The service hover time is represented as follows: in, and They represent drones u At the service terminal i Departure time and arrival time, Indicates terminal i The amount of data transmitted; S303. The energy consumption during the hovering period consists of the energy consumption of both transmission and hovering, expressed as: (9) in, Indicates the hovering power of the drone; S304. During the flight of the UAV to the hovering point, let the set of sequence pairs representing the order of all terminals be . Sequence pairs Indicates from the terminal i To the terminal j The optimal state is when the start and end points are fixed, and the drone flies at a constant speed. Indicates that the drone is from the terminal i To the terminal j Flight speed, Indicates the flight power of the drone; [This refers to the drone's] u From the terminal i Flying to the Terminal j The time consumption is expressed as: (10) S305. Flight energy consumption is simplified as follows: (11) in, , and These represent constants related to drones; S306. Divide the time cost of each terminal into hovering virtual time and flight virtual time, and refer to hovering virtual time and flight virtual time as hovering VTC and flight VTC, respectively; wherein, each VTC segment includes the actual duration and the energy consumption charging time of that stage; let This indicates the charging power for the drone. u With terminal i VTC during the hovering phase of communication is represented as: (12) S307. From the terminal i Flying to the Terminal j Flight VTC is represented as: (13) in, and These represent constants related to drones; S308. Consider the charging pile as terminal 0, and the set of terminals including the charging pile is... Since the charging time has already been calculated in VTC, the hovering time and energy consumption of each drone during the charging phase are set to 0, and the scheduling variable... A value of 1 indicates that the charging station has been scheduled, and the charging station status is set as follows: (14) 5. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 4, characterized in that: In step S4, the energy consumption of drone swarm flight and hovering is considered through the Traveling Salesman Problem (TSPE), and the optimization problem is updated. The specific steps are as follows: S401. Let the terminal scheduling variable be... This indicates a rechargeable drone. Can it be done? Internal completion terminal i Data acquisition; assuming binary variables Indicates the order of access to the terminal, if This indicates that there is a problem with the service process from the terminal. i Fly to Terminal j The route, and vice versa ;set up This indicates changes in energy consumption; this stage of drone... Energy consumption changes, including terminal j The previous collection point was the terminal. i The constraints (6c) and (6f) on the flight trajectory of the UAV u in the optimization problem P1 are equivalent to: (15a) (15b) (15c) (15d) Equation (15a) indicates that the number of times the drone departs from the charging station is equal to the number of times it returns to the charging station, ensuring that the drone... The starting and ending points of each sub-trip are charging stations; Equations (15b) and (15c) represent Terminal scheduling variables The constraints; Equation (15d) represents It is a binary variable; S402. Transform the energy consumption constraint (6d) in optimization problem P1 into: using TSPE. (16a) (16b) (16c) Wherein, equation (16a) represents when hour, drones during the time period For the terminal i The energy consumption used is equal to the sum of the energy consumption during hovering and flight; Equation (16b) indicates that when the drone... The battery is fully charged when the drone departs from the charging station; Equation (16c) represents the drone. The energy consumed does not exceed ; S403. Transform the time constraint equation (6b) in optimization problem P1 into equation (17), which represents the sum of VTC during all hovering and flight phases over the data lifetime. Within, that is: (17) S404. Transform optimization problem P1 into optimization problem P2: (14) (15a)~Equation (15d), Equation (16a)~Equation (16c), Equation (17) in, (18b) ensures that the terminal scheduling variable is a binary variable, and equations (18c) and (18d) ensure that the UAV... The transmission power and flight speed are within the maximum constraints. Indicates drone Maximum flight speed.

6. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 5, characterized in that: In step S5, combined with K - The system employs mean clustering, feature selection, and stochastic optimization to construct the influence factor for each terminal. Based on the terminal's influence factor, a stochastic optimization algorithm is used to calculate the priority of each terminal relative to each drone. The specific steps are as follows: S501. Assume the amount of data to be transmitted by the terminal. Feature one: the distance between the terminal and the charging station Feature two, K -The size of the cluster to which the terminal belongs after -means clustering is feature three. K -The silhouette coefficient of the terminal after mean clustering is feature four; let the geometric center of the terminal already served by the UAV be the feature five; find the geometric center of the remaining terminals already served by the UAV, and let the distance between the terminal and the geometric center be the feature six; S502. Normalize the above 6 features for each terminal to construct the terminal's influence factor; S503. Calculate the priority of each terminal for each drone using a random selection algorithm.

7. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 6, characterized in that: The random selection algorithm described in step S503 is as follows: S5101. Based on six features, six influencing factors were obtained after normalization. ; S5102. Randomly generated N Group impact factor weighting reorganization Calculate the maximum number of terminals served by different weighted reconfigurations under the current scenario conditions, and record the current maximum number of services. The optimal corresponding weight reorganization ; S5103. If optimal weight reorganization From now on N The group's weight values ​​have been updated; repeat step 5102. If the optimal weight reorganization is successful... From now on N Maximum number of services in the group's weight value No longer changing, let For this scenario weight value Execute step 5104; S5104. Solve for the weighted sum of the influence factors in step 5101, and set it as the priority of the terminal. .

8. The terminal scheduling method of the UAV swarm-assisted data acquisition system according to claim 7, characterized in that: The conditions for the optimization problem P3 are: st Equations (14), (15a) to (15d), (16a) to (16c), and (18c) to (18d).

9. A terminal scheduling system for an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system, characterized in that: The method for implementing the method as described in any one of claims 1 to 8 includes a model building module, an initialization optimization problem module, a flight hovering cost module, a TSPE module, a random selection module, an optimization update module, and a solution output module; The model building module is used to build a model of an unmanned aerial vehicle (UAV) swarm-assisted data acquisition system. The initialization optimization problem module is used to construct an optimization problem for terminal scheduling of the UAV swarm assisted data acquisition system model, assuming that the channel between the UAV swarm and each terminal it serves follows a line-of-sight fading model. The aforementioned flight hovering cost module is used to decompose the mission of the UAV swarm into two stages, flight and hovering, using a flight-hovering communication protocol, and to calculate the time and energy costs of the UAV swarm's flight and hovering. The TSPE module is used to consider the energy consumption of drone swarm flight and hovering through the Traveling Salesman Problem (TSPE) and to update and optimize the problem. The random selection module is used to combine K - The system employs mean clustering, feature selection, and stochastic optimization to construct the influence factor for each terminal; based on the terminal's influence factor, a stochastic optimization algorithm is used to calculate the priority of each terminal relative to each drone. The optimization and update module is used to optimize the power, speed, and trajectory of the drone swarm, and further update and optimize the problem; The solution output module is used to iteratively optimize the UAV swarm, solve the further updated optimization problem, and obtain the terminal scheduling scheme of the UAV swarm-assisted data acquisition system.

Images