Trajectory optimization and resource management method in multi-uav empowered digital twin network

Abstract

The application discloses a kind of multi-unmanned aerial vehicle empowerment's trajectory optimization and resource management method in digital twin network, obtain the transmission data of unmanned aerial vehicle and data center, determine the transmission power when unmanned aerial vehicle independent work and collaborative work;According to the transmission power when unmanned aerial vehicle independent work and collaborative work, determine the energy consumed by unmanned aerial vehicle to complete a data collection;Obtain digital twin system information age, wherein digital twin system information age is defined as the sum of digital twin information age and data information age;With the weighted sum of digital twin system information age and energy consumption in the whole task completion time T as the goal of minimization, construct target optimization function;Construct decision model to optimize optimization function, obtain the optimal operation strategy of unmanned aerial vehicle.Can optimize multi-unmanned aerial vehicle collaborative sensing, communication and overlapping area data, reduce data redundancy and energy consumption.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication networks, and in particular to a trajectory optimization and resource management method in a multi-unmanned aerial vehicle enabled digital twin network. BACKGROUND

[0002] With the rapid development of artificial intelligence, big data and cloud computing, the state has introduced a series of policies related to the exploration and construction of digital twin cities. Digital twin (DT) technology is penetrating into various industries in cities. This technology can assess transportation risks and intelligent scheduling in cities.

[0003] Digital twin refers to the use of physical entity models, sensor updates, and historical operation data to map the real physical world into a virtual space, so that the entire life cycle of the corresponding physical entity can be mapped from the virtual space. The technology replicates physical entity objects and creates their virtual copies in digital space, i.e., twins. During the entire life cycle of the digital twin, the physical entity object and its corresponding virtual twin continuously interact and evolve. Digital twin technology was initially applied in aerospace, through digital simulation to monitor and predict the status of spacecraft in real time, thereby improving the safety and reliability of spacecraft materials.

[0004] Unmanned aerial vehicles (UAVs) are important tools for building information service air platforms. By carrying payloads with different capabilities, UAVs can perform communication, remote sensing, storage, and computing functions for different information systems. Compared with traditional terrestrial wireless communication networks, UAVs can effectively supplement ground network infrastructure and perform wireless access network tasks, effectively addressing the inconsistency between ground base station distribution and business needs. Moreover, UAVs can better serve the new paradigm of user-centricity due to their mobility and flexibility.

[0005] This system first considers the trajectory optimization and resource management research problem in the context of multi-unmanned aerial vehicle enabled digital twin network. When applying digital twin technology to smart city construction, one challenge is that the coverage range of fixed sensors is limited by the terrain and cannot cover all areas of the city. In addition, there may be some obstructions between fixed sensors and base stations, resulting in weak communication link transmission rates. Therefore, multiple UAVs carry sensors to perceive the data required by the digital twin system. However, multiple UAVs collaborating to perceive the data required by the digital twin system may result in repeated coverage of the UAV perception area in some cases, i.e., the data of the same perception target may be perceived by multiple UAVs at the same time, resulting in data redundancy and increased energy consumption. SUMMARY

[0006] The technical problem to be solved by the present application is that multiple unmanned aerial vehicles may cause repeated coverage of unmanned aerial vehicle sensing areas in some cases during cooperative sensing, resulting in data redundancy and increased energy consumption.

[0007] The present application is implemented by the following technical solutions:

[0008] The present application provides a multi-unmanned aerial vehicle enabled digital twin network trajectory optimization and resource management method, comprising the following specific steps:

[0009] Obtain the transmission data of the unmanned aerial vehicle and the data center, and determine the transmission power of the unmanned aerial vehicle when working independently and when working cooperatively;

[0010] Determine the energy consumed by the unmanned aerial vehicle to complete one data collection according to the transmission power when working independently and when working cooperatively;

[0011] Obtain the digital twin system information age, wherein the digital twin system information age is defined as the sum of the digital twin information age and the data information age;

[0012] Construct a target optimization function to minimize the weighted sum of the digital twin system information age and energy consumption within the entire task completion time T;

[0013] Construct a decision model to optimize the optimization function and obtain the optimal operation strategy of the unmanned aerial vehicle.

[0014] The application considers the independent working state and the cooperative working state of the multiple unmanned aerial vehicles, obtains the energy consumed by the unmanned aerial vehicles to complete data collection once, adopts deep reinforcement learning to minimize the weighted sum of the digital twin system information age and the energy consumption within the entire task completion time T as the goal, constructs a target optimization function to optimize the unmanned aerial vehicle running track, and based on this, the multiple unmanned aerial vehicle cooperative sensing, communication and overlapping area data can be optimized, and data redundancy and energy consumption can be reduced.

[0015] Further, the transmission power of the unmanned aerial vehicle when working independently and when working cooperatively includes:

[0016] The transmission rate of the unmanned aerial vehicle between independent transmission and the data center;

[0017] The time consumed by the unmanned aerial vehicle to transmit all data in the sensing area when transmitting independently;

[0018] The cooperative transmission rate between the unmanned aerial vehicles;

[0019] The time consumed by the unmanned aerial vehicles to transmit data cooperatively.

[0020] Further, the calculation step of the transmission rate of the unmanned aerial vehicle between independent transmission and the data center includes:

[0021] Obtain the transmission data of the unmanned aerial vehicle and the data center, and determine the line-of-sight transmission probability and non-line-of-sight transmission probability of the unmanned aerial vehicle and the data center;

[0022] According to the line-of-sight transmission probability and the non-line-of-sight transmission probability of the unmanned aerial vehicle and the data center, determine the path loss of the line-of-sight transmission and the path loss of the non-line-of-sight transmission;

[0023] According to the path loss of the line-of-sight transmission and the path loss of the non-line-of-sight transmission, determine the average path loss of the unmanned aerial vehicle when transmitting;

[0024] Obtain the transmission power of the unmanned aerial vehicle when transmitting independently, and determine the transmission rate of the unmanned aerial vehicle between independent transmission and the data center in combination with the average path loss;

[0025] The calculation step of the time consumed by the unmanned aerial vehicle to transmit all data in the sensing area when transmitting independently includes:

[0026] According to the transmission rate between the unmanned aerial vehicle and the data center when transmitting data independently, the data sensing strategy of the unmanned aerial vehicle, and the state update data of the gth sensing target at time slot t, determine the time consumed by the unmanned aerial vehicle to transmit all data in the sensing area when transmitting independently;

[0027] The calculation step of the cooperative transmission rate between the UAVs comprises:

[0028] According to the average path loss and the transmission power of the UAVs in the cooperative transmission stage, the cooperative transmission rate between the UAVs is determined.

[0029] The calculation step of the time consumed by the UAVs for cooperative transmission of data comprises:

[0030] According to the data sensing strategy of the UAV, whether the sensing target g is repeatedly covered by multiple UAVs, and the state update data of the gth sensing target in the time slot t, the time consumed by the UAVs for cooperative transmission of data is determined.

[0031] Further, the determination of the energy consumed by the UAV for completing one data collection specifically comprises:

[0032] The flight data of the UAV is obtained, and the flight power of the UAV is determined;

[0033] According to the flight power of the UAV and the flight time of the UAV, the energy consumed by the UAV during flight is determined.

[0034] According to the time consumed by the UAV for independently transmitting all data in the sensing area of the UAV, the transmission power of the UAV during independent transmission, the time consumed by the UAV for cooperative transmission of data, and the transmission power of the UAV during cooperative transmission, the transmission energy consumption of the UAV is determined.

[0035] According to the energy consumed by the UAV during flight and the transmission energy consumption of the UAV, the energy consumed by the UAV for completing one data collection is determined.

[0036] Further, the acquisition step of the digital twin information age specifically comprises:

[0037] Based on the time consumed by the UAV for independently transmitting all data in the sensing area of the UAV and the time consumed by the UAV for cooperative transmission of data, the calculation time required for processing the data transmitted by the UAV is determined.

[0038] In combination with the latest time slot of the base station from the updated digital twin data of the current time slot t, the information age of the digital twin in the time slot t is determined.

[0039] Further, the acquisition step of the data information age specifically comprises:

[0040]

[0041] Wherein, represents the latest time slot of the UAV from which the data generated by the sensing target g is collected in the current time slot t.

[0042] Further, the construction of the decision model specifically comprises:

[0043] acquire unmanned aerial vehicle state data, construct a state function;

[0044] acquire unmanned aerial vehicle action data, construct an action function;

[0045] Based on the state function and the action function, an optimal unmanned aerial vehicle operation strategy is obtained by constructing a reward function in combination with an optimization objective.

[0046] Further, the target optimization function specifically includes:

[0047]

[0048] s.t.C1:

[0049] C2:

[0050] C3:

[0051] C4:V min ≤V u ≤V max

[0052] C5:

[0053] C6:

[0054] C7:

[0055] wherein α and β represent weights, C1 represents an information age threshold of each perception target data update, represents the data information age generated by the perception target g; C2 represents the energy constraint of each unmanned aerial vehicle, represents the energy consumed by the unmanned aerial vehicle u for completing one data collection; C3 specifies the movement range of the unmanned aerial vehicle, indicating that each unmanned aerial vehicle can only move within the specified area; C4 specifies the speed range of each unmanned aerial vehicle; C5 avoids collision between unmanned aerial vehicles, R in represents the minimum safe distance between any two unmanned aerial vehicles; C6 constrains the maximum transmission power of each unmanned aerial vehicle, represents the transmission power of the unmanned aerial vehicle u in the cooperative transmission phase, represents the transmission power of the unmanned aerial vehicle u in the cooperative transmission phase; C7 indicates that the data collection strategy of the unmanned aerial vehicle is a binary variable, which can only take 0 or 1.

[0056] Further, the constructing a decision model and optimizing the function specifically includes:

[0057] acquire unmanned aerial vehicle state data, construct a state function;

[0058] acquire unmanned aerial vehicle action data, and construct an action function;

[0059] construct a reward function based on the state function and the action function, in combination with an optimization target, to obtain an optimal operation strategy of the unmanned aerial vehicle.

[0060] Further, the construction of the state function specifically includes:

[0061] constructing the state function based on the position of the unmanned aerial vehicle at time slot t, the data size generated by the perception target, the digital twin information age, the information age of all perception targets, the information age threshold of all perception targets, and the remaining energy of the unmanned aerial vehicle;

[0062] The state function is:

[0063]

[0064] wherein represents the remaining energy set of each unmanned aerial vehicle at time slot t, represents the position of the unmanned aerial vehicle u at time slot t, represents the position information of the perception target g, represents the data information age of the perception target g, represents the data information age threshold of the perception target g, A dt,t represents the information age of the digital twin.

[0065] Further, the construction of the action function specifically includes:

[0066] Based on the current state s and the experience of other agents, the embodiment constructs the action function by designing the flight angle of all unmanned aerial vehicles at time slot t, the flight speed of all unmanned aerial vehicles at time slot t, the independent transmission power of all unmanned aerial vehicles, and the cooperative transmission power of all unmanned aerial vehicles.

[0067] The action function is:

[0068] a t =(V t ,Θ t ,P s,t ,P c,t )

[0069] wherein represents the flight speed of all unmanned aerial vehicles, represents the flight angle of all unmanned aerial vehicles at time slot t, represents the independent transmission power of all unmanned aerial vehicles at time slot t, represents the cooperative transmission power of all unmanned aerial vehicles at time slot t.

[0070] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0071] The present application considers the independent working state and the cooperative working state of the multiple unmanned aerial vehicles by collecting the data of the multiple unmanned aerial vehicles flying to the sensing target by carrying sensors, obtains the energy consumed by the unmanned aerial vehicle to complete a data collection, adopts deep reinforcement learning to minimize the weighted sum of the digital twin system information age and the energy consumption within the entire task completion time T as the target, constructs a target optimization function, optimizes the unmanned aerial vehicle running track, and based on this, can optimize the multiple unmanned aerial vehicle cooperative sensing, communication and overlapping area data, reduce data redundancy and energy consumption.

[0072] The present application can ensure the freshness of the digital twin model and reduce energy consumption under the premise of ensuring the real-time data of the data points and not increasing the cost, and has strong application value and development potential. BRIEF DESCRIPTION OF DRAWINGS

[0073] In order to more clearly illustrate the technical scheme in the example embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments, and it should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation to the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor under the premise of the drawings.

[0074] Figure 1 The optimization method in the embodiment of the present application;

[0075] Figure 2 The system model schematic diagram in the embodiment of the present application;

[0076] Figure 3 The overlapping area schematic diagram in the embodiment of the present application;

[0077] Figure 4 The collaborative scheme schematic diagram in the embodiment of the present application;

[0078] Figure 5 The information age schematic diagram in the embodiment of the present application;

[0079] Figure 6 The system optimization problem algorithm convergence graph in the embodiment of the present application. DETAILED DESCRIPTION

[0080] In order to make the objects, technical solutions and advantages of the present application clearer, further detailed description will be made to the present application in combination with embodiments and drawings, the illustrative embodiments and the description thereof are only used to explain the present application, and do not limit the present application.

[0081] As a possible implementation, as shown in Figure 1 , the embodiment provides a trajectory optimization and resource management method in a multi-UAV enabled digital twin network, comprising the following specific steps: acquiring transmission data of the UAV and the data center, and determining transmission power of the UAV when working independently and when working cooperatively; according to the transmission power of the UAV when working independently and when working cooperatively, determining energy consumed by the UAV for completing a data collection; acquiring digital twin system information age, wherein the digital twin system information age is defined as the sum of digital twin information age and data information age; taking the weighted sum of the digital twin system information age and the energy consumption in the whole task completion time T as the target, constructing a target optimization function; constructing a decision model to optimize the optimization function, and obtaining an optimal operation strategy of the UAV. In order to ensure that the UAV knows all the real-time requirements of the data sensing target for data, the real-time performance of the present scheme is embodied by the sensing target data information age, therefore, multiple UAVs fly to the sensing target to collect data required for constructing the digital twin model by carrying sensors, and the data in the overlapping area is divided, finally, all the UAVs transmit the collected data to the base station BS, the base station processes the received data, and finally constructs the digital twin model, which is used to optimize the multi-UAV cooperative sensing, communication and overlapping area data, and reduce data redundancy and energy consumption.

[0082] Specifically, as shown in Figure 2 , the embodiment system considers the scenario of multi-UAV (Unmanned Aerial Vehicle, UAV) assisted digital twin. In this scenario, there are UAVs to perform a cooperative sensing task, and the whole task flight time T is discretized into t equal time slots, and the duration of each time slot is δ0. The position of the UAV u in the time slot t is denoted as Meanwhile, the embodiment considers that there are sensing targets in the environment, and the UAV needs to collect data of the sensing targets in the environment and transmit the data to the base station, and the position of the sensing target is defined as q g = [x g , y g , 0], and in addition, the position of the base station is denoted as q b = [x b , y b,H]. When the location of the perceived target is within the sensing range of the UAV, the UAV can collect information data about the perceived target. In addition, in order to prevent collisions between UAVs, each UAV is defined as having a circular safety protection area with a radius of R in , other UAVs cannot hover in this area, so the following constraints are imposed on the u-th UAV and the j-th UAV:

[0083] In some possible embodiments, such as Figure 3 and Figure 4 As shown in the figure, the transmission power of drones when working independently and collaboratively, including: the transmission rate between the drone in independent transmission and the data center; the time consumed by the drone to sense all data in the area when transmitting independently; the transmission rate of collaborative work between drones; and the time consumed by collaborative data transmission between drones.

[0084] Considering the link between the UAV and the data center, including line-of-sight (LoS) transmission and non-line-of-sight (NLoS) transmission, the LoS link between the UAV and the data center depends on factors such as the elevation angle, distance, and environment between the UAV and the data center. At time slot t, the probability of LoS transmission and non-line-of-sight transmission between UAV u and the data can be defined as:

[0085]

[0086] P NLoS (d u,b t )=1-P LoS (d u,b t )

[0087] Among them, a and b are constants related to the environment, h is the flight altitude of the UAV, and d u,b t is the distance between the UAV u and the base station.

[0088] The path loss for line-of-sight transmission and non-line-of-sight transmission can be expressed as:

[0089]

[0090]

[0091] Among them, f c is the carrier frequency, η NLoS and η LoS are the additional attenuation coefficients due to LoS ​​and NLoS connections, respectively, and c is the speed of light in a vacuum. Therefore, the average path loss is defined as:

[0092]

[0093] Thus, the transmission rate between the UAV u and the data center when the UAV u transmits independently is:

[0094]

[0095] where denotes the transmission power of the UAV u when transmitting independently, σ 2 denotes the noise power, denotes the bandwidth allocated to the u-th UAV at the base station.

[0096] As Figure 3 shown, a sensing target can be covered by multiple UAVs, thus the targets within the sensing range of the UAVs can be divided into two types, namely, independent sensing targets and overlapping sensing targets. For the independent sensing targets, the UAV u transmits the update data of the independent sensing targets to the base station independently, and a binary variable is defined to represent the independent collection strategy of the u-th UAV, where indicates that the u-th UAV will collect the data of the g-th sensing target independently at time slot t, otherwise A binary variable is defined to represent whether the g-th sensing target is covered by multiple UAVs, where indicates that the g-th sensing target is covered by the UAV u at time slot t, and the UAVs will cooperatively transmit the state update data of the g-th sensing target, otherwise 0.

[0097] Thus, the time for the UAV u to independently transmit the data consumed in its sensing area is defined as:

[0098]

[0099] where, denotes the state update data of the g-th sensing target at time slot t.

[0100] As Figure 4 shown, the multi-UAV cooperative assisted digital twin scheme includes one sensing phase and two transmission phases. First, the UAVs fly to their pre-set sensing positions at the same time, and after all the UAVs arrive at the sensing positions, the sensing information is started to be sensed simultaneously. Then, the UAVs start to transmit the data information of the independent sensing targets in their sensing range, and after the UAVs finish transmitting the data information of the independent sensing targets, the second phase of the UAVs starts to cooperatively transmit the data information of the overlapping sensing targets by forming a virtual MISO. In order to reduce the data transmission time, the UAVs with larger channel gains will transmit more data.

[0101] After the UAVs transmit their independent sensing data, the UAVs start the second phase of cooperative transmission. The cooperative transmission rate between the UAVs can be expressed as:

[0102]

[0103] where represents the transmission power of the UAV u in the cooperative transmission phase, B represents the total bandwidth of the system, σ 2 represents the Gaussian white noise power.

[0104] Therefore, the time consumed by the UAV u to cooperatively transmit data is:

[0105]

[0106] After the base station receives the data transmitted by the UAVs, it starts to process the data. Therefore, the calculation time required to process the data transmitted by the UAVs is:

[0107]

[0108] where f t,DT is the calculation resource of the base station when processing the data transmitted by the UAVs at time slot t.

[0109] As Figure 5 shown in the embodiment, the target of the study is the weighted sum of the information age and the energy consumption of the system. Therefore, the updated data of the sensing target is received by the UAVs, and then the latest data is transmitted by the UAVs to the base station. Finally, the base station processes the data from all the UAVs. Therefore, the information age of the digital twin at time slot t in this embodiment is defined as A t,dt = t - u t , where u t represents the latest time slot of the updated digital twin data of the base station from the current time slot t. At time slot t1, the base station updates the digital twin model of the system after processing the data from the UAVs. At time slot t2, the base station receives the latest data information from the UAVs again, and replaces the original digital twin model with the latest digital twin model after processing the data at time slot t3. Therefore, at time slot t3, the information age of the digital twin can be expressed as Δ = t3 - t1.

[0110] The data information age of the sensing target is defined as the time difference of the UAVs collecting the data of the same sensing target. Therefore, at time slot t, the data information age of the sensing target g is expressed by the following formula:

[0111]

[0112] where, represents the latest time slot of the UAVs collecting the data of the sensing target g from the current time slot t.

[0113] In the process of data sensing task of the UAV, the consumed energy of the UAV includes the following aspects: flight energy consumption, transmission energy consumption, the power of which is respectively represented as Therefore, the flight power of the UAV u can be represented as:

[0114]

[0115] Wherein, P0 is the blade profile power of the UAV, P1 is the sensing power of the UAV in hovering state, U tip is the tip speed of the UAV blade, v0 is the average rotor sensing speed of the UAV in hovering state, d0 is the fuselage drag ratio, s is the rotor solidity of the UAV, V is the flight speed of the UAV. ρ represents air density, and A is the rotor disc area of the UAV.

[0116] The energy consumed by the UAV u in flight can be represented as: Wherein T fly represents the flight time of the UAV u, which is related to the position of the UAV u in continuous two hovering states.

[0117] Therefore, the transmission energy consumption of the UAV u can be represented as:

[0118]

[0119] According to the above analysis, the energy consumed by the UAV u to complete one data collection is:

[0120]

[0121] According to the above analysis of the wireless channel model of the system and the model construction of the system information age, the system information age optimization problem under dynamic cooperative transmission is defined as the minimum total cost problem, and the formula is as follows:

[0122]

[0123] s.t.C1:

[0124] C2:

[0125] C3:

[0126] C4:V min ≤V u ≤V max

[0127] C5:

[0128] C6:

[0129] C7:

[0130] where a and b represent the weights, C1 represents the information age threshold of each perception target data update, denotes the data information age generated by the perception target g; C2 represents the energy constraint of each UAV, denotes the energy consumed by the UAV u to complete one data collection; C3 specifies the movement range of the UAV, indicating that each UAV can only move within the specified area; C4 specifies the speed range of each UAV; C5 avoids collision between UAVs, R in denotes the minimum safe distance between any two UAVs; C6 constrains the maximum transmit power of each UAV, denotes the transmit power of the UAV u in the cooperative transmission phase, denotes the transmit power of the UAV u in the cooperative transmission phase; C7 indicates that the data collection strategy of the UAV is a binary variable, which can only take 0 or 1.

[0131] For the above optimization problem, in the embodiment, we do not consider solving the problem directly from a mathematical point of view, but consider solving the problem through deep reinforcement learning (DRL) to obtain a feasible UAV trajectory and resource allocation scheme.

[0132] The scheme will use the MADDPG algorithm based on centralized training and distributed execution to perform distributed trajectory optimization strategy for the multi-UAV assisted digital twin network. The strategy first performs centralized training on multiple UAVs according to the global state information of the system, and after the training is completed, each UAV can perform optimal trajectory optimization and resource allocation according to the local state information observed by itself.

[0133] The state s, action a and system reward r are described as follows:

[0134] In order to facilitate the representation of the real-time state of the multi-UAV assisted digital twin system, the embodiment selects the position of the UAV at time slot t, the data amount generated by the perception target, the digital twin information age, the information age of all perception targets, the information age threshold of all perception targets, and the remaining energy of the UAV to form a state space. Therefore, the state of the entire system at time slot t can be represented as:

[0135]

[0136] where represents the remaining energy set of each UAV at time slot t, represents the location of the UAV u at time slot t, represents the location information of the perception target g, represents the data information age of the perception target g, represents the data information age threshold of the perception target g, A dt,t represents the information age of the digital twin.

[0137] Similarly, based on the current state s and the experience of other agents, the embodiment designs a system action space composed of the flight angle of all UAVs at time slot t, the flight speed of all UAVs at time slot t, the independent transmission power of all UAVs, and the cooperative transmission power of all UAVs. Therefore, the action of the system at time slot t can be represented as:

[0138] a t = (V t , Θ t , P s,t , P c,t )

[0139] wherein, represents the flight speed of the UAV u, represents the flight angle of all UAVs at time slot t, represents the independent transmission power of all UAVs at time slot t, represents the cooperative transmission power of all UAVs at time slot t.

[0140] Finally, the reward function design of the multi-UAV assisted digital twin system, the essence of using deep reinforcement learning algorithm to solve the optimization problem is to convert the optimization problem into a system cumulative reward maximization problem. In the content studied in this embodiment, the final optimization target mainly includes three parts: the information age of the digital twin, the data information age, and the energy consumption of the UAV. Therefore, the reward function of this embodiment is related to the above three parts, and the instantaneous reward value of the UAV u at time slot t is designed as:

[0141]

[0142] As shown in Figure 6 , the system performance superiority of the present application is verified by simulation experiment. The system simulation parameters are set as follows: the site area is 500m x 500m, the number of UAVs is 4, the number of perception targets is 30, the initial energy of each UAV is 2e4J, the flight height of the UAV is 50m, the maximum flight speed of the UAV is 50m / s, the transmission power of the UAV is 0.1w, the information age threshold of the perception target update data (unit, s) is between 20-90, the entire data collection period T=300s, the bandwidth B=10MHZ, and the base station computing capacity is 70GHZ.

[0143] The above detailed description of the specific embodiments of the present application is provided for the purpose of further explaining the objects, technical solutions and advantages of the present application, and it should be understood that the above description is only a specific embodiment of the present application and is not used to limit the protection scope of the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A trajectory optimization and resource management method in a multi-UAV-enabled digital twin network, characterized in that: The specific steps include: Obtain transmission data between drones and data centers to determine the transmission power of drones when working independently and collaboratively; Determine the energy consumed by the drones to complete one data collection operation based on the transmission power of the drones when working independently and collaboratively; Obtain the information age of the digital twin system, where the information age of the digital twin system is defined as the sum of the digital twin information age and the data information age; The objective is to minimize the weighted sum of the information age and energy consumption of the digital twin system within the entire task completion time T, and construct the target optimization function; Construct the decision model objective optimization function for optimization to obtain the optimal operation strategy of the UAV; Determining the energy consumed by the drone to complete one data collection operation specifically includes: Obtain the flight data of the drone and determine the flight power of the drone; Determine the energy consumed by the drone during flight based on the drone's flight power and flight time; The transmission energy consumption of the drone is determined based on the time it takes for the drone to independently transmit all the data in its sensing area, the transmission power of the drone when performing independent transmission, the time it takes for the drone to collaboratively transmit data, and the transmission power of the drone when performing collaborative transmission; Based on the energy consumed by the drone during flight and the transmission energy consumption of the drone, determine the energy consumed by the drone to complete one data collection; The steps for obtaining the digital twin information age specifically include: Determine the computational time required to process the data sent by the drones based on the time it takes for the drones to independently transmit all the data in their sensing area and the time it takes for the drones to collaboratively transmit data. Determine the information age of the time slot digital twin based on the latest time slot of the digital twin data updated by the base station from the current time slot; The steps of obtaining the data information age specifically include: in, It represents the latest time slot of the data generated by the UAV’s collection of the perception target g from the current time slot t; The target optimization function specifically includes: in, and Represent weights respectively, C1 represents the information age threshold for each perception target data update, represents the age of data information generated by the perception target g; C2 represents the energy constraint of each UAV, Indicates drone The energy consumed to complete a data collection; C3 specifies the movement range of the UAV, indicating that each UAV can only move within the specified area; C4 specifies the speed range of each UAV; C5 avoids collisions between UAVs. Indicates the minimum safe distance between any two drones; C6 constrains the maximum transmission power of each drone. Indicates drone The transmission power during independent transmission, Indicates drone The transmission power in the cooperative transmission phase; C7 indicates that the data collection strategy of the drone is a binary variable, which can only take 0 or 1. The binary definition variable represents the independent collection strategy of the first UAV, when When, it indicates that at time slot t The drone will independently collect data on the g-th perception target, otherwise , binary definition variables Indicates whether the perceived target g is repeatedly covered by multiple drones, where Indicates that the target is detected by the drone at time slot t Repeated coverage, the UAVs will cooperate to transmit the status update data of the perception target g, otherwise it is 0.

2. The trajectory optimization and resource management method in a multi-UAV-enabled digital twin network according to claim 1 is characterized in that: The construction of the decision model target optimization function for optimization specifically includes: Get the drone status data and build the status function; Obtain drone action data and construct action functions; Based on the state function and action function, combined with the optimization goal, a reward function is constructed to obtain the optimal operation strategy of the drone.

3. The trajectory optimization and resource management method in a multi-UAV-enabled digital twin network according to claim 2 is characterized in that: The construction state function specifically includes: Based on the position of the UAV at time slot t, the amount of data generated by the perceived target, the information age of the digital twin, the information age of all perceived targets, the information age threshold of all perceived targets, and the remaining energy of the UAV, a state function is constructed; The state function is: in, represents the remaining energy set of each UAV at time slot t, Indicates that the drone in time slot t Location, Indicates the location information of the perceived target g, Represents the age of data information of the perceived target g, Represents the age threshold of the data information of the perceived target g, Represents the information age of the digital twin.

4. The trajectory optimization and resource management method in a multi-UAV-enabled digital twin network according to claim 2 is characterized in that: The construction action function specifically includes: Based on the current state s and the agent's experience, the action function is constructed by designing the flight angles of all drones at time slot t, the flight speeds of all drones at time slot t, the independent transmission power of all drones, and the collaborative transmission power of all drones. The action function is: in, Representative drone The flight speed, represents the flight angle of all UAVs in time slot t, represents the independent transmission power of all UAVs in time slot t, represents the collaborative transmission power of all UAVs in time slot t.

Images