A Synchronous Data Sensing Method, Device and Equipment Based on Edge Multi-Point Collaboration

Abstract

The present invention provides a synchronous data perception method, device and equipment based on edge multi-point cooperation. Among them, the method includes: obtaining first perception parameters and second perception parameters when an unmanned aerial vehicle (UAV) and a ground physical device perform data perception on a perception target; determining a synchronous data utility evaluation model during data perception according to the first perception parameters and the second perception parameters; determining an optimal flight trajectory set of the UAV according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model; updating the optimal flight trajectory according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory; synchronously perceiving data information of the target by the ground physical device and the UAV based on the target flight trajectory and feeding it back to an edge server. The solution of the present invention can accurately perceive a perception target with real-time changes through the collaborative action of the UAV and the ground physical device, so as to improve the accuracy and coverage of data perception.

Description

Technical Field

[0001] The present invention relates to the technical field of computer information processing, and particularly to a synchronous data sensing method, device and equipment based on edge multi-point cooperation. Background Art

[0002] In the context of the rapid development of current digital twin technology, the importance of synchronous data sensing has become increasingly prominent. With the wide application of edge computing, the Internet of Things and drone-assisted communication, it has become the key to improving the intelligent level of the system to sense environmental data in real time, accurately and efficiently. Only by ensuring the timeliness and consistency of data can a digital twin model be accurately constructed to achieve accurate mapping and dynamic optimization of the physical world. During the task execution process, if there are delays or inconsistencies in data sensing, it will not only affect the accuracy of computing offloading and resource scheduling, but may also lead to decision-making errors and reduce the overall system performance. Therefore, at present, constructing an accurate, efficient and low-latency synchronous data sensing mechanism has become a research topic that has received much attention.

[0003] However, the sensing schemes in the existing synchronous data sensing technologies adopt the method of single base station or Internet of Things device sensing, which is difficult to meet the sensing requirements of the system for time-varying sensing targets and achieve true synchronization. In particular, the ground sensing devices at fixed positions are greatly affected by line-of-sight blocking and environmental factors, and cannot comprehensively sense the targets. Especially when facing time-varying sensing targets, the synchronous data sensing schemes in the existing technologies cannot effectively sense the sensing targets, cannot dynamically allocate resources well, and are extremely likely to result in poor overall system performance. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a synchronous data sensing method, device and equipment based on edge multi-point cooperation, which can accurately sense time-varying sensing targets through the collaborative action of drones and ground physical devices, so as to improve the accuracy and coverage of data sensing.

[0005] To solve the above technical problem, an embodiment of the present invention provides a synchronous data sensing method based on edge multi-point cooperation, including:

[0006] Obtaining a first sensing parameter when a drone senses a sensing target during flight and a second sensing parameter when a ground physical device senses the sensing target;

[0007] Determining a synchronous data utility evaluation model when the drone and the ground physical device perform data sensing according to the first sensing parameter and the second sensing parameter;

[0008] Determine an optimal flight trajectory set of the UAV according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set includes multiple different optimal flight trajectories;

[0009] Update the optimal flight trajectories in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory;

[0010] According to the ground physical device and the UAV based on the target flight trajectory, synchronously sense the data information of the sensed target and feedback it to the edge server.

[0011] In one embodiment, according to the first sensing parameter and the second sensing parameter, determining a synchronous data utility evaluation model for data sensing by the UAV and the ground physical device includes:

[0012] Determine the index values for data sensing by the UAV and the ground physical device according to the first sensing parameter and the second sensing parameter;

[0013] Determine the synchronous data utility evaluation model according to the index values.

[0014] In one embodiment, the synchronous data utility evaluation model is constructed by the following formula:

[0015] ;

[0016] where represents the sensing completeness index value, represents the sensing accuracy index value, represents the sensing timeliness index value, represents the sensing energy efficiency index value, where respectively represent the weight coefficients of the corresponding index values, and .

[0017] In one embodiment, according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, determining an optimal flight trajectory set of the UAV includes:

[0018] Generate an initial flight trajectory set of the UAV according to a preset random generation algorithm;

[0019] Update the initial flight trajectory set according to the preset initialization trajectory algorithm to obtain an updated flight trajectory set, where the updated flight trajectories in the updated flight trajectory set correspond one-to-one to the initial flight trajectories in the initial flight trajectory set;

[0020] Determine the optimal flight trajectory set according to the synchronization data utility evaluation model, the initial flight trajectory set, and the updated flight trajectory set.

[0021] In one embodiment, updating the initial flight trajectory set according to the preset initialization trajectory algorithm to obtain an updated flight trajectory set, including:

[0022] According to the formula: Perform an update process on each initial flight trajectory in the initial flight trajectory set once to obtain a first flight trajectory corresponding one-to-one to the initial flight trajectory; where represents the initial flight coordinate; represents the first flight coordinate; , respectively represent , the weight coefficients of, and are both random numbers between 0 and 1; is the weight coefficient of fel; represents a normal distribution with a mean of 0 and a variance of 1; represents the coordinate vector difference between the optimal initial flight coordinate in the initial flight trajectory set and the initial flight coordinate at the current moment; represents the coordinate vector difference between a randomly selected coordinate from the first three optimal initial flight coordinates, the mean of all initial flight coordinates, and the median of the initial flight coordinates in the initial flight trajectory set and the initial flight coordinate at the current moment; represents the coordinate vector difference between the UAV at a random position and the initial flight coordinate at the current moment;

[0023] According to the formula: Perform a secondary update process on the first flight trajectory to obtain an updated flight trajectory set composed of updated flight trajectories corresponding one-to-one to the first flight trajectory; where represents the updated flight coordinate; represents the helical flight parameter of the UAV; represents the Levy flight parameter of the UAV; M represents the adaptive parameter; represents a random factor, randomly taking values between 0 and 1, and is used to select the position update formula.

[0024] In one embodiment, determining the optimal flight trajectory set according to the synchronization data utility evaluation model, the initial flight trajectory set, and the updated flight trajectory set, including:

[0025] According to the synchronous data utility evaluation model, determine the initial utility value corresponding to each initial flight trajectory in the initial flight trajectory set and the updated utility value corresponding to each updated flight trajectory in the updated flight trajectory set;

[0026] According to the initial utility value and the updated utility value, perform screening processing on the initial flight trajectory set and the updated flight trajectory set to obtain the optimal flight trajectory set.

[0027] In one embodiment, updating the optimal flight trajectory in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory, including:

[0028] Perform forward diffusion noise addition and reverse diffusion noise reduction processing on the optimal flight trajectory in the optimal flight trajectory set according to the preset diffusion model to obtain a flight trajectory after diffusion processing;

[0029] Determine the trajectory adjustment action of the flight trajectory after diffusion processing according to the preset depth gradient model;

[0030] Adjust the flight trajectory after diffusion processing according to the trajectory adjustment action to obtain the target flight trajectory.

[0031] An embodiment of the present invention further provides a synchronous data perception device based on edge multi-point collaboration, which is characterized in that it includes:

[0032] An acquisition module, configured to acquire the first perception parameter when the drone performs data perception on the perception target during flight and the second perception parameter when the ground physical device performs data perception on the perception target;

[0033] A processing module, configured to determine a synchronous data utility evaluation model for the drone and the ground physical device to perform data perception according to the first perception parameter and the second perception parameter; determine an optimal flight trajectory set of the drone according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set includes multiple different optimal flight trajectories; update the optimal flight trajectory in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory; synchronously perceive data information of the perception target by the ground physical device and the drone based on the target flight trajectory and feedback it to the edge server.

[0034] An embodiment of the present invention further provides a computing device, including:

[0035] A memory, configured to store one or more programs;

[0036] One or more processors for executing the one or more programs to implement the method as described above.

[0037] An embodiment of the present invention further provides a computer-readable storage medium, in which a program is stored, and when the program is executed by a processor, the method as described above is implemented.

[0038] The above solution of the present invention has at least the following beneficial effects:

[0039] The synchronous data perception method, device and equipment based on edge multi-point cooperation provided by the above solution of the present invention, wherein the method includes: obtaining a first perception parameter when a drone performs data perception on a perception target during flight and a second perception parameter when a ground physical device performs data perception on the perception target; determining a synchronous data utility evaluation model for the drone and the ground physical device to perform data perception according to the first perception parameter and the second perception parameter; determining an optimal flight trajectory set of the drone according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set contains multiple different optimal flight trajectories; updating the optimal flight trajectories in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory; synchronously perceiving data information of the perception target by the ground physical device and the drone based on the target flight trajectory and feeding it back to the edge server. The solution provided by the above embodiment of the present invention can accurately perceive a perception target with real-time changes through the cooperation of the drone and the ground physical device, so as to improve the accuracy and coverage of data perception. Description of the Drawings

[0040] Figure 1 is a flowchart of the synchronous data perception method based on edge multi-point cooperation provided by an embodiment of the present invention;

[0041] Figure 2 is a perception system diagram of a drone-assisted edge network provided by an optional embodiment of the present invention;

[0042] Figure 3 is a schematic block diagram of the modules of the synchronous data perception device based on edge multi-point cooperation provided by an embodiment of the present invention;

[0043] Figure 4 is a schematic block diagram of an electronic device provided by an embodiment of the present invention;

[0044] Figure 5 is a schematic block diagram of a computing device provided by an embodiment of the present invention. Detailed Embodiments

[0045] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0046] In the following description, certain specific details are set forth for the purpose of illustrating various disclosed embodiments to provide a thorough understanding of the various disclosed embodiments. However, those skilled in the relevant art will recognize that the embodiments can be practiced without one or more of these specific details. In other instances, well-known devices, structures, and techniques associated with the present application may not be shown or described in detail so as not to unnecessarily obscure the description of the embodiments.

[0047] References to "one embodiment" or "an embodiment" throughout the specification mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, the appearances of "in one embodiment" or "in an embodiment" throughout the specification do not necessarily all refer to the same embodiment. Additionally, the particular features, structures, or characteristics may be combined in any manner in one or more embodiments.

[0048] In the following description, for the purpose of clearly showing the structure and working mode of the present invention, many directional terms will be used for description. However, words such as "front", "rear", "left", "right", "outer", "inner", "outward", "inward", "up", "down", etc. should be understood as convenient terms and should not be understood as limiting terms.

[0049] As Figure 1 shown, an embodiment of the present invention provides a synchronous data perception method based on edge multi-point collaboration, including:

[0050] Step 11, obtaining a first perception parameter when the unmanned aerial vehicle (UAV) performs data perception on a perception target during flight and a second perception parameter when a ground physical device performs data perception on the perception target;

[0051] Step 12, determining a synchronous data utility evaluation model for the UAV and the ground physical device to perform data perception according to the first perception parameter and the second perception parameter;

[0052] Step 13, determining an optimal flight trajectory set of the UAV according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set includes multiple different optimal flight trajectories;

[0053] Step 14: Update the optimal flight trajectory in the optimal flight trajectory set according to the preset diffusion model and the preset depth gradient model to obtain the target flight trajectory;

[0054] Step 15: The ground physical device and the drone based on the target flight trajectory synchronously sense the data information of the target and feedback it to the edge server.

[0055] In this embodiment, the mobility of the drone is utilized to quickly move to the target area to achieve effective sensing of the target in the dead corner; through the collaborative sensing of the drone and the ground physical device, data can be collected from different angles and heights, which can improve the accuracy and coverage of data sensing. By the collaborative action of the drone and the ground physical device to sense the target data, it can effectively balance the resources, delay and energy consumption of the sensing system, and avoid the problems that in the existing synchronous data sensing scheme, the single base station or Internet of Things device sensing method is difficult to meet the sensing requirements of the system for the target with real-time changes and cannot comprehensively sense the target.

[0056] Here, considering indicators such as sensing accuracy, sensing completeness, sensing timeliness, and sensing energy efficiency during data sensing to ensure data quality and network efficiency, the optimal flight trajectory set of the drone in the static environment can be determined through the preset initialization trajectory algorithm and the synchronous data utility evaluation model constructed based on the first sensing parameter and the second sensing parameter, so that the flight trajectory of the drone is as optimal as possible in terms of sensing completeness, integrity, timeliness, and energy efficiency; here, the first sensing parameter may include the sensing frequency when the drone senses the target at any moment, the first distance between the drone and the target, the first sensing factor (the influencing factor of the external environment on the sensing quality), the maximum sensing range of the drone (depending on the performance of the drone itself, when the target exceeds the maximum sensing range of the drone, the drone cannot sense the target at this time), and the first time required for the drone to perform a sensing operation; the second sensing parameter may include the second distance between the ground physical device and the target, the second sensing factor (a positive parameter for evaluating the sensing detection quality according to environmental conditions), and the second time required for the ground physical device to perform a sensing operation.

[0057] The optimal flight trajectory set contains multiple different ideal optimal flight trajectories; since the energy, sensing resources, etc. of the drone are dynamically changing during flight, in this embodiment, the optimal flight trajectory in the optimal flight trajectory set is updated through the preset diffusion model and the preset depth gradient model to obtain a high-quality target flight trajectory that can meet real-time dynamic changes, and at the same time avoid problems such as limitations brought by the single optimization strategy usually adopted in the prior art for trajectory update.

[0058] Furthermore, the UAV flies according to the target flight trajectory, and continuously senses the sensing target in cooperation with the ground physical device during the flight, and sends the synchronized sensing data to the edge server, so that the edge server processes the collected data status and generates a digital twin of the sensing target, thereby reflecting the real-time status of the sensing target.

[0059] The synchronous data sensing method provided by the above embodiments of the present invention can be applied to the sensing system of the UAV-assisted edge network; as Figure 2 shown, the sensing system may include UAVs, ground physical devices, an edge server, and a sensing target. When the system works specifically, the ground physical device and the UAV continuously sense the sensing target, and send the synchronized sensing data to the edge server; the edge server processes the collected data status and generates a digital twin of the sensing target to reflect the real-time status of the sensing target. In the sensing system, use to represent the index set of time.

[0060] In an optional embodiment of the present invention, the above step 12 may include:

[0061] Step 121, determine the index values when the UAV and the ground physical device perform data sensing according to the first sensing parameter and the second sensing parameter;

[0062] Step 122, determine the synchronous data utility evaluation model according to the index values.

[0063] Here, the index values when the UAV and the ground physical device perform data sensing may include average sensing completeness, average sensing accuracy, sensing timeliness, and sensing energy efficiency. Among them, sensing completeness directly affects the completeness and reliability of the synchronous data during the sensing process of the UAV-assisted edge network, and determines whether the sensing system can comprehensively capture environmental information; sensing completeness refers to the sensing mapping ratio of each element attribute of the sensing target by the system, reflecting the overall coverage rate of the sensing target by the system. Sensing accuracy refers to the sensing accuracy of the attributes (such as position, shape, category, etc.) of the sensing target during the sensing process of the sensing system. Sensing timeliness is used to characterize the effectiveness of the sensing data. Since the state of the sensing target changes with time, the effectiveness of the sensing data will rapidly decline over time. Obsolete information not only cannot provide effective support for the sensing system, but may even lead to incorrect decisions. Sensing energy efficiency represents the energy utilization rate of the entire sensing system during the process of sensing the sensing target (that is, the quantity and quality of sensing completed per unit energy consumption).

[0064] Based on average perception completeness, average perception accuracy, perception timeliness, and perception energy efficiency, comprehensively evaluate the quality of UAV perception data from four dimensions; at the same time, construct a synchronous data utility evaluation model based on the index values determined by the first perception parameter and the second perception parameter, and use the maximum value corresponding to the synchronous data utility evaluation model as the target value during subsequent UAV flight trajectory update and adjustment to ensure the accuracy and effectiveness of perception data when flying based on the finally updated and adjusted target flight trajectory.

[0065] In an alternative embodiment of the present invention, when determining the index value based on the first perception data and the second perception data, first construct a perception model for constructing perception data through the first perception data and the second perception data. The perception model includes a UAV perception model and a ground physical device perception model; wherein, the UAV perception model is used to evaluate the perception ability of the UAV, and the ground physical device perception model is used to evaluate the perception ability of the ground physical device; here, the perception ability can be characterized by the perception accuracy, perception latency, perception energy consumption, etc. of the UAV or the ground physical device; here, the specific process of constructing the perception model is as follows:

[0066] (1) UAV perception model:

[0067] Step 1211a, let represent the perception frequency of the UAV at time , then the first perception accuracy of the UAV at time for the perception target can be expressed as:

[0068] ;

[0069] Among them, represents the first perception accuracy; represents the first perception factor, and according to the change of the environment, the value of also changes accordingly; represents the maximum perception range of the UAV ; represents the first distance between the UAV and the perception target at time

[0070] Step 1212a, let be the first time required for the UAV to perform a perception operation, then at time t, the first perception latency of the UAV can be expressed as:

[0071] ;

[0072] Step 1213a, according to the first time and the sensing frequency, the first sensing energy consumption of the UAV at time t can be determined :

[0073] .

[0074] (2) Ground physical device sensing model:

[0075] Step 1211b, let represent the size of the sensing data of the ground physical device at time t ; Use the probability model to evaluate the second sensing accuracy of the ground physical device. At time t, the ground physical device for the sensing target the successful sensing probability is expressed as:

[0076] ;

[0077] Among them, represents the second sensing factor; represents the distance between the ground physical device at time t and the sensing target ; Since the ground physical device can perform multiple sensing operations simultaneously in a sensing process (to increase the possibility of successful sensing), so at time t, the ground physical device for the sensing target the successful sensing probability after multiple sensing operations is expressed as:

[0078] ;

[0079] Among them, represents the number of sensing times performed by the ground physical device at time t for the sensing target , let be the maximum number of sensing times of a ground physical device at a moment, then the constraint condition that the number of sensing times should satisfy is: ;

[0080] Then according to the successful sensing probability after multiple sensing operations, at time, the second sensing accuracy of the ground physical device for the sensing target can be expressed as:

[0081] ;

[0082] Step 1212b, let Represents ground physical equipment The second time required to perform a sensing operation is Moment, ground physical equipment Second perceived delay It can be expressed as:

[0083] ;

[0084] Step 1213b, based on the perceived data size ,exist Ground physical equipment Second perceived energy consumption It is expressed as:

[0085] ;

[0086] in, Represents ground physical equipment Perceive the energy consumption per bit of synchronous data.

[0087] Here, in order to ensure the quality of the perceived target, the ground physical equipment The probability of successful perception should not be lower than a lower limit. Represents ground physical equipment The minimum probability of successful perception is required, and the perception probability should meet the following constraints:

[0088] ; Where N represents the total number of ground physical devices in the perception system.

[0089] In an optional embodiment of the present invention, after the drone and the ground physical device perceive the perceived target, they need to transmit the synchronous perception data to the ground edge server respectively; here, it is necessary to build a corresponding communication model and transmit the data through the communication model; the specific process of building the communication model is as follows:

[0090] Step 1214a, constructing an air-to-ground channel model between the drone and the ground edge server: Since the drone is flying in the air during the mission, the possibility of line-of-sight link communication is greatly increased during the communication with the ground edge server; at the same time, the existence of obstacles such as buildings also leads to the generation of non-line-of-sight links;

[0091] Here, the expected path loss for communication between UAV u and ground edge server k can be determined based on the air-to-ground probability model: It is expressed as:

[0092] ;

[0093] in, , respectively represent the line-of-sight link communication probability and the non-line-of-sight link communication probability between the UAV and the edge server at a certain moment; represents the path loss between the UAV and the edge server under line-of-sight link communication at a certain moment; represents the path loss between the UAV and the edge server under non-line-of-sight link communication at a certain moment; the line-of-sight link probability can be expressed as:

[0094] ;

[0095] wherein, , both represent environmental parameters and change with the environment; represents the included angle degree between the UAV and the edge server at a certain moment; ;

[0096] and can be respectively expressed as:

[0097] ;

[0098] ;

[0099] wherein, represents the carrier frequency of the air-to-ground channel; represents the distance between the UAV and the edge server at a certain moment; is the speed of light; and respectively represent the average additional path loss of the line-of-sight link and the non-line-of-sight link.

[0100] At a certain moment, the first transmission rate at which the UAV uses the air-to-ground channel to transmit the sensed data to the edge server can be expressed as: ;

[0101] wherein, represents the bandwidth of the communication channel, ​Indicates at the moment when the drone and the edge server The signal-to-noise ratio between them can be specifically expressed as: ; where represents Gaussian white noise with a mean of 0, is the edge server at the moment when the received drone The first transmission power of, which can be expressed as: ; where is the second transmission power of the drone, represents the antenna gain of the edge server .

[0102] Therefore, the first transmission delay of the drone at the moment can be expressed as:

[0103] ;

[0104] where represents the data size sensed by the drone for all potential sensing targets.

[0105] Step 1214b, constructing a ground-to-ground channel model: After the ground physical device completes a round of sensing of the sensing target, the ground physical device will transmit the data to the edge server to update the target status information, etc. Here, the th ground physical device transmits the sensed data to the edge server The transmission rate of can be expressed as:

[0106] ;

[0107] where represents the bandwidth of the transmission channel, represents the ground physical device at the moment of transmission power, is the ground physical device The channel gain between and the edge server.

[0108] Furthermore, at the moment, the second transmission delay of the ground physical device can be expressed as:

[0109] ;

[0110] where represents the data size sensed by the ground physical device .

[0111] When the drone and the ground physical devices transmit the sensed data to the edge server, the edge server needs to process the sensed data. The computing delay of the edge server at time t can be expressed as:

[0112] ;

[0113] where, represents the data size transmitted from all ground physical devices and drones to the edge server at time, and represents the computing power of the edge server. Since the edge server is located on the ground and has an infinite energy supply, the energy consumption problem of the edge server is not considered.

[0114] In an optional embodiment of the present invention, based on the above - constructed models, the metric values during data sensing by the drone and the ground physical devices can be specifically expressed as:

[0115] At time, for the sensing target , the sensing completeness can be expressed as:

[0116] ; where, represents the number of element attributes of the th sensing target sensed by the drone or the ground sensing device at time; represents the number of all element attributes of the th sensing target at time. When , it means that all element attributes of the sensing target are successfully sensed, and the sensing completeness is the highest; on the contrary, when , it means that the sensing is completely failed;

[0117] Furthermore, the average sensing completeness of the entire sensing system can be expressed as:

[0118] ;

[0119] At time, the accuracy of the th sensing target can be expressed as:

[0120] ;

[0121] where, and are the accuracy weights sensed by the drone and the ground sensing device respectively, and satisfy the constraint condition: ; Indicates at the moment when the drone for the th perception target; Indicates at the moment when the ground perception device for the th perception target; It can be expressed as: ; where represents the perception success rate of the ground perception device at time slot t for the th perception target;

[0122] Furthermore, the average perception accuracy of the entire perception system can be expressed as:

[0123] .

[0124] At the moment, the perception timeliness of the entire perception system can be expressed as:

[0125] ; where represents the size of the synchronized perception data sensed by the entire system at the moment; represents the transmission and calculation delay, which can be specifically expressed by the following formula:

[0126] .

[0127] At the moment, the perception energy efficiency of the entire perception system can be expressed as:

[0128] ; where represents the size of the data sensed by the drone and the ground perception device at the moment; represents the total energy consumed by the entire synchronized data perception, transmission, and calculation at the moment.

[0129] In an optional embodiment of the present invention, based on the above index values, a synchronized data utility evaluation model can be constructed, which can be specifically expressed as:

[0130] ;

[0131] where represents the average perception completeness, represents the average perception accuracy, represents the perception timeliness, Denote the perceived energy efficiency respectively represent the weight coefficients of the corresponding index values, and .

[0132] For all potential sensing targets, the optimization goal is to maximize the completeness and accuracy of the sensing targets and minimize the latency and energy consumption to improve the system timeliness and energy efficiency, which can be specifically expressed as: .

[0133] Quantify the quality and value of the synchronized sensing data in the UAV-assisted edge network by the synchronized data utility evaluation model to better and more comprehensively sense the sensing targets; comprehensively consider the factors of target completeness, target accuracy, target timeliness, and target energy efficiency, optimize the system resource allocation, and improve the sensing efficiency and network synchronization performance.

[0134] In an optional embodiment of the present invention, step 13 may include:

[0135] Step 131: Generate an initial flight trajectory set of the UAV according to a preset random generation algorithm.

[0136] Here, the initial flight trajectory generated based on the preset random generation algorithm can be specifically expressed as:

[0137] ; Assume that each UAV flies 100 steps (corresponding to 100 moments), and each moment corresponds to a flight coordinate. Then an initial flight trajectory corresponds to 100 flight coordinates, that is, an initial flight trajectory can be obtained by connecting 100 flight coordinates in chronological order;

[0138] Here, the initial flight coordinates of the UAV at each moment can be randomly generated through the following preset random generation algorithm:

[0139] ;

[0140] Among them, respectively represent the initial coordinate values of the UAV on the X-axis, Y-axis, and Z-axis at time t, which can be specifically expressed as:

[0141] ;

[0142] ;

[0143] ;

[0144] Among them, , , respectively represent the boundaries in the three directions of the X-axis, Y-axis, and Z-axis to prevent the generated initial solution from exceeding the boundaries; represents a random number between 0 and 1, which is used to randomize the initial coordinates; Since represents a random number, and multiple different coordinate values may be generated at the corresponding moment, so multiple different initial flight trajectories can be obtained.

[0145] In an alternative embodiment of the present invention, the above step 13 further includes:

[0146] Step 132, according to a preset initialization trajectory algorithm, perform an update process on the initial flight trajectory set to obtain an updated flight trajectory set, and the updated flight trajectories in the updated flight trajectory set correspond one-to-one with the initial flight trajectories in the initial flight trajectory set.

[0147] Here, based on the preset initialization trajectory algorithm, the initial flight trajectories in the initial flight trajectory set are updated. Specifically, the initial flight coordinates in each initial flight trajectory are updated to obtain the optimal flight coordinates, so as to ensure that the optimal flight trajectory can be obtained subsequently. Here, the update process of the preset initialization trajectory algorithm mainly includes two parts: the exploration stage and the exploitation stage. Among them, the exploration stage of the algorithm mainly reflects its powerful global search ability, fast convergence, and flexibility. Through extensive search, the exploration stage can effectively avoid local optima and ensure finding multiple potential good solutions (that is, the optimal flight coordinates), especially outstanding in dealing with complex, multi-objective optimization problems. The exploitation stage of the algorithm is mainly to further refine and optimize the solutions obtained in the exploration stage to improve the quality and accuracy of the solutions. In the exploitation stage, the algorithm converges to the global optimal solution through local search and fine-tuning, and usually uses heuristic methods or more strict constraint conditions to improve the performance of the solutions. Its advantage is that it can deeply optimize the key parts by concentrating resources, narrow the search range, and improve the accuracy of the solutions. In addition, the exploitation stage can effectively avoid unnecessary calculations, reduce the search space, thereby accelerating the convergence process and improving the optimization efficiency.

[0148] In an alternative embodiment of the present invention, the above step 132 may include:

[0149] Step 1321, in the exploration stage, according to the formula: perform an update process on each initial flight trajectory in the initial flight trajectory set once to obtain the first flight trajectories corresponding one-to-one to the initial flight trajectories; where represents the initial flight trajectory; represents the first flight trajectory; 、 respectively represent 、 The weight coefficients, all of which are random numbers between 0 and 1; represents the coordinate vector difference between the optimal initial flight coordinate in the initial flight trajectory set and the initial flight coordinate at the current moment; represents the coordinate vector difference between a randomly selected coordinate from the first three optimal initial flight coordinates in the initial flight trajectory set, the mean of all initial flight coordinates, and the median of the initial flight coordinates and the initial flight coordinate at the current moment; these two coordinate vector differences are used to guide the update of the initial flight coordinate at the current moment, enabling the algorithm to find the optimal flight trajectory as soon as possible; is the weight coefficient of represents a normal distribution with a mean of 0 and a variance of 1; represents the coordinate vector difference between the UAV at a random position and the initial flight coordinate at the current moment; here, , and can be expressed as:

[0150] is the weight coefficient of fel; represents a normal distribution with a mean of 0 and a variance of 1; fel represents the difference in distance between the UAV at a random position and the current position; here, , and fel can be expressed as:

[0151] ;

[0152] ;

[0153] ;

[0154] Among them, represents a randomly selected coordinate from the first three optimal initial flight coordinates in the initial flight trajectory set, the mean of all initial flight coordinates, and the median of the initial flight coordinates, represents a random position randomly selected in the initial flight trajectory set, represents the optimal initial flight coordinate in the initial flight trajectory set, and this optimal initial flight coordinate can be obtained by calculating the value of each initial flight coordinate, and determining the coordinate corresponding to the maximum value as the optimal initial flight coordinate.

[0155] Step 1322, in the development stage, the first flight trajectory can be processed for secondary update according to the formula: to obtain an updated flight trajectory set composed of updated flight trajectories corresponding one-to-one to the first flight trajectory; among them, Indicates updating the flight coordinates; Indicates the spiral flight parameters of the UAV; Indicates the Levy flight parameters of the UAV; M represents the adaptive parameter; Indicates a random factor, randomly taking values between 0 and 1, and is used to select the position update formula.

[0156] Here, the spiral flight parameters of the UAV Indicates that in order to obtain the optimal solution faster, the UAV updates its position in a spiral flight mode. The spiral flight parameters can be specifically expressed as:

[0157] ;

[0158] Among them, L represents the spiral contraction factor, which determines the convergence speed during spiral flight, and Z represents the spiral angle parameter, which determines the updated position during spiral flight; and Can be specifically expressed as:

[0159] ;

[0160] ;

[0161] Among them, Is the current iteration number, Is the maximum iteration number. Indicates the adaptive parameter and can be expressed as:

[0162] ;

[0163] The Levy flight parameters of the UAV Can be expressed as:

[0164] ;

[0165] Among them, Is the constant 1.5, u represents the random scale of the Levy flight step, that is, the basic size of the step, and v represents the heavy-tailed characteristic of the Levy distribution, which controls the possibility of long-distance jumps in the step. Both obey a specific normal distribution, Can be respectively expressed as:

[0166] ;

[0167] ;

[0168] Among them, Indicates a normal distribution with a mean of 0 and a variance of 1.

[0169] In an alternative embodiment of the present invention, the above step 13 further includes:

[0170] Step 133. Determine the optimal flight trajectory set according to the synchronization data utility evaluation model, the initial flight trajectory set, and the updated flight trajectory set.

[0171] Specifically, the above step 133 may include:

[0172] Step 1331. Determine the initial utility value corresponding to each initial flight trajectory in the initial flight trajectory set and the updated utility value corresponding to each updated flight trajectory in the updated flight trajectory set according to the synchronization data utility evaluation model;

[0173] Step 1332. Screen the initial flight trajectory set and the updated flight trajectory set according to the initial utility value and the updated utility value to obtain the optimal flight trajectory set.

[0174] In this embodiment, for the initial flight trajectory before update and the updated flight trajectory after update, the initial utility value corresponding to the initial flight trajectory before update and the updated utility value corresponding to the updated flight trajectory after update are respectively determined through the synchronization data utility evaluation model; if the updated utility value corresponding to the updated flight trajectory after update is greater than the initial utility value corresponding to the initial flight trajectory before update, then determine the updated flight trajectory as the optimal flight trajectory; otherwise, determine the initial flight trajectory as the optimal flight trajectory; the above process can be expressed as:

[0175] ;

[0176] Wherein, represents the optimal flight trajectory; represents the updated utility value corresponding to the updated flight trajectory after update; represents the initial utility value corresponding to the initial flight trajectory before update.

[0177] In an alternative embodiment of the present invention, the above step 14 may include:

[0178] Step 141. Perform forward diffusion noise addition and reverse diffusion noise reduction processing on the optimal flight trajectories in the optimal flight trajectory set in sequence according to the preset diffusion model to obtain the flight trajectories after diffusion processing;

[0179] Step 142. Determine the trajectory adjustment actions of the flight trajectories after diffusion processing according to the preset depth gradient model;

[0180] Step 143. Adjust the flight trajectories after diffusion processing according to the trajectory adjustment actions to obtain the target flight trajectories.

[0181] In this embodiment, the data corresponding to the optimal flight trajectory in the optimal flight trajectory set is modeled through a preset diffusion model, so as to generate a more optimized, smooth and constraint-compliant flight trajectory. The diffusion process perturbs the optimal flight trajectory with noise and reconstructs and optimizes the flight trajectory through the reverse process, enabling the UAV to generate a more efficient and adaptable flight trajectory in future tasks. Its role is to enhance the generalization ability of flight trajectory planning, reduce the dependence on single optimization, and generate a better flight trajectory by combining historical experience; thereby improving the perception accuracy, reducing the energy consumption, and enhancing the adaptability of the UAV in a dynamic environment.

[0182] Here, the optimal flight trajectory is represented by the original trajectory data and noise is gradually added to the original trajectory data until it is finally transformed into a standard Gaussian distribution. For the given original trajectory data , the forward diffusion process can be expressed as:

[0183] ;

[0184] where represents the trajectory data after noise addition in the th round; is a factor controlling the attenuation of the original trajectory data, which can be expressed as , is a preset noise increase parameter; represents the noise intensity; represents the identity matrix, indicating that the noise in each dimension is independently and identically distributed; represents the multi-dimensional normal distribution, indicating the state transition at each step in the diffusion process.

[0185] After rounds of noise addition, the noise version trajectory at any time can be expressed as:

[0186] ;

[0187] where ;

[0188] After forward diffusion and noise addition, it is necessary to gradually denoise from the standard Gaussian noise trajectory data to restore the original trajectory data . The key to the reverse denoising process is to train a neural network to predict the denoised trajectory, which can be specifically expressed as:

[0189]

[0190] where denotes the predicted denoised mean, denotes the predicted denoised variance, denotes the neural network parameters (which can be obtained through training). Here, can be set as a fixed value, so the denoising network only needs to predict ; Using Bayesian derivation, we can get:

[0191] ;

[0192] where, denotes the trajectory data corresponding to the original trajectory after denoising; denotes the noise estimation value output by the neural network; denotes the standard deviation of the random noise added during the reverse denoising process, which is used to ensure the randomness of denoising and make the sampling process diverse. This formula means that at each step of denoising, the noise is subtracted from the current standard Gaussian noise trajectory data to gradually restore the trajectory data corresponding to the original trajectory.

[0193] In an optional embodiment of the present invention, in order to train the neural network , the mean square loss can be minimized for training, which can be specifically expressed as:

[0194] ;

[0195] where, denotes the noise added during the forward diffusion process.

[0196] Then the specific training steps are explained as follows:

[0197] Step 1411, randomly sample the original trajectory data from the historical trajectories obtained by the intelligent optimization algorithm ;

[0198] Step 1412, randomly select a diffusion time step ;

[0199] Step 1413, generate noise , and calculate ;

[0200] Step 1414, use the neural network to predict the noise ;

[0201] Step 1415, calculate the loss and update the neural network parameters .

[0202] In an alternative embodiment of the present invention, based on the forward diffusion noise addition and reverse diffusion noise reduction processes performed on the optimal flight trajectories in the optimal flight trajectory set by a preset diffusion model, adjustment can be made through a preset deep gradient model to enable the drone to adapt to changes in a dynamic environment and optimize the synchronous data utility evaluation model. Preferably, the preset deep gradient model can be a Deep Deterministic Policy Gradient (DDPG) network model. The specific adjustment process is as follows:

[0203] Step 1421: Sample the trajectory data corresponding to the original trajectory after denoising and convert it into a state sequence: ;

[0204] Step 1422: Initialize the DDPG network model;

[0205] Step 14221: Initialize the Actor network ( ): Input the state , and output the action , which is used for adjusting the drone trajectory.

[0206] Step 14222: Initialize the Critic network ( ): Input the state and the action , and output the Q value, which is used for evaluating the utility.

[0207] Step 14223: Initialize the experience replay buffer , which stores past trajectory data.

[0208] Step 14224: Initialize the target Actor and Critic networks for delayed update to stabilize the training.

[0209] Step 1423: Interaction and data collection:

[0210] First, select the initial state from the trajectories generated by the diffusion model, and then the Actor network generates a trajectory adjustment action, which can be specifically expressed as:

[0211] ;

[0212] where is the exploration noise to ensure the diversity of the policy. Execute the action , and update the drone trajectory, which can be expressed as:

[0213] .

[0214] Step 1424, calculate the reward (that is, the synchronous data utility evaluation model ), the calculation method can be expressed as:

[0215] ;

[0216] Step 1425, store the data into the experience pool .

[0217] Step 1426, update the Actor network and the Critic network, and sample from the experience replay pool to obtain a batch of data .

[0218] Step 1427, calculate the Q value of the target, which can be expressed as:

[0219]

[0220] where, is the discount factor, and are the target networks for delayed update.

[0221] Step 1428, update the Critic network: update the Critic network through training with the mean squared error loss, which can be specifically expressed as:

[0222] ;

[0223] Update the Actor network by maximizing the Q value of the trajectory adjustment strategy through policy gradient training, which can be specifically expressed as:

[0224] ;

[0225] Update the target network, which can be specifically expressed as:

[0226] ;

[0227] where, takes a value of 0.005, and after the DDPG training converges, the final optimized target flight trajectory is obtained.

[0228] The synchronous data perception method based on edge multi-point collaboration provided by the above embodiments of the present invention, aiming at the perception requirements of unmanned aerial vehicles (UAVs) in the edge computing environment, obtains the first perception parameter and the second perception parameter when the UAV and ground physical devices perform data perception on the perception target; determines the synchronous data utility evaluation model during data perception according to the first perception parameter and the second perception parameter; determines the optimal flight trajectory set of the UAV according to the preset initialization trajectory algorithm and the synchronous data utility evaluation model; updates the optimal flight trajectory according to the preset diffusion model and the preset depth gradient model to obtain the target flight trajectory; synchronously perceives the data information of the target by the ground physical device and the UAV based on the target flight trajectory and feeds it back to the edge server; comprehensively evaluates the quality of the UAV's perceived data from four dimensions of perception accuracy, perception completeness, perception timeliness, and perception energy efficiency based on the optimization strategy of the synchronous data utility evaluation model. Compared with the prior art that only focuses on the optimization goals of the shortest path or the lowest energy consumption, the solution of the present invention takes into account both data quality and computational resource utilization, enabling the UAV to dynamically adjust the trajectory to maximize the data synchronization utility when performing the perception task, and improving the availability and decision-making value of the data in the edge computing environment. Through the multi-stage optimization strategy, accurate perception of the real-time changing perception target can be achieved, and the accuracy and coverage of data perception can be improved; in addition, this solution can achieve efficient and low-energy UAV trajectory planning in complex dynamic environments, providing better technical support for the application of UAVs in scenarios such as disaster monitoring, smart cities, and industrial inspections.

[0229] As Figure 3 shown, an embodiment of the present invention also provides a synchronous data perception 30 based on edge multi-point collaboration, including:

[0230] An acquisition module 31, configured to acquire the first perception parameter when the UAV performs data perception on the perception target during flight and the second perception parameter when the ground physical device performs data perception on the perception target;

[0231] A processing module 32, configured to determine the synchronous data utility evaluation model when the UAV and the ground physical device perform data perception according to the first perception parameter and the second perception parameter; determine the optimal flight trajectory set of the UAV according to the preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set includes multiple different optimal flight trajectories; update the optimal flight trajectory in the optimal flight trajectory set according to the preset diffusion model and the preset depth gradient model to obtain the target flight trajectory; synchronously perceive the data information of the target by the ground physical device and the UAV based on the target flight trajectory and feed it back to the edge server.

[0232] It should be noted that this device corresponds to the method for synchronously perceiving data based on edge multi-point collaboration of the above data. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0233] As Figure 4 shown, an embodiment of the present invention further provides an electronic device 50, including: a memory 51 for storing one or more computer programs; one or more processors 52 for executing one or more computer programs. When the computer programs are run by the processor, the method for synchronously perceiving data based on edge multi-point collaboration 10 as described above is executed. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects. The electronic device 50 is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown in the present invention, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described and / or claimed in the present invention.

[0234] As Figure 5 shown, the electronic device 50 is manifested as a computing device, or a computer system, and may include a CPU 501 (computing unit), which can execute various appropriate actions and processes according to the computer programs stored in the ROM 502 (read-only memory) or the computer programs loaded from the storage unit 508 into the random access RAM 503 (memory). In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. The I / O interface 505 (input / output interface) is also connected to the bus 504.

[0235] Multiple components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disc, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0236] The CPU 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the CPU 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The CPU 501 executes the various methods and processes described above. For example, in some embodiments, the edge multi-point collaboration-based synchronous data perception method 10 can be implemented as a computer software program that is tangibly included in a computer-readable storage medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the CPU 501, one or more steps of the edge multi-point collaboration-based synchronous data perception method 10 described above can be executed. Alternatively, in other embodiments, the CPU 501 can be configured to execute the edge multi-point collaboration-based synchronous data perception method 10 in any other suitable manner (e.g., by means of firmware).

[0237] Embodiments of the present invention also provide a computer-readable storage medium that stores instructions which, when run on a computer, cause the computer to execute the edge multi-point collaboration-based synchronous data perception method 10 as described above. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0238] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraint conditions of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0239] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0240] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be in electrical, mechanical or other forms.

[0241] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0242] In addition, each functional unit in various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0243] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. And the aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0244] In addition, it should be noted that in the devices and methods of the present invention, obviously, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present invention. Moreover, the steps of performing the above series of processes can naturally be executed in chronological order according to the described order, but it is not necessary to be executed in chronological order. Some steps can be executed in parallel or independently of each other. For those of ordinary skill in the art, it is understandable that all or any steps or components of the method and device of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in the form of hardware, firmware, software, or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0245] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a well-known general-purpose device. Therefore, the object of the present invention can also be achieved only by providing a program product containing program code for implementing the method or device. That is to say, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be noted that in the devices and methods of the present invention, obviously, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present invention. Moreover, the steps of performing the above series of processes can naturally be executed in chronological order according to the described order, but it is not necessary to be executed in chronological order. Some steps can be executed in parallel or independently of each other.

[0246] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A synchronous data perception method based on edge multi-point collaboration, characterized in that, Including: Obtaining a first sensing parameter when the drone performs data sensing on a sensing target during flight and a second sensing parameter when a ground physical device performs data sensing on the sensing target; Determining a synchronous data utility evaluation model for the drone and the ground physical device to perform data sensing according to the first sensing parameter and the second sensing parameter; Determining an optimal flight trajectory set of the drone according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set includes multiple different optimal flight trajectories; Updating the optimal flight trajectories in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory; Based on the drone of the target flight trajectory and collaborating with the ground physical device to synchronously sense the data information of the sensing target, and feeding back the data information to the edge server; Among them, determining a synchronous data utility evaluation model for the drone and the ground physical device to perform data sensing according to the first sensing parameter and the second sensing parameter includes: Determining index values for the drone and the ground physical device to perform data sensing according to the first sensing parameter and the second sensing parameter; Determining the synchronous data utility evaluation model according to the index values; The synchronous data utility evaluation model is constructed by the following formula: ； Among them, represents the value of the perception completeness index, represents the value of the perception accuracy index, represents the value of the perception timeliness index, represents the value of the perception energy efficiency index, among which, respectively represent the weight coefficients of the corresponding index values, and .

2. The synchronous data perception method based on edge multi-point collaboration according to claim 1, wherein Determining an optimal flight trajectory set of the drone according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model includes: Generating an initial flight trajectory set of the drone according to a preset random generation algorithm; Performing an update process on the initial flight trajectory set according to the preset initialization trajectory algorithm to obtain an updated flight trajectory set, where the updated flight trajectories in the updated flight trajectory set correspond one-to-one to the initial flight trajectories in the initial flight trajectory set; Determining the optimal flight trajectory set according to the synchronous data utility evaluation model, the initial flight trajectory set, and the updated flight trajectory set.

3. The synchronous data perception method based on edge multi-point collaboration according to claim 2, wherein Performing an update process on the initial flight trajectory set according to the preset initialization trajectory algorithm to obtain an updated flight trajectory set, including: According to the formula: Perform an update process on each initial flight trajectory in the set of initial flight trajectories to obtain a first flight trajectory corresponding one-to-one to the initial flight trajectory; where represents the initial flight coordinates; represents the first flight coordinates; , respectively represent , the weight coefficients of, and are both random numbers between 0 and 1; is the weight coefficient of fel; represents a normal distribution with a mean of 0 and a variance of 1; represents the coordinate vector difference between the optimal initial flight coordinates and the initial flight coordinates at the current moment in the set of initial flight trajectories; represents the coordinate vector difference between a randomly selected coordinate from the first three optimal initial flight coordinates, the mean of all initial flight coordinates, and the median of the initial flight coordinates in the set of initial flight trajectories and the initial flight coordinates at the current moment; represents the coordinate vector difference between the UAV at a random position and the initial flight coordinates at the current moment; According to the formula: Perform a secondary update process on the first flight trajectory to obtain an updated flight trajectory set composed of updated flight trajectories corresponding one-to-one to the first flight trajectory; where represents the updated flight coordinates; represents the spiral flight parameters of the UAV; represents the Levy flight parameters of the UAV; M represents the adaptive parameter; represents a random factor, randomly taking values between 0 and 1, and is used to select the position update formula.

4. The synchronous data perception method based on edge multi-point collaboration according to claim 2, characterized in that, Determining the optimal flight trajectory set according to the synchronous data utility evaluation model, the initial flight trajectory set, and the updated flight trajectory set, including: Determining an initial utility value corresponding to each initial flight trajectory in the initial flight trajectory set and an updated utility value corresponding to each updated flight trajectory in the updated flight trajectory set according to the synchronous data utility evaluation model; Performing a screening process on the initial flight trajectory set and the updated flight trajectory set according to the initial utility value and the updated utility value to obtain the optimal flight trajectory set.

5. The synchronous data perception method based on edge multi-point collaboration according to claim 3, characterized in that, Updating the optimal flight trajectories in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain a target flight trajectory, including: Perform forward diffusion noise addition and reverse diffusion noise reduction processing on the optimal flight trajectories in the optimal flight trajectory set according to the preset diffusion model, so as to obtain the flight trajectories after diffusion processing; Determine the trajectory adjustment actions of the flight trajectories after diffusion processing according to the preset depth gradient model; Adjust the flight trajectories after diffusion processing according to the trajectory adjustment actions to obtain the target flight trajectories.

6. A synchronous data perception device based on edge multi-point collaboration, characterized in that, Comprising: An acquisition module, configured to acquire a first sensing parameter when the unmanned aerial vehicle performs data sensing on a sensing target during flight and a second sensing parameter when a ground physical device performs data sensing on the sensing target; A processing module, configured to determine a synchronous data utility evaluation model when the unmanned aerial vehicle and the ground physical device perform data sensing according to the first sensing parameter and the second sensing parameter; determine an optimal flight trajectory set of the unmanned aerial vehicle according to a preset initialization trajectory algorithm and the synchronous data utility evaluation model, where the optimal flight trajectory set includes multiple different optimal flight trajectories; update the optimal flight trajectories in the optimal flight trajectory set according to a preset diffusion model and a preset depth gradient model to obtain target flight trajectories; the unmanned aerial vehicle based on the target flight trajectories cooperates with the ground physical device to synchronously sense data information of the sensing target, and feeds back the data information to the edge server; Wherein, determining the synchronous data utility evaluation model when the unmanned aerial vehicle and the ground physical device perform data sensing according to the first sensing parameter and the second sensing parameter includes: Determine the index values when the unmanned aerial vehicle and the ground physical device perform data sensing according to the first sensing parameter and the second sensing parameter; Determine the synchronous data utility evaluation model according to the index values; The synchronous data utility evaluation model is constructed by the following formula: ； Among them, represents the value of the perception completeness index, represents the value of the perception accuracy index, represents the value of the perception timeliness index, represents the value of the perception energy efficiency index, where respectively represent the weight coefficients of the corresponding index values, and .

7. A computing device, characterized in that, Comprising: A memory, configured to store one or more programs; One or more processors, configured to execute the one or more programs to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A program is stored in the computer-readable storage medium, and when the program is executed by the processor, the method according to any one of claims 1 to 5 is implemented.

Images