Multi-unmanned aerial vehicle power transmission line inspection method and system based on compressed communication and fedadmm optimization
By using compressed communication and FedADMM optimization, the problems of power consumption and communication power loss in multi-UAV inspection were solved, realizing efficient and safe UAV inspection, reducing energy consumption and improving inspection efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SOUTHEAST UNIV
- Filing Date
- 2025-07-10
- Publication Date
- 2026-06-26
AI Technical Summary
In multi-drone power line inspections, each drone has limited power. Long-term, wide-range flights lead to power consumption, which becomes a key factor limiting inspection efficiency, increasing inspection costs and potentially causing drones to lose contact or crash. Furthermore, existing communication methods result in excessive power loss.
A method based on compressed communication and FedADMM optimization is adopted to reduce communication power consumption through local update of UAV data, compressed transmission and total-end aggregation process. This includes UAV determining the energy consumption objective function, compressed transmission and total-end aggregation, and designing a special FedADMM algorithm to ensure a globally consistent solution.
While ensuring a globally consistent solution, it significantly reduces power consumption caused by communication, improves the efficiency of UAV inspection, reduces transmission volume, lowers energy consumption, and avoids the risk of insufficient power for UAVs.
Smart Images
Figure CN120640249B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of information transmission and intelligent control, specifically to a multi-UAV power line inspection technology based on compressed communication and FedADMM optimization. Background Technology
[0002] In modern power systems, transmission lines, as critical infrastructure for power transmission, are essential for ensuring the reliability of power supply through their safe and stable operation. With the continuous expansion of power grids and the increasingly wide coverage of transmission lines, multi-UAV transmission line inspection technology, with its advantages of high efficiency, flexibility, and ability to cover complex terrain, has gradually become a research hotspot and development trend in the field of transmission line inspection. However, each UAV has limited battery power, while inspection tasks often require long-term, wide-area flight operations, making power consumption a key factor limiting the efficiency of UAV inspections. These problems not only increase inspection costs but may also lead to UAVs being unable to complete tasks due to insufficient power, or even risks such as loss of contact or crashes in remote areas, resulting in equipment damage and data loss. Therefore, the requirement for low communication power consumption while ensuring accurate UAV data transmission is extremely important.
[0003] For achieving a globally consistent solution across multiple drones, this invention proposes a FedADMM algorithm. Under this algorithm, each drone relies solely on its own information and that of the central terminal. It updates its parameters locally based on the central terminal's information and sends its information back to the central terminal. The central terminal then aggregates the data and sends it to each drone, thus achieving a final consistent solution. However, this framework often results in significant power consumption due to the large amount of data transmitted, making it highly insecure. Summary of the Invention
[0004] Purpose of the invention: The purpose of this invention is to propose a multi-UAV power line inspection method and system based on compressed communication and FedADMM optimization, which can reduce the power loss caused by communication while ensuring that each UAV obtains a globally consistent solution.
[0005] Technical solution: To achieve the above objectives, the present invention adopts the following technical solution:
[0006] In a first aspect, the present invention provides a multi-UAV power line inspection method based on compressed communication and FedADMM optimization, comprising UAV data local updating, compressed transmission, and central aggregation processes; the UAV data local updating includes:
[0007] The i-th drone determines its own energy consumption objective function;
[0008] The data transmitted by the i-th drone satisfy σi Greater than 0 indicates The importance of; Let be the physical parameters required for communication by the i-th UAV. For the accumulated error, α i For constant weights, f i (·) represents the energy consumption objective function. These are the initial physical parameters. Represents the gradient;
[0009] The central terminal randomly selects a subset of drones to calculate global parameters. And broadcast to the drones in this subset, among which This represents the sum of all weights, where m represents the total number of drones. These are compressed variables and compressed data, respectively.
[0010] Received The i-th drone first internally updates the error tolerance value. Make it satisfy in
[0011] Given a range of [0.5, 1), ensure the error tolerance value approaches 0; then update the compression variables. in Given a positive real number, update the communication parameters. Make it satisfy in ‖·‖ represents the 2-norm; the cumulative error is updated last. and data transmission and compressed data Where C(·) is the compression function; and for drones that do not receive a signal, its parameters remain unchanged;
[0012] The aggregation process at the central terminal includes: after receiving the transmission data from the selected subset of UAVs, the central terminal updates...
[0013] This represents the final data from the drone.
[0014] Preferably, the central terminal generates α based on the priority of each drone's mission. i The higher the priority, the better. i The larger, and α i >0.
[0015] As a preferred option, the central terminal randomly selects a subset of drones in the following manner. in All drones are selected, while a subset of drones is randomly selected when k>0. The goal is to ensure that all drones are selected at least once in each sequence containing s0 sets. k and k0 represent the number of local iterations for the drone and the communication cycle between the drone and the central terminal, respectively. This indicates rounding up to the nearest integer.
[0016] Preferably, the compression function C(x) satisfies E c [‖C(x)-x‖ 2 ]≤r‖x‖ 2 ,r∈[0,1),E C [·] represents the expectation of the internal randomness of the random compression operator C.
[0017] Secondly, the present invention provides a multi-UAV power line inspection method based on compressed communication and FedADMM optimization, which differs from the method in the first aspect in that: the global parameters are calculated by the central terminal according to the following formula: The drone's local parameters include To compress variables, For the new compressed data; update the local parameters according to the following formula: Internal update error tolerance value Make it satisfy Next, update the compression variables. And update communication parameters Make it satisfy in Then update the cumulative error. and data transmission and compressed data Finally, update the new local parameters. and γ is a positive real number.
[0018] Thirdly, the present invention provides a multi-UAV power line inspection system based on compressed communication and FedADMM optimization, including multiple UAV nodes, which form a transmission network with the central terminal to constitute a UAV swarm; each UAV node includes a sensor for monitoring the UAV status and a client data transmission module, the client data transmission template includes a client data local update unit and a compressed transmission unit, and the central terminal is provided with an aggregation unit;
[0019] The client data local update unit is used for:
[0020] The i-th drone determines its own energy consumption objective function;
[0021] The data transmitted by the i-th drone satisfy σ i Greater than 0 indicates The importance of; Let be the physical parameters required for communication by the i-th UAV. For the accumulated error, α i For constant weights, f i (·) represents the energy consumption objective function. These are the initial physical parameters. Represents the gradient;
[0022] Receive global parameters calculated by the central terminal The i-th drone, of which This represents the sum of all weights, where m represents the total number of drones. These are compressed variables and compressed data, respectively; first, the error tolerance value is updated internally. Make it satisfy in Given a range of [0.5, 1), ensure the error tolerance value approaches 0; then update the compression variables. in Given a positive real number, update the communication parameters. Make it satisfy in ‖·‖ represents the 2-norm; the cumulative error is updated last. and data transmission and compressed data Where C(·) is the compression function; and for drones that do not receive a signal, its parameters remain unchanged;
[0023] The compressed transmission unit is used to: when communicating with the central terminal, the i-th UAV will transmit data... Compress to Then compress the data and compressed variables Transmitted to the central terminal;
[0024] The aggregation unit is configured to: after receiving the transmission data from the selected subset of UAVs, update... This represents the final data from the drone.
[0025] Fourthly, the present invention provides a multi-UAV power line inspection system based on compressed communication and FedADMM optimization, which differs from the system in the third aspect in that: in the aggregation unit, the global parameters are calculated by the central terminal according to the following formula:
[0026] The local update unit for client data includes the following local parameters for the drone: To compress variables, For the new compressed data; update the local parameters according to the following formula: Internal update error tolerance value Make it satisfy Next, update the compression variables. And update communication parameters Make it satisfy in Then update the cumulative error. and data transmission and compressed data Finally, update the new local parameters. and γ is a positive real number.
[0027] Fifthly, the present invention provides a multi-UAV power line inspection system based on compressed communication and FedADMM optimization, comprising multiple UAV nodes, which form a transmission network with the central terminal to constitute a UAV swarm;
[0028] Each UAV node includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of local updating and compressed transmission of UAV data in the multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization described in the first or second aspect.
[0029] The central terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the central terminal aggregation step in the multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization described in the first or second aspect.
[0030] In a sixth aspect, the present invention provides a computer system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of local updating and compressed transmission of UAV data, or the step of central aggregation, in the multi-UAV power line inspection method based on compressed communication and FedADMM optimization described in the first or second aspect.
[0031] In a seventh aspect, the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of local updating and compressed transmission of UAV data, or the step of central aggregation, in the multi-UAV power line inspection method based on compressed communication and FedADMM optimization described in the first or second aspect.
[0032] Beneficial Effects: This invention proposes a multi-UAV power line inspection method based on compressed communication and FedADMM optimization. The information transmitted by the UAVs is compressed, and a special FedADMM algorithm is designed to successfully reduce communication volume without compromising the effectiveness of the original algorithm. Since this invention is based on an improved FedADMM algorithm and is a distributed algorithm, it solves the problem of excessive load on traditional transmission methods when the data is transmitted close to the central server node, compared to general centralized algorithms. Experimental analysis confirms that this invention simultaneously ensures consistency and efficiency. Attached Figure Description
[0033] Figure 1 This is a schematic diagram illustrating an application scenario of an embodiment of the present invention.
[0034] Figure 2 This is a flowchart of a method according to an embodiment of the present invention.
[0035] Figure 3 This is an experimental simulation diagram from an embodiment of the present invention.
[0036] Figure 4 The figure shows the simulation results in an embodiment of the present invention. Detailed Implementation
[0037] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.
[0038] Example 1
[0039] like Figure 1 The image shows an application scenario of a multi-UAV power line inspection method based on compressed communication and FedADMM optimization provided by an embodiment of the present invention. Using the method of this embodiment, the ground control station only needs to aggregate the compressed parameters to know the overall optimal parameter results and adjust the communication parameters in real time in order to better reduce the amount of transmission and reduce energy consumption.
[0040] like Figure 2As shown in the figure, this invention provides a multi-UAV power line inspection method based on compressed communication and FedADMM optimization. First, multiple UAVs determine their energy consumption objective function and update local parameters in real time for subsequent compression processing. Then, the UAVs compress and transmit data to the ground control station (central terminal) to reduce communication overhead. Finally, the ground control station (central terminal) performs an aggregation process, aggregating the data sent by the UAVs and broadcasting its own information to the UAVs to achieve a final average consistency value for electricity consumption information. The specific process is as follows:
[0041] 1. Local updates of UAV data: Local updates of UAV data mainly include measuring the energy consumption objective function, generating transmission weights, and updating local data using the FedADMM optimization algorithm.
[0042] Step 101: Measure the energy consumption objective function: The i-th drone determines its own energy consumption objective function, expressed as f, based on its mission completion status, remaining battery power, and inspection distance. i (w). For example, the objective function could be Where d i The number of samples (e.g., a positive integer in [50, 100]), t represents the t-th sample, and <, ·> represent the inner product between vectors. For an n-dimensional column vector, is a real number, and w represents the physical parameters that need to be communicated, including the drone's current speed, acceleration, remaining battery power, and other status information in various directions.
[0043] Step 102 Transmission Weight Generation: The i-th drone will generate a constant denoted as α. i The weights are generated as follows:
[0044] The central server generates α based on the priority of each drone's mission. i Generally, the higher the priority, the better. i The larger, and α i >0.
[0045] Then, for the k-th iteration, the data used for transmission is denoted as... satisfy Where σ i Representation of real numbers greater than 0 The importance of; Let be the physical parameters of the i-th drone, such as its speed. For the accumulated error, in f i (w) in The gradient at that point.
[0046] Step 103 Local Parameter Update: The central terminal will randomly select n drones to form a subset. in k0 represents the communication cycle (number of communication intervals) between the drone and the central terminal, and then the data transmitted by each drone... Compress as Calculate global parameters in Let m represent the sum of all weights and m represent the total number of drones. The selection process involves broadcasting the message to the drones in this subset via a wireless communication module, using the following method:
[0047] Select all drones, and randomly select a subset of drones when k>0, then satisfy the following sequence in each group containing s0 sets. Ensure that each drone is selected at least once.
[0048] Received Each drone i will execute the FedADMM optimization algorithm to internally update the error tolerance value. Make it satisfy in Given a range of [0.5, 1), ensure the error tolerance value approaches 0; then update the compression variables. in Update the communication parameters using a positive real number. Make it satisfy in f i (w) in The gradient at the given point is used to update the cumulative error at the end. as well as C(x) represents a compression function, such as a floor function.
[0049] For drone i that did not receive a signal, its parameters... It remains unchanged.
[0050] 2. Data Compression Process: This process involves the drone compressing local information. At this point, all data has been updated to the latest format, and only the communication compression algorithm needs to be applied. The process is as follows:
[0051] Step 201 in the set The i-th drone in the process will process the uploaded data. Compress the data and update it internally. Finally obtained The transmitted data is sent to the central terminal. k0 represents the communication cycle or communication interval; the compression function C(x) should satisfy E C [‖C(x)-x‖ 2 ]≤r‖x‖ 2 E C [·] denotes the expectation of the internal randomness of the random compression operator C, and ||·| denotes the 2-norm and r∈[0,1). It is an n-dimensional vector consisting entirely of zeros. for
[0052] 3. Ground control station aggregation process: This process involves the ground control station aggregating UAV information and then broadcasting it to some UAV nodes to achieve a consistent value. At this stage, the UAV information has already been compressed, and only the FedADMM algorithm needs to be applied. The process is as follows:
[0053] Step 301 When k = {0, k0, 2k0, 3k0, ...}, the main terminal receives the set After the data uploaded by the drone is updated, This represents the final data from the drone.
[0054] Step 302: The central terminal will randomly select n drones to form a subset. in Then put Broadcast to this subset of drones via wireless communication module; randomly selected as follows:
[0055] Select all drones, and randomly select a subset of drones when k>0, then satisfy the following sequence in each group containing s0 sets. Ensure that each drone is selected at least once.
[0056] Example 2
[0057] The present invention provides a multi-UAV power line inspection method based on compressed communication and FedADMM optimization. The main difference between this method and the method in Embodiment 1 is that different update technologies are used.
[0058] Specifically, the drone's local parameters also include To compress variables, For the new compressed data; update the local parameters according to the following formula: Internal update error tolerance value Make it satisfy in Given a range of [0.5, 1), the error tolerance value is guaranteed to approach 0; then the compressed variable is updated. in Given a positive real number, update the communication parameters. Make it satisfy in and f i (w) in gradient at σ i The value is a real number greater than 0, and then the cumulative error is updated. and data transmission and compressed data Finally, update the new local parameters. and γ is a positive real number; Ground control station aggregation process: global parameters are calculated according to the following formula: When k = {0, k0, 2k0, 3k0, ...}, the central terminal receives the set After the data uploaded by the drone is updated, This represents the final data from the drone.
[0059] The aforementioned update technique performs secondary compression on the transmitted data, allowing the compression error to approach zero more quickly, while its communication volume is only twice that of the method in Example 1, thus enabling communication to achieve consistent results more quickly.
[0060] To verify the effectiveness of Examples 1 and 2, the following methods are used as follows: Figure 3 Taking a cluster of 50 drones as an example, a simulation experiment was conducted. Figure 3 Each UAV transmits communication parameters to the ground control station along the arrows, and the determined energy consumption objective function is a linear regression problem. For example... Figure 4 As shown, where quantize indicates that the compression function C(·) is an unbiased quantization function, norm-sign indicates that the compression function C(·) is a norm-signed compression function, topk indicates that the compression function C(·) is a Top-k sparsity function, and uniform indicates that the compression function C(·) is a uniform quantization function, we can find that these compression functions, whether in Example 1 or Example 2, after a period of information transmission and update, reduce the error. It tends to 0, where express The average value Indicate f(w) in The gradient indicates that each user has reached the optimal point and achieved an average consistency result, while significantly reducing the communication throughput (BIT) compared to the original algorithm. In summary, this invention can reduce data transmission and energy consumption while ensuring that each user receives an average consistency result.
[0061] Example 3
[0062] Based on the same inventive concept as Embodiment 1, this embodiment of the invention provides a multi-UAV power line inspection system based on compressed communication and FedADMM optimization, comprising multiple UAV nodes. These UAV nodes form a transmission network with a central terminal, constituting a UAV swarm. Each UAV node includes a sensor and a client data transmission module. The sensor monitors the UAV's battery level, location, task completion status, and other status information. The client data transmission module includes a local client data update unit and a compressed transmission unit. The central terminal is equipped with an aggregation unit. Wherein:
[0063] The client-side local data update unit is used for:
[0064] The i-th drone determines its own energy consumption objective function;
[0065] The data transmitted by the i-th drone satisfy σ i Greater than 0 indicates The importance of; Let be the physical parameters required for communication by the i-th UAV. For the accumulated error, α i For constant weights, f i (·) represents the energy consumption objective function. These are the initial physical parameters. Represents the gradient;
[0066] Receive global parameters calculated by the central terminal The i-th drone, of which This represents the sum of all weights, where m represents the total number of drones. These are compressed variables and compressed data, respectively; first, the error tolerance value is updated internally. Make it satisfy in Given a range of [0.5, 1), ensure the error tolerance value approaches 0; then update the compression variables. in Given a positive real number, update the communication parameters. Make it satisfy in ‖·‖ represents the 2-norm; the cumulative error is updated last. and data transmission and compressed data Where C(·) is the compression function; and for drones that do not receive a signal, its parameters remain unchanged;
[0067] The compressed transmission unit is used for: when communicating with the central terminal, the i-th UAV will compress the transmitted data. Compress to Then compress the data and compressed variables Transmitted to the central terminal;
[0068] The aggregation unit is used to: update the data transmitted by the selected subset of UAVs after the main unit receives the transmitted data. This represents the final data from the drone.
[0069] Example 4
[0070] Based on the same inventive concept as Embodiment 2, the present invention provides a multi-UAV power line inspection system based on compressed communication and FedADMM optimization, which differs from Embodiment 3 in that:
[0071] In the aggregation unit, the global parameters are calculated by the total end according to the following formula: In the client-side data local update unit, the drone's local parameters include: To compress variables, For the new compressed data; update the local parameters according to the following formula: Internal update error tolerance value Make it satisfy Next, update the compression variables. And update communication parameters Make it satisfy in Then update the cumulative error. and data transmission and compressed data Finally, update the new local parameters. and γ is a positive real number.
[0072] Example 5
[0073] This invention provides a multi-UAV power line inspection system based on compressed communication and FedADMM optimization, comprising multiple UAV nodes, which form a transmission network with the central terminal, constituting a UAV swarm;
[0074] Each UAV node includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of local update and compressed transmission of UAV data in the multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization in Embodiment 1 or Embodiment 2.
[0075] The central terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the central terminal aggregation step in the multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization in Embodiment 1 or Embodiment 2.
[0076] Example 6
[0077] The present invention provides a computer system including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of local updating and compressed transmission of UAV data in the multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization in Embodiment 1 or Embodiment 2, or the step of central aggregation.
[0078] Example 7
[0079] This invention provides a computer program product, comprising a computer program that, when executed by a processor, implements the steps of local updating and compressed transmission of UAV data, or the step of central aggregation, in the multi-UAV power line inspection method based on compressed communication and FedADMM optimization according to Embodiment 1 or Embodiment 2.
[0080] It should be noted that the above-described embodiments only illustrate some implementation methods of the present invention, and their description should not be construed as limiting the scope of the present invention. It should be pointed out that those skilled in the art can make several improvements without departing from the concept of the present invention, and these improvements should all fall within the protection scope of the present invention.
Claims
1. A multi-UAV power line inspection method based on compressed communication and FedADMM optimization, characterized in that: This includes the process of local updating of drone data, compressed transmission, and aggregation at the central terminal; the local updating of drone data includes: The i-th drone determines its own energy consumption objective function; The data transmitted by the i-th drone satisfy σ i Greater than 0 indicates The importance of; Let be the physical parameters required for communication by the i-th UAV. For the accumulated error, α i For constant weights, f i (·) represents the energy consumption objective function. These are the initial physical parameters. Represents the gradient; The central terminal randomly selects a subset of drones to calculate global parameters. And broadcast to the drones in this subset, among which This represents the sum of all weights, where m represents the total number of drones. These are compressed variables and compressed data, respectively. Received The i-th drone first internally updates the error tolerance value. Make it satisfy Where θ i Given a range of [0.5, 1), ensure the error tolerance value approaches 0; then update the compression variables. in Given a positive real number, update the communication parameters. Make it satisfy in ‖·‖ represents the 2-norm; the cumulative error is updated last. and data transmission and compressed data Where C(·) is the compression function; and for drones that do not receive a signal, its parameters remain unchanged; The aggregation process at the central terminal includes: after receiving the transmission data from the selected subset of UAVs, the central terminal updates... This represents the final data from the drone.
2. The multi-UAV power line inspection method based on compressed communication and FedADMM optimization according to claim 1, characterized in that: The central server generates α based on the priority of each drone's mission. i The higher the priority, the better. i The larger, and α i >0.
3. The multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization according to claim 1, characterized in that: The main unit randomly selects drones to form a subset in the following manner. in All drones are selected, while a subset of drones is randomly selected when k>
0. The goal is to ensure that all drones are selected at least once in each sequence containing s0 sets. k and k0 represent the number of local iterations for the drone and the communication cycle between the drone and the central terminal, respectively. This indicates rounding up to the nearest integer.
4. The multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization according to claim 1, characterized in that: The compression function C(x) satisfies E c [‖C(x)-x‖ 2 ]≤r‖x‖ 2 ,r∈[0,1),E C [·] represents the expectation of the internal randomness of the random compression operator C.
5. The multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization according to claim 1, characterized in that: The global parameters are calculated using the following formula: The drone's local parameters include To compress variables, For the new compressed data; update the local parameters according to the following formula: Internal update error tolerance value Make it satisfy Next, update the compression variables. And update communication parameters Make it satisfy in Then update the cumulative error. and data transmission and compressed data Finally, update the new local parameters. and γ is a positive real number.
6. A multi-UAV power line inspection system based on compressed communication and FedADMM optimization, characterized in that: It includes multiple drone nodes, which form a transmission network with the central terminal, constituting a drone swarm; each drone node includes sensors for monitoring drone status and a client data transmission module. The client data transmission module includes a client data local update unit and a compressed transmission unit, and the central terminal is equipped with an aggregation unit. The client data local update unit is used for: The i-th drone determines its own energy consumption objective function; The data transmitted by the i-th drone satisfy σ i Greater than 0 indicates The importance of; Let be the physical parameters required for communication by the i-th UAV. For the accumulated error, α i For constant weights, f i (·) represents the energy consumption objective function. These are the initial physical parameters. Represents the gradient; Receive global parameters calculated by the central terminal The i-th drone, of which This represents the sum of all weights, where m represents the total number of drones. These are compressed variables and compressed data, respectively; first, the error tolerance value is updated internally. Make it satisfy Where θ i Given a range of [0.5, 1), ensure the error tolerance value approaches 0; then update the compression variables. in Given a positive real number, update the communication parameters. Make it satisfy in ‖·‖ represents the 2-norm; the cumulative error is updated last. and data transmission and compressed data Where C(·) is the compression function; and for drones that do not receive a signal, its parameters remain unchanged; The compressed transmission unit is used to: when communicating with the central terminal, the i-th UAV will transmit data... Compress to Then compress the data and compressed variables Transmitted to the central terminal; The aggregation unit is configured to: after receiving the transmission data from the selected subset of UAVs, update... This represents the final data from the drone.
7. The multi-UAV power line inspection system based on compressed communication and FedADMM optimization according to claim 6, characterized in that: In the aggregation unit, the global parameters are calculated by the main end according to the following formula: The local update unit for client data includes the following local parameters for the drone: To compress variables, For the new compressed data; update the local parameters according to the following formula: Internal update error tolerance value Make it satisfy Next, update the compression variables. And update communication parameters Make it satisfy in Then update the cumulative error. and data transmission and compressed data Finally, update the new local parameters. and γ is a positive real number.
8. A multi-UAV power line inspection system based on compressed communication and FedADMM optimization, characterized in that: It includes multiple drone nodes, and the multiple drone nodes form a transmission network with the central terminal, constituting a drone swarm; Each UAV node includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of local updating and compressed transmission of UAV data in the multi-UAV power line inspection method based on compressed communication and FedADMM optimization according to any one of claims 1-5. The central terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the central terminal aggregation step in the multi-UAV transmission line inspection method based on compressed communication and FedADMM optimization according to any one of claims 1-5.
9. A computer system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of local updating and compressed transmission of UAV data, or the step of central aggregation, in the multi-UAV power line inspection method based on compressed communication and FedADMM optimization according to any one of claims 1-5.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of local updating and compressed transmission of UAV data, or the step of central aggregation, in the multi-UAV power line inspection method based on compressed communication and FedADMM optimization according to any one of claims 1-5.