Low-altitude economic unmanned aerial vehicle communication optimization method based on machine learning

Abstract

The invention relates to the field of unmanned aerial vehicle communication, in particular to a low-altitude economic unmanned aerial vehicle communication optimization method based on machine learning, which comprises the following steps: establishing an actual effective communication range through a communication attenuation gradient, and obtaining a multi-layer communication projection area of an unmanned aerial vehicle under plane projection of a map model information unit; other communication units and user units are covered according to the effective projection area, path intention information of other unmanned aerial vehicles is received under the communication projection area, and the communication units and the unmanned aerial vehicles in the effective communication range are relay communication units. According to the invention, by setting a low-altitude linkage communication mode, unmanned aerial vehicle groups can be linked with one another to form unmanned aerial vehicle chain connection in the region, and then the whole network chain is expanded to a certain extent according to intention paths of a plurality of different unmanned aerial vehicles on the basis of original path planning, so that the coverage of the whole chain signal is enhanced, and meanwhile, the unmanned aerial vehicle chain connection is realized. And the optimal communication distance of the unmanned aerial vehicle can be further enhanced.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle communication, and in particular to a low-altitude economical unmanned aerial vehicle communication optimization method based on machine learning. Background Art

[0002] The low-altitude economy is a comprehensive economic form that relies on low-altitude airspace and is driven by various low-altitude flight activities of various manned and unmanned aircraft, radiating and driving the integrated development of related fields. Among them, the application of unmanned aircraft, especially drones, is becoming a trend that cannot be ignored.

[0003] The drone algorithms of the existing technology usually involve optimization designs under current communications. However, as dynamic mobile devices, drones are inherently unstable in communication coverage signals. Whether it is the user's location or the drone's location, as well as the automated connection through the cellular network, all pose challenges to the drone's communication strength. In the face of low-altitude economy, how to achieve stable signal transmission in the low-altitude range is also a trend that drones need to develop. Especially in areas with poor signals, too many and dense buildings will lead to more situations at corners. Relying solely on the instant drone path algorithm to avoid obstacles will obviously put a heavy load on the drone's overall system. Obstacle avoidance, as the last preventive mechanism, obviously requires a pre-linked communication optimization method to reduce collisions in the air. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a low-altitude economical UAV communication optimization method based on machine learning.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect of the present invention, a method for optimizing communication of a low-altitude economical UAV based on machine learning is provided, comprising:

[0007] The actual effective communication range is established through the communication attenuation gradient, and the multi-layer communication projection area of ​​the UAV under the plane projection of the map model information unit is obtained;

[0008] Cover other communication units and user units according to the effective projection area, receive the path intention information of other drones under the communication projection area, and serve as relay communication units with drones within the effective communication range;

[0009] The optimal signal path planning is established based on the machine learning algorithm, the signal sources in the environment are divided into fixed and variable quantities, and the signal source positioning information is continuously updated iteratively.

[0010] As a preferred technical solution of the present invention, the machine learning algorithm includes a drone path planning algorithm and a particle swarm algorithm, wherein the particle swarm algorithm is used to calculate the information of the received variable communication source, the variable communication source includes a drone relay unit, a mobile communication source and an unstable signal source, and the particle swarm algorithm includes the following steps:

[0011] A. Initialization: Generate an initial communication particle swarm, assign a random initial position and signal speed to each particle, and dynamically collect the degree of signal change;

[0012] B. Calculate the adaptive range of the signal source: by establishing the optimal communication range, let the communication of a single particle expand the default area, and re-mark the position that enters the optimal communication range when moving as the actual effective communication range;

[0013] C. Update the information based on the path intention: Build a link network of the variable signal source, and change the original UAV's path according to the actual state of the link network, where the edge of the link network tends to shrink inward, and the inner side of the link network tends to expand outward.

[0014] As a preferred technical solution of the present invention, the fixed quantity includes a user signal source, a base station signal source and a satellite communication signal, and the drone also includes a map model information unit and a system communication module.

[0015] As a preferred technical solution of the present invention, the map model information unit further includes an airborne radar, wherein the map model information unit includes buildings, geographical location heights and newly added objects in the air, and the newly added objects in the air are identified by the airborne radar;

[0016] The map model information unit is marked with communication range values ​​at different height intervals. The reduction of communication range values ​​is based on the obstruction settings of buildings and newly added objects in the air.

[0017] In a second aspect of the present invention, a low-altitude economic UAV communication optimization system based on machine learning is provided, comprising a user-end system, wherein the user-end system is linked to a system communication module of the UAV, and the user-end system comprises:

[0018] A communication module, used to communicate with a single drone and link multiple drones with the same mission;

[0019] The terminal module is used to connect to the server terminal and connect to the link network by connecting to the server terminal. The server terminal is used to connect or disconnect drones in the same area under the link network;

[0020] The communication optimization module is used to update the actual communication distance of the drone.

[0021] In a third aspect of the present invention, there is provided an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for topology optimization of a drone communication network as claimed in any one of claims 1 to 4 are implemented, or the functions of the system as claimed in claim 5 are implemented.

[0022] In a fourth aspect of the present invention, a non-transitory computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the drone communication network topology optimization method as described in any one of claims 1 to 4, or the functions of the system as described in claim 5.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1: The present invention sets up a low-altitude linkage communication method to allow drone groups to link with each other, forming a chain connection of drones in the area, and then expands the overall network chain to a certain extent based on the original path planning according to the intended paths of multiple different drones. While enhancing the overall chain signal coverage, it can also further enhance the optimal communication distance of the drones.

[0025] 2: Based on the above method, the present invention further designs an overall communication system. After a single server establishes a cloud network, a chain relay effect can be generated within the area based on a fixed signal source. Based on the onboard wireless module and communication module carried by the drone itself, the communication module can be used as a signal relay network within the range while on the move based on the signal enhancement module, and the system is controlled by the server as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0027] Figure 1 It is a schematic diagram of the UAV communication optimization process of the present invention;

[0028] Figure 2 It is a schematic diagram of the UAV communication optimization system of the present invention. DETAILED DESCRIPTION

[0029] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0030] Example 1

[0031] like Figure 1-2As shown, the present invention provides a low-altitude economical UAV communication optimization method based on machine learning, comprising:

[0032] The actual effective communication range is established through the communication attenuation gradient, and the multi-layer communication projection area of ​​the UAV under the plane projection of the map model information unit is obtained;

[0033] Cover other communication units and user units according to the effective projection area, receive the path intention information of other drones under the communication projection area, and serve as relay communication units with drones within the effective communication range;

[0034] The optimal signal path planning is established based on the machine learning algorithm, the signal sources in the environment are divided into fixed and variable quantities, and the signal source positioning information is continuously updated iteratively.

[0035] The machine learning algorithm includes the UAV path planning algorithm and the particle swarm algorithm. The particle swarm algorithm is used to calculate the information of the received variable communication source. The variable communication source includes the UAV relay unit, the mobile communication source and the unstable signal source. The particle swarm algorithm includes the following steps:

[0036] A. Initialization: Generate an initial communication particle swarm, assign a random initial position and signal speed to each particle, and dynamically collect the degree of signal change;

[0037] B. Calculate the adaptive range of the signal source: by establishing the optimal communication range, let the communication of a single particle expand the default area, and re-mark the position that enters the optimal communication range when moving as the actual effective communication range;

[0038] v i+1 =ω*v i +c1*rand()*(pbest i -x i )+c2*rand()*(gbest i -x i )

[0039] Where i = 1, 2, ..., n, n represents the total number of particle swarms; v i Represents the current particle speed, v i+1 Indicates the particle speed of the next iteration; the function rand() is used to generate a random number between 0 and 1, x i Represents the current position coordinates of the particle. The learning factors c1 and c2 are usually assigned fixed values; if v i Exceeding the allowable value V max , then v i Set to V max; pbesti represents the best position reached by an individual particle, while gbesti represents the global optimal position reached by the entire particle group;

[0040] As for the inertia factor ω, when the UAV is connected to the link network based on the communication module, the data sent by other UAVs is used as the basic calculation.

[0041] C. Update the information based on the path intention: construct a link network of the variable signal source, and change the original UAV path according to the actual state of the link network, where the edge of the link network tends to shrink inward, and the inner side of the link network tends to expand outward;

[0042] The Fratar model is used to calculate the trend-type flight distribution matrix of different flight routes in each region in the future, as shown in the following formula:

[0043]

[0044] In the formula,

[0045] Q ij ——The distribution of flights from area i to area j in a certain forecast period in the future;

[0046] Q 0ij ——The current departure volume of flights from zone i to zone j;

[0047] G j ——The growth multiple of the flight distribution intention in area j;

[0048] F i ——The growth multiple of the flight distribution occurrence in zone i;

[0049] n——the total number of regional flight intervals;

[0050] Q aj ——Flight distribution intention quantity in area j at characteristic time;

[0051] Q 0aj ——The current flight distribution formation volume in area j;

[0052] Q pi ——Flight distribution intention quantity in area i at characteristic time;

[0053] Q 0pi ——The current traffic volume in zone i;

[0054] L i ——The position coefficient of zone i with respect to all j zones;

[0055] L j —location coefficient of zone j for all zones i;

[0056] According to the above prediction, if there are few flights in the area, the route planning can be directly based on the signal source generated by the fixed quantity to achieve the best communication optimization effect. For example, if there is only a single drone in the area, the multi-hop wireless network deployment optimization based on the particle swarm algorithm can be directly adopted;

[0057] Therefore, under normal circumstances, when the flight units in the area are relatively crowded, the flight equipment will plan the path in order to obtain a better flight strategy to optimize the communication. At this time, in order to ensure the best optimization of the route, it will often avoid the more crowded sections. At this time, the flight intention of the target location can be calculated directly based on the above formula, and the link network is allocated to each different UAV through the target path in time sequence. When the same path is congested, priority is given to ensuring the passing of UAVs at the best communication height at the path location, so that continuous communication lines are formed around the UAVs when they pass by, and then other flight routes can be added around the communication lines to form an expansion of the link network with the range signal relay generated at the communication lines. For example, when there are UAVs passing at different heights on the path on one side of a building, priority is given to ensuring that the UAVs at this height pass continuously and form a signal relay point, and then a flight path is set on the other side of the building so that other UAVs can pass synchronously with the passing UAVs to form a new passing route.

[0058] Fixed quantities include user signal sources, base station signal sources and satellite communication signals. The drone also includes map model information units and system communication modules.

[0059] The map model information unit also includes an airborne radar, wherein the map model information unit includes buildings, geographical location heights, and newly added objects in the air, and the newly added objects in the air are identified by the airborne radar;

[0060] The map model information unit is marked with communication range values ​​at different height intervals. The reduction of communication range values ​​is based on the obstruction settings of buildings and newly added objects in the air.

[0061] In another embodiment, the present invention provides a low-altitude economical UAV communication optimization system based on machine learning, including a user-end system, the user-end system is linked to the system communication module of the UAV, and the user-end system includes:

[0062] A communication module, used to communicate with a single drone and link multiple drones with the same mission;

[0063] The terminal module is used to connect to the server terminal and connect to the link network by connecting to the server terminal. The server terminal is used to connect or disconnect drones in the same area under the link network;

[0064] The communication optimization module is used to update the actual communication distance of the drone.

[0065] In another embodiment, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the functions of the combined method embodiment or other corresponding system embodiments are implemented, which will not be described in detail here.

[0066] In another embodiment, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the functions of the method embodiment or other corresponding system embodiments described are implemented, which will not be repeated here.

[0067] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A low-altitude economic UAV communication optimization method based on machine learning, characterized in that: include: The actual effective communication range is established through the communication attenuation gradient, and the multi-layer communication projection area of ​​the UAV under the plane projection of the map model information unit is obtained; Cover other communication units and user units according to the effective projection area, receive the path intention information of other drones under the communication projection area, and serve as relay communication units with drones within the effective communication range; The optimal signal path planning is established based on the machine learning algorithm, the signal sources in the environment are divided into fixed and variable quantities, and the signal source positioning information is continuously updated iteratively.

2. According to the machine learning-based low-altitude economic UAV communication optimization method of claim 1, it is characterized in that: The machine learning algorithm includes a drone path planning algorithm and a particle swarm algorithm, wherein the particle swarm algorithm is used to calculate the information of the received variable communication source, the variable communication source includes a drone relay unit, a mobile communication source and an unstable signal source, and the particle swarm algorithm includes the following steps: A. Initialization: Generate an initial communication particle swarm, assign a random initial position and signal speed to each particle, and dynamically collect the degree of signal change; B. Calculate the adaptive range of the signal source: by establishing the optimal communication range, let the communication of a single particle expand the default area, and re-mark the position that enters the optimal communication range when moving as the actual effective communication range; C. Update the information based on the path intention: Build a link network of the variable signal source, and change the original UAV's path according to the actual state of the link network, where the edge of the link network tends to shrink inward, and the inner side of the link network tends to expand outward.

3. The low-altitude economic UAV communication optimization method based on machine learning according to claim 2 is characterized in that: The fixed quantities include user signal sources, base station signal sources and satellite communication signals, and the drone also includes a map model information unit and a system communication module.

4. The low-altitude economic UAV communication optimization method based on machine learning according to claim 3 is characterized in that: The map model information unit also includes an airborne radar, wherein the map model information unit includes buildings, geographical location heights, and newly added objects in the air, and the newly added objects in the air are identified by the airborne radar; The map model information unit is marked with communication range values ​​at different height intervals. The reduction of communication range values ​​is based on the obstruction settings of buildings and newly added objects in the air.

5. According to claim 1, a low-altitude economic UAV communication optimization system based on machine learning includes a user-side system, characterized in that: The user end system is linked to the system communication module of the drone, and the user end system includes: A communication module, used to communicate with a single drone and link multiple drones with the same mission; The terminal module is used to connect to the server terminal and connect to the link network by connecting to the server terminal. The server terminal is used to connect or disconnect drones in the same area under the link network; The communication optimization module is used to update the actual communication distance of the drone.

6. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for topology optimization of a UAV communication network as claimed in any one of claims 1 to 4 or the functions of the system as claimed in claim 5 are implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for optimizing the topology of a drone communication network as described in any one of claims 1 to 4 or the functions of the system as described in claim 5 are implemented.

Images