A method and system for cooperative dynamic obstacle avoidance for a swarm of drones

Abstract

The application relates to a cooperative dynamic obstacle avoidance method and system for a UAV group, and relates to the technical field of UAV obstacle avoidance, which comprises the following steps: acquiring UAV group information and delivery goods information, determining UAV application data and a UAV application number of each UAV according to the UAV group information, constructing a UAV matrix based on the UAV application number, binding the delivery goods information with the UAV matrix, and obtaining UAV matrix data. A survey UAV layer is determined according to the UAV matrix, real-time visual field images are collected, obstacle analysis is performed on the real-time visual field images and the UAV application data, if there are aerial obstacles interfering with the UAV matrix, static and dynamic analysis is performed on the aerial obstacles, the UAV matrix is optimized and adjusted for obstacle avoidance flight, and an optimized and adjusted UAV matrix is obtained. The UAV matrix is controlled based on the optimized and adjusted UAV matrix to evolve the aerial obstacle avoidance. The application improves the efficiency and stability of UAV group goods delivery.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle obstacle avoidance, in particular to a cooperative dynamic obstacle avoidance method and system for a group of unmanned aerial vehicles. BACKGROUND

[0002] In the field of logistics distribution, a group of unmanned aerial vehicles (a multi-unmanned aerial vehicle cooperative system) is gradually becoming a key force for solving the "last mile" distribution problem due to its high efficiency and flexibility. In particular, in urban low-altitude distribution, remote area material transportation, medical emergency and other scenarios, the group of unmanned aerial vehicles can cross traffic congestion, terrain restrictions and other obstacles to achieve rapid and accurate delivery of goods.

[0003] However, during the logistics distribution process, the group of unmanned aerial vehicles may encounter dynamic obstacles in the air. Currently, the group of unmanned aerial vehicles faces dynamic obstacles based on real-time perception and cooperative decision-making dynamic path re-planning strategies, thereby achieving dynamic obstacle avoidance of the group of unmanned aerial vehicles and ensuring that the group of unmanned aerial vehicles can safely deliver the carried goods to the designated location. However, in the process of obstacle avoidance, the corresponding unmanned aerial vehicle needs to adjust its pose, and the goods carried on the unmanned aerial vehicle will also be sent to shake and oscillate, which not only directly threatens the physical integrity of the goods, especially for fragile goods, slight vibration can cause damage or performance degradation, increasing logistics costs and affecting customer satisfaction. At the same time, the frequent shaking of the goods also interferes with the flight stability of the unmanned aerial vehicle, further increasing the difficulty of obstacle avoidance and energy consumption, and even in extreme cases, triggering a chain reaction, affecting the cooperative operation efficiency and safety of the entire group of unmanned aerial vehicles. SUMMARY

[0004] To solve at least one of the above technical problems, the present application provides a cooperative dynamic obstacle avoidance method and system for a group of unmanned aerial vehicles.

[0005] In a first aspect, the present application provides a cooperative dynamic obstacle avoidance method for a group of unmanned aerial vehicles, which adopts the following technical solution:

[0006] A cooperative dynamic obstacle avoidance method for a group of unmanned aerial vehicles, comprising:

[0007] Obtaining group of unmanned aerial vehicle information and delivery goods information carried by each unmanned aerial vehicle in the group of unmanned aerial vehicles;

[0008] Determining an application data for each unmanned aerial vehicle and an unmanned aerial vehicle application number according to the group of unmanned aerial vehicle information;

[0009] Constructing a unmanned aerial vehicle matrix based on the unmanned aerial vehicle application number, and binding the delivery goods information with the unmanned aerial vehicle matrix to obtain unmanned aerial vehicle matrix data;

[0010] Determine a detection UAV layer according to the UAV matrix, and collect real-time visual field images of the UAV matrix in the goods distribution process based on the detection UAV layer;

[0011] Perform obstacle analysis on the real-time visual field images and the UAV application data, determine whether there is an aerial obstacle in the real-time visual field images that interferes with the UAV matrix, if there is, perform static and dynamic analysis on the aerial obstacle, and based on the analysis results and the UAV matrix data, perform obstacle avoidance flight optimization adjustment on the UAV matrix to obtain an optimized and adjusted UAV matrix;

[0012] Control the UAV matrix to evolve based on the optimized and adjusted UAV matrix to avoid aerial obstacles.

[0013] By adopting the above technical scheme, the unmanned aerial vehicle group information and the delivery goods information carried by each unmanned aerial vehicle are obtained. Based on the above information, the overall situation of the unmanned aerial vehicle group and the specific tasks of each unmanned aerial vehicle can be accurately understood, ensuring that the entire delivery process is orderly carried out, avoiding confusion caused by missing or incorrect information, and improving the initial accuracy and reliability of the delivery task. According to the obtained unmanned aerial vehicle group information, the application data of each unmanned aerial vehicle and the unmanned aerial vehicle application number are determined. By clearly defining the application data and number, the precise positioning and parameter setting of the individual unmanned aerial vehicle can be realized, laying the foundation for building an orderly unmanned aerial vehicle matrix, making the subsequent matrix operation more accurate and efficient. Based on the unmanned aerial vehicle application number, the unmanned aerial vehicle matrix is constructed, and the delivery goods information is bound with the unmanned aerial vehicle matrix to obtain the unmanned aerial vehicle matrix data. The construction of the unmanned aerial vehicle matrix can integrate the dispersed unmanned aerial vehicles into an organic whole, and through reasonable matrix arrangement, the collaborative work between the unmanned aerial vehicles can be realized, improving the delivery efficiency. Binding the delivery goods information with the matrix makes the position and task of each unmanned aerial vehicle in the matrix more clear, avoiding the confusion of task allocation. The obtained unmanned aerial vehicle matrix data provides comprehensive information for subsequent real-time monitoring and adjustment, ensuring that the unmanned aerial vehicles can be accurately scheduled according to the actual situation during the delivery process, improving the flexibility and response speed of the entire delivery system. According to the unmanned aerial vehicle matrix, the reconnaissance unmanned aerial vehicle layer is determined, and based on the reconnaissance unmanned aerial vehicle layer, the real-time field of view images of the unmanned aerial vehicle matrix in the goods delivery process are collected. By collecting real-time field of view images, the air situation around the unmanned aerial vehicle matrix is obtained in real time, including possible obstacles, weather changes and other information. These real-time images provide intuitive basis for subsequent obstacle analysis and obstacle avoidance decision-making, enabling the unmanned aerial vehicles to accurately perceive the surrounding situation in complex environments, discovering potential dangers in advance, providing strong protection for safe delivery, and effectively reducing the risk in the delivery process. Obstacle analysis is performed on the real-time field of view images and the unmanned aerial vehicle application data to determine whether there are air obstacles that interfere with the unmanned aerial vehicle matrix in the real-time field of view images. If there are, static and dynamic analysis is performed on the air obstacles, and based on the analysis results and the unmanned aerial vehicle matrix data, the unmanned aerial vehicle matrix is optimized and adjusted for obstacle avoidance flight to obtain the optimized and adjusted unmanned aerial vehicle matrix. Obstacle analysis can quickly and accurately identify threats, and static and dynamic analysis can further understand the motion state of the obstacles. Combined with the unmanned aerial vehicle matrix data, the obstacle avoidance optimization adjustment can plan the best obstacle avoidance path for each unmanned aerial vehicle according to the actual situation, enabling the unmanned aerial vehicle matrix to efficiently adjust the flight state while ensuring the safety of the delivery goods when encountering obstacles, avoiding collision accidents, ensuring the smooth progress of the delivery task, and improving the safety and reliability of the delivery. Based on the optimized and adjusted unmanned aerial vehicle matrix, the unmanned aerial vehicle matrix is controlled to avoid obstacles in the air. Ensure that the unmanned aerial vehicles accurately perform obstacle avoidance actions according to the established scheme. Thus, not only can the obstacles be successfully avoided, but also the impact on the delivery goods can be minimized, maintaining the efficiency and stability of the delivery.

[0014] The application can be further configured in a preferred example that: the obstacle analysis on the real-time field of view image and the UAV application data determines whether there is an aerial obstacle in the real-time field of view image that interferes with the UAV matrix, comprising:

[0015] According to the UAV application data, a UAV three-dimensional model of the UAV matrix in the flight process is built;

[0016] According to the delivery goods information, a three-dimensional simulation processing is performed on the delivery goods, and a goods three-dimensional model obtained after the processing is model fitted with the UAV three-dimensional model according to a loading correspondence and a preset ratio to obtain a delivery three-dimensional model;

[0017] The real-time field of view image is subjected to image fusion processing to obtain a processed real-time field of view image;

[0018] The processed real-time field of view image is input into a preset obstacle model for detection to determine whether there is an obstacle in the real-time field of view image, and if there is, the relative position relationship between the obstacle and the investigation UAV layer in the real-time field of view image is determined, and whether the obstacle reaches the preset flight range of the investigation UAV layer is determined according to the relative position relationship;

[0019] If the obstacle does not reach the preset flight range of the investigation UAV layer, the obstacle evolution of the relative position relationship at different time nodes is determined, the obstacle evolution is predicted and deduced, the future obstacle evolution between the obstacle and the investigation UAV layer in a future time period is determined, an evolution three-dimensional model of the obstacle is built according to the future obstacle evolution, and the evolution three-dimensional model is model fitted with the delivery three-dimensional model according to the relative position relationship to obtain a monitoring three-dimensional model;

[0020] According to the monitoring three-dimensional model, it is determined whether there is an aerial obstacle in the real-time field of view image that interferes with the UAV matrix.

[0021] The application can be further configured in a preferred example that: the image fusion processing on the real-time field of view image to obtain a processed real-time field of view image comprises:

[0022] The real-time field of view image is subjected to edge filtering processing to obtain a primary layer and a secondary layer, the primary layer is a scene basic outline layer in the real-time field of view image, and the secondary layer is a scene detail outline layer in the real-time field of view image;

[0023] The primary layer is divided according to a preset division rule to obtain a plurality of layer sub-regions;

[0024] collecting a sub-region histogram and a sub-region flatness corresponding to each sub-region of the image layer and a global histogram and a global flatness corresponding to the primary image layer, and determining a primary tone curve and a light and dark degree enhancement curve corresponding to the primary image layer according to the global histogram and the global flatness;

[0025] determining a secondary tone curve and a curve fusion weight corresponding to each sub-region according to the sub-region histogram and the sub-region flatness corresponding to each sub-region of the image layer, fusing the primary tone curve and the secondary tone curve to obtain a tone fusion curve;

[0026] fusing the tone fusion curve and the light and dark degree enhancement curve based on the curve fusion weight, and adjusting each sub-region in the primary image layer according to the fused visual field adjustment curve to obtain an adjusted primary image layer;

[0027] fusing the adjusted primary image layer and the secondary image to obtain a processed real-time visual field image.

[0028] In a preferred example, the application can be further configured to: the static and dynamic analysis of the aerial obstacles, and based on the analysis result and the UAV matrix data, the UAV matrix is optimized and adjusted to avoid obstacles to obtain an optimized and adjusted UAV matrix, including:

[0029] determining the impact point area of the aerial obstacles and the UAV matrix in the future time period according to the monitoring three-dimensional model, and determining the evolving UAV involved in the dynamic adjustment in the UAV matrix based on the impact point area;

[0030] determining the first cargo information corresponding to each evolving UAV based on the UAV matrix data, and inputting the first cargo information into a preset cargo distribution standard model for identification to obtain a first cargo sensitive value corresponding to each cargo in the first cargo information;

[0031] determining whether the first cargo sensitive value is greater than a preset sensitive value, if yes, taking the cargo corresponding to the first cargo sensitive value as a marker point, measuring a boundary distance value of each region boundary of the impact point area, and determining a minimum boundary distance value and a target region boundary corresponding to the minimum boundary distance value according to the boundary distance value;

[0032] determining the second cargo information corresponding to the non-evolving UAV located at the target region boundary based on the UAV matrix data, and inputting the second cargo information into a preset cargo distribution standard model for identification to obtain a second cargo sensitive value corresponding to each cargo in the second cargo information;

[0033] determining whether the second cargo sensitivity value is greater than a preset sensitivity value, if not, calculating the relative distance between each non-evolution unmanned aerial vehicle in the target area boundary and the marker point, determining the target non-evolution unmanned aerial vehicle with the closest relative distance to the marker point according to the relative distance, and generating a position replacement instruction based on the target non-evolution unmanned aerial vehicle and the unmanned aerial vehicle corresponding to the marker point, to control the target non-evolution unmanned aerial vehicle and the unmanned aerial vehicle corresponding to the marker point to replace positions, to obtain a position-replaced unmanned aerial vehicle matrix;

[0034] analyzing the evolution of the evolution unmanned aerial vehicle in the position-replaced unmanned aerial vehicle matrix and the unmanned aerial vehicle in the position-replaced unmanned aerial vehicle matrix corresponding to the non-target area boundary, to obtain a matrix evolution route;

[0035] optimizing and adjusting the position-replaced unmanned aerial vehicle matrix according to the matrix evolution route to obtain an optimized and adjusted unmanned aerial vehicle matrix.

[0036] In a preferred example, the application can be further configured to:

[0037] If the second cargo sensitivity value is greater than the preset sensitivity value, the third cargo information corresponding to the non-evolution unmanned aerial vehicle of each area boundary is determined one by one according to the boundary distance value, and the third cargo information is input into a preset cargo distribution standard model for identification to obtain a third cargo sensitivity value corresponding to each cargo in the third cargo information;

[0038] determining whether the third cargo sensitivity value is greater than a preset sensitivity value, if still greater than the preset sensitivity value, determining the center point position of the unmanned aerial vehicle matrix and the impact center point position of the impact point area, and connecting the center point position and the impact center point position to obtain a matrix movement vector line;

[0039] generating a matrix obstacle avoidance route according to the movement vector line to control the unmanned aerial vehicle matrix to move along the flight route.

[0040] In a preferred example, the application can be further configured to:

[0041] After detecting that the unmanned aerial vehicle matrix flies through the aerial obstacle, the optimized and adjusted unmanned aerial vehicle matrix is backtracked according to the matrix evolution route or the unmanned aerial vehicle matrix is backtracked according to the movement vector line to obtain a backtracked unmanned aerial vehicle matrix.

[0042] The application can be further configured in a preferred example as follows: the unmanned aerial vehicle matrix controlled based on the optimized adjustment is further configured to evolve in an aerial obstacle avoidance manner, and the application further comprises:

[0043] Network communication information for controlling the unmanned aerial vehicle matrix is collected, and network state detection is performed on the network communication information to determine whether the network state in the network communication information meets a preset network state. If not, regional network information in a preset range is obtained with the unmanned aerial vehicle matrix as the center, network channel information is determined according to the regional network information, and a network channel group other than a current application channel in the network communication information is determined based on the network channel information;

[0044] The network channel group is tested for signal transmission one by one to obtain a signal transmission rate corresponding to each network channel;

[0045] The signal transmission rate is analyzed by screening and combination, and the network channels are reorganized into a channel group according to a transmission order corresponding to each signal transmission rate in the network channel group information after screening and combination, to obtain a splicing network channel for sending control instructions to the unmanned aerial vehicle matrix.

[0046] In a second aspect, the application provides a cooperative dynamic obstacle avoidance system for a group of unmanned aerial vehicles, which adopts the following technical solution:

[0047] A cooperative dynamic obstacle avoidance system for a group of unmanned aerial vehicles comprises:

[0048] An information acquisition module is configured to acquire group information of the unmanned aerial vehicles and delivery goods information carried by each unmanned aerial vehicle in the group of unmanned aerial vehicles;

[0049] An application determination module is configured to determine, according to the group information of the unmanned aerial vehicles, application data of each unmanned aerial vehicle and an unmanned aerial vehicle application number;

[0050] A matrix generation module is configured to construct an unmanned aerial vehicle matrix based on the unmanned aerial vehicle application number, and bind the delivery goods information with the unmanned aerial vehicle matrix to obtain unmanned aerial vehicle matrix data;

[0051] An image acquisition module is configured to determine, according to the unmanned aerial vehicle matrix, a reconnaissance unmanned aerial vehicle layer, and acquire real-time field of view images of the unmanned aerial vehicle matrix in a goods delivery process based on the reconnaissance unmanned aerial vehicle layer;

[0052] An obstacle analysis module is configured to perform obstacle analysis on the real-time field of view images and the unmanned aerial vehicle application data to determine whether there is an aerial obstacle interfering with the unmanned aerial vehicle matrix in the real-time field of view images. If there is, the aerial obstacle is analyzed statically and dynamically, and the unmanned aerial vehicle matrix is optimized and adjusted for obstacle avoidance flight based on the analysis result and the unmanned aerial vehicle matrix data to obtain an optimized and adjusted unmanned aerial vehicle matrix.

[0053] An obstacle avoidance optimization module is configured to control the UAV matrix to evolve in air obstacle avoidance based on the optimized adjusted UAV matrix.

[0054] In a possible implementation, when the obstacle analysis module analyzes the real-time field of view image and the UAV application data to determine whether there is an air obstacle in the real-time field of view image that interferes with the UAV matrix, the obstacle analysis module is specifically configured to:

[0055] A UAV three-dimensional model of the UAV matrix in flight is built according to the UAV application data;

[0056] A three-dimensional simulation of the delivery goods is performed according to the delivery goods information, and a three-dimensional model of the goods obtained after the processing is model fitted with the UAV three-dimensional model according to a loading correspondence and a preset ratio to obtain a delivery three-dimensional model;

[0057] The real-time field of view image is subjected to image fusion processing to obtain a processed real-time field of view image;

[0058] The processed real-time field of view image is input into a preset obstacle model for detection to determine whether there is an obstacle in the real-time field of view image, and if there is, a relative position relationship between the obstacle in the real-time field of view image and the investigation UAV layer is determined, and whether the obstacle reaches a preset flight range of the investigation UAV layer is determined according to the relative position relationship;

[0059] If the obstacle does not reach the preset flight range of the investigation UAV layer, an evolution of the obstacle at different time nodes is determined, the evolution of the obstacle is predicted and deduced, a future evolution of the obstacle between the obstacle and the investigation UAV layer in a future time period is determined, an evolution three-dimensional model of the obstacle is built according to the future evolution of the obstacle, and the evolution three-dimensional model is model fitted with the delivery three-dimensional model according to the relative position relationship to obtain a monitoring three-dimensional model;

[0060] Whether there is an air obstacle in the real-time field of view image that interferes with the UAV matrix is determined according to the monitoring three-dimensional model.

[0061] In another possible implementation, when the obstacle analysis module performs image fusion processing on the real-time field of view image to obtain a processed real-time field of view image, the obstacle analysis module is specifically configured to:

[0062] perform edge filtering processing on the real-time field-of-view image to obtain a first-level layer and a second-level layer, the first-level layer being a scene basic outline layer in the real-time field-of-view image, and the second-level layer being a scene detail outline layer in the real-time field-of-view image;

[0063] divide the first-level layer according to a preset division rule to obtain a plurality of layer sub-regions;

[0064] collect a sub-region histogram and a sub-region flatness corresponding to each layer sub-region, and a global histogram and a global flatness corresponding to the first-level layer, and determine a first-level tone curve and a light-dark degree enhancement curve corresponding to the first-level layer according to the global histogram and the global flatness;

[0065] determine a second-level tone curve and a curve fusion weight corresponding to each sub-region according to the sub-region histogram and the sub-region flatness corresponding to each layer sub-region, perform fusion processing on the first-level tone curve and the second-level tone curve to obtain a tone fusion curve;

[0066] fuse the tone fusion curve and the light-dark degree enhancement curve based on the curve fusion weight, and adjust each sub-region in the first-level layer according to the fused field-of-view adjustment curve to obtain an adjusted first-level layer;

[0067] perform fusion processing on the adjusted first-level layer and the second-level layer to obtain a processed real-time field-of-view image.

[0068] In another possible implementation, when the obstacle analysis module performs static and dynamic analysis on the aerial obstacle, and performs obstacle avoidance flight optimization adjustment on the UAV matrix based on the analysis result and the UAV matrix data to obtain an optimized and adjusted UAV matrix, the obstacle analysis module is specifically configured to:

[0069] determine an impact point area of the aerial obstacle and the UAV matrix in a future time period according to the monitoring three-dimensional model, and determine an evolved UAV involved in dynamic adjustment in the UAV matrix based on the impact point area;

[0070] determine first cargo information corresponding to each evolved UAV based on the UAV matrix data, and input the first cargo information into a preset cargo distribution standard model for identification to obtain a first cargo sensitivity value corresponding to each cargo in the first cargo information;

[0071] determining whether the first goods sensitivity value is greater than a preset sensitivity value, if yes, taking goods corresponding to the first goods sensitivity value as a marker point, measuring a boundary distance value of each region boundary of the impact point area, and determining a minimum boundary distance value and a target region boundary corresponding to the minimum boundary distance value according to the boundary distance value;

[0072] determining second goods information corresponding to a non-evolution drone located at the target region boundary based on the drone matrix data, and inputting the second goods information into a preset goods distribution standard model for identification to obtain a second goods sensitivity value corresponding to each goods in the second goods information;

[0073] determining whether the second goods sensitivity value is greater than a preset sensitivity value, if no, calculating a relative distance between each non-evolution drone in the target region boundary and the marker point, determining a target non-evolution drone with a closest relative distance to the marker point according to the relative distance, and generating a position replacement instruction based on a drone corresponding to the target non-evolution drone and the marker point to control the drone corresponding to the target non-evolution drone and the marker point to replace positions to obtain a position-replaced drone matrix;

[0074] performing intercalation evolution analysis on an evolution drone in the position-replaced drone matrix and a drone not corresponding to the target region boundary in the position-replaced drone matrix to obtain a matrix evolution route;

[0075] performing obstacle avoidance flight optimization adjustment on the position-replaced drone matrix according to the matrix evolution route to obtain an optimized and adjusted drone matrix.

[0076] In another possible implementation, when determining whether the second goods sensitivity value is greater than a preset sensitivity value, the obstacle analysis module is specifically used for:

[0077] if the second goods sensitivity value is greater than the preset sensitivity value, determining third goods information corresponding to a non-evolution drone of each region boundary according to the boundary distance value one by one, inputting the third goods information into a preset goods distribution standard model for identification to obtain a third goods sensitivity value corresponding to each goods in the third goods information;

[0078] determining whether the third goods sensitivity value is greater than a preset sensitivity value, if yes, determining a center point position of the drone matrix and an impact center point position of the impact point area, and connecting the center point position and the impact center point position in a position line to obtain a matrix movement vector line;

[0079] The matrix obstacle avoidance route is generated according to the moving vector line, and the UAV matrix is controlled to move along the flight route.

[0080] In another possible implementation, the system further includes a matrix backtracking module, wherein,

[0081] The matrix backtracking module is configured to, when it is detected that the UAV matrix flies through the aerial obstacle, backtrack the UAV matrix according to the matrix evolution route or the moving vector line to obtain a backtracked UAV matrix.

[0082] In another possible implementation, the system further includes a channel combination module, a channel testing module, and a channel recombination module, wherein,

[0083] The channel combination module is configured to collect network communication information for controlling the UAV matrix, detect a network state of the network communication information, determine whether the network state of the network communication information meets a preset network state, if not, acquire regional network information in a preset range with the UAV matrix as a center, determine network channel information according to the regional network information, and determine a network channel group other than a current application channel in the network communication information based on the network channel information.

[0084] The channel testing module is configured to test the network channel group one by one to obtain a signal transmission rate corresponding to each network channel.

[0085] The channel recombination module is configured to analyze the signal transmission rates by screening and combination, recombine the network channels according to a transmission order corresponding to each signal transmission rate in the screened and combined network channel group information, and obtain a splicing network channel for sending a control instruction to the UAV matrix.

[0086] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0087] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned method for cooperative dynamic obstacle avoidance of a UAV group when executing the computer program.

[0088] In a fourth aspect, the present application provides a computer storage medium, which adopts the following technical solution:

[0089] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the above-mentioned method for cooperative dynamic obstacle avoidance of a UAV group.

[0090] In summary, the present application has the following beneficial technical effects:

[0091] The drone group information and the delivery goods information carried by each drone are obtained. Based on the above information, the overall situation of the drone group and the specific tasks of each drone can be accurately understood, ensuring that the entire delivery process is orderly carried out, avoiding confusion caused by missing or incorrect information, and improving the initial accuracy and reliability of the delivery task. According to the obtained drone group information, the application data and the drone application number of each drone are determined. By specifying the application data and the number, the precise positioning and parameter setting of the individual drone can be realized, laying the foundation for building an orderly drone matrix, making the subsequent matrix operation more accurate and efficient. Based on the drone application number, a drone matrix is constructed, and the delivery goods information is bound with the drone matrix to obtain drone matrix data. The construction of the drone matrix can integrate the scattered drones into an organic whole, and through reasonable matrix arrangement, the cooperative work between drones can be realized, improving the delivery efficiency. Binding the delivery goods information with the matrix makes the position and task of each drone in the matrix more clear, avoiding the confusion of task allocation. The obtained drone matrix data provides comprehensive information for subsequent real-time monitoring and adjustment, ensuring that the drones can be accurately scheduled according to the actual situation during the delivery process, improving the flexibility and response speed of the entire delivery system. According to the drone matrix, a reconnaissance drone layer is determined, and based on the reconnaissance drone layer, real-time field of view images of the drone matrix during the goods delivery process are collected. By collecting real-time field of view images, the air situation around the drone matrix is obtained in real time, including possible obstacles, weather changes and other information. These real-time images provide intuitive basis for subsequent obstacle analysis and obstacle avoidance decision-making, enabling the drone to accurately perceive the surrounding situation in complex environments, discovering potential dangers in advance, providing strong protection for safe delivery, and effectively reducing the risk in the delivery process. Obstacle analysis is performed on the real-time field of view images and the drone application data to determine whether there are air obstacles that interfere with the drone matrix in the real-time field of view images. If there are, static and dynamic analysis of the air obstacles is performed, and based on the analysis results and the drone matrix data, the drone matrix is optimized and adjusted for obstacle avoidance flight to obtain the optimized and adjusted drone matrix. Obstacle analysis can quickly and accurately identify threats, and static and dynamic analysis can further understand the motion state of the obstacles. Combined with the obstacle avoidance optimization adjustment based on the drone matrix data, the best obstacle avoidance path can be planned for each drone according to the actual situation, so that the drone matrix can efficiently adjust the flight state while ensuring the safety of the delivery goods when encountering obstacles, avoiding collision accidents and ensuring the smooth progress of the delivery task, improving the safety and reliability of the delivery. Based on the optimized and adjusted drone matrix, the drone matrix is controlled to avoid obstacles in the air. Ensure that the drones accurately perform obstacle avoidance actions according to the established scheme. Thus, not only can the obstacles be successfully avoided, but also the impact on the delivery goods can be minimized, maintaining the efficiency and stability of the delivery. BRIEF DESCRIPTION OF DRAWINGS

[0092] Figure 1 is a flow chart of a method for cooperative dynamic obstacle avoidance of a UAV group in one embodiment of the present application.

[0093] Figure 2 is a structural schematic diagram of a system for cooperative dynamic obstacle avoidance of a UAV group in one embodiment of the present application.

[0094] Figure 3 is a principle block diagram of an electronic device in one embodiment of the present application. DETAILED DESCRIPTION

[0095] The following will be described in detail below with reference to the accompanying drawings. Figure 1 to the accompanying drawings Figure 3 The present application will be further described in detail.

[0096] The present embodiments are merely explanatory of the present application, and are not a limitation of the present application, and those skilled in the art can make modifications to the present embodiments without creative contribution, but as long as within the scope of the claims of the present application, are protected by the patent law.

[0097] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme of the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0098] In addition, the term "and / or" in this paper is only to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. In addition, the character " / " in this paper, unless otherwise specified, generally represents an "or" relationship between the associated objects before and after.

[0099] The embodiments of the present application will be further described in detail below with reference to the drawings of the specification.

[0100] The embodiment of the present application provides a cooperative dynamic obstacle avoidance method for a UAV group, which is executed by an electronic device, and the electronic device can be a server or a terminal device. The server can be an independent physical server, a server cluster composed of multiple physical servers or a distributed system, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer and the like, but is not limited to this. The terminal device and the server can be directly or indirectly connected through wired or wireless communication, and the embodiment of the present application does not limit this. As shown in the following formula (1), the method comprises the following steps. Figure 1

[0101] Step S10: Obtain UAV group information and delivery goods information carried by each UAV in the UAV group.

[0102] In the present application, the UAV group refers to a set composed of multiple UAVs, and the UAVs can work cooperatively to complete a specific task such as logistics delivery. The UAV group information is used to represent the relevant data of the whole UAV group, including but not limited to the number, position, flight state, task type and UAV attribute parameters of the UAVs. The delivery goods information carried by each UAV refers to the detailed information of the goods loaded on each UAV for delivery, such as the type, weight, fragility, volume and destination of the goods.

[0103] For the embodiment of the present application, multiple types of sensors are installed on the UAV, such as a GPS module for obtaining position information and an acceleration sensor for obtaining flight acceleration data. A unique RFID tag is provided for each goods. A ground control station is built, and corresponding data receiving and processing equipment is provided to establish a communication link with the UAV. During the execution of the task, the UAV transmits the information collected by the sensors and the information read from the goods RFID tag to the ground control station in real time through the communication link, and the ground control station stores and analyzes the received information. It can be understood that other ways can also be used to obtain the UAV group information and the delivery goods information carried by each UAV in the UAV group, such as using satellite communication technology to make the UAV directly communicate with the satellite and transmit the information to the ground receiving station, or using blockchain technology to ensure the authenticity and non-tamperability of the information. Each UAV uploads the information to the blockchain network, and relevant parties can obtain the required information from the blockchain. Here, no limitation is made.

[0104] Step S11: Determine the application data of each UAV and the UAV application number according to the UAV group information.

[0105] ​In the embodiments of the present application, the UAV application data is used to represent the specific data related to the application of a single UAV in the UAV fleet, including the geometric shape of the UAV (such as the length, width, and height of the fuselage, the shape and size of the wings, etc.), the appearance features (such as color, material texture), the internal structure (such as engine position, cargo compartment layout), and the flight attitude related parameters (such as angle and motion data during takeoff, cruising, and landing), etc. Through these data, the three-dimensional form of the UAV in the actual scene can be accurately restored. The UAV application number is a unique identification number assigned to each UAV, which is used to accurately distinguish and identify different UAVs in the system, facilitating the management, scheduling of UAVs, and the recording and tracking of related data. For example, in a UAV fleet consisting of 10 UAVs, each UAV has its own independent application number, such as UAV-001, UAV-002, etc., and the three-dimensional model constructed based on the application data of each UAV will differ due to the different models and functions of the UAVs. For example, the three-dimensional models of cargo UAVs and reconnaissance UAVs will have obvious differences in cargo compartment structure and reconnaissance equipment layout.

[0106] Step S12: Constructing a UAV matrix based on the UAV application number, and binding the delivery goods information with the UAV matrix to obtain UAV matrix data.

[0107] In the embodiments of the present application, the UAV matrix is a logical structure for orderly arranging and organizing UAVs, constructed according to different dimensions (such as spatial position, function type, task priority, etc.). For example, it can be constructed into a three-dimensional grid matrix according to spatial position, with each cell corresponding to a UAV, facilitating the overall control and collaborative work of the UAV fleet. The UAV matrix data is a comprehensive data set formed after binding operation, which not only contains the structure information of the UAV matrix, but also contains the delivery goods information bound to each matrix position corresponding to the UAV. Through these data, the position of each UAV in the matrix and the goods it needs to deliver can be clearly understood. For example, in an e-commerce UAV delivery system, there are 10 UAVs, each with a unique application number. By constructing a 2-row 5-column UAV matrix, the 10 UAVs are arranged at different positions in the matrix, and the goods information of different customer orders is bound to the corresponding UAVs in the matrix, thus forming the UAV matrix data.

[0108] For the embodiments of the present application, a graphical interface tool is used to draw a layout of a UAV matrix on a computer. In the layout, each cell represents a UAV position, and the cells are filled by inputting the UAV application number, thereby visually constructing the UAV matrix. At the same time, the real UAVs are controlled synchronously according to the constructed UAV matrix, so that the real UAVs can be arranged according to the UAV matrix in the layout. Then a goods information management platform is established, which can import delivery goods information and provide query and filtering functions. By setting filtering conditions such as delivery time range, goods weight interval, etc., the goods information is filtered. An automatic binding script is developed, which reads the number information in the UAV matrix layout and the filtering results in the goods information management platform, and automatically binds the goods information with the UAVs in the UAV matrix according to a certain algorithm (such as the shortest delivery path algorithm). After the binding is completed, the binding result is displayed in the graphical interface, and the relevant data is exported as a UAV matrix data table, which contains detailed information such as UAV number, goods name, destination, etc.

[0109] Step S13: Determine the investigation UAV layer according to the UAV matrix, and collect real-time field of view images of the UAV matrix in the goods delivery process based on the investigation UAV layer.

[0110] For the embodiments of the present application, when the UAV matrix is constructed, a position identifier is set for each UAV, which contains its row and column information in the matrix. For example, a combination of numbers and letters, such as "A1", "B2", etc. is used to represent the position of the UAV. By writing a position recognition program, the program reads the position identifier of each UAV, filters out the UAVs located in the front row (or column, depending on the arrangement direction of the matrix) of the matrix, and determines them as the investigation UAV layer. Each UAV in the investigation UAV layer is equipped with a high-definition camera, and the camera is ensured to be normally connected with the communication module of the UAV. An image acquisition and transmission system is developed, which can receive image data collected by the camera in real time, and transmit the image to the control center through wireless communication technology (such as Wi-Fi, 4G / 5G, etc.). During the goods delivery process, the image acquisition and transmission system is started, and the UAVs in the investigation UAV layer continuously collect field of view images and transmit them back to the control center.

[0111] Step S14: Obstacle analysis is performed on the real-time field of view images and UAV application data to determine whether there are air obstacles interfering with the UAV matrix in the real-time field of view images. If there are, static and dynamic analysis is performed on the air obstacles, and based on the analysis results and the UAV matrix data, the UAV matrix is optimized and adjusted for obstacle avoidance flight, to obtain the optimized and adjusted UAV matrix.

[0112] Specifically, the UAV matrix is built according to the UAV application data, the three-dimensional model of the UAV in the flight process is built, the three-dimensional simulation processing is performed on the delivery goods according to the delivery goods information, the three-dimensional model of the goods after the processing is matched with the three-dimensional model of the UAV according to the loading correspondence and a preset proportion to obtain a delivery three-dimensional model. The real-time field of view image is subjected to image fusion processing to obtain a processed real-time field of view image. The processed real-time field of view image is input into a preset obstacle model for detection to determine whether there is an obstacle in the real-time field of view image. If there is, the relative position relationship between the obstacle and the investigation UAV layer in the real-time field of view image is determined, and whether the obstacle reaches the preset flight range of the investigation UAV layer is determined according to the relative position relationship. If the obstacle does not reach the preset flight range of the investigation UAV layer, the evolution of the obstacle at different time nodes is determined, the evolution of the obstacle is predicted and deduced, the future obstacle evolution between the obstacle and the investigation UAV layer in a future time period is determined, an evolution three-dimensional model of the obstacle is built according to the future obstacle evolution, and the evolution three-dimensional model is matched with the delivery three-dimensional model according to the relative position relationship to obtain a monitoring three-dimensional model. Whether there is an aerial obstacle interfering with the UAV matrix is determined according to the monitoring three-dimensional model.

[0113] In the embodiments of the present application, the loading correspondence specifies the specific placement position and manner of the goods on the UAV, for example, whether the goods are placed in the cargo compartment of the UAV or fixed externally through a mounting device. The preset proportion is a scale set according to the size relationship between the actual UAV and the goods to ensure the accuracy and reasonableness of the model matching. By matching the models according to the loading correspondence and the preset proportion, the three-dimensional model of the goods and the three-dimensional model of the UAV are integrated together to form a complete virtual model reflecting the actual loading condition. The preset obstacle model is a model trained in advance through machine learning or image recognition technology, which can identify various possible obstacles in the image, such as buildings, trees, other flying objects, etc. The preset flight range is a safety space area required to be maintained for safe flight of the investigation UAV layer. According to the relative position relationship between the obstacle and the investigation UAV layer, it is determined whether the obstacle enters the preset range, so as to evaluate the potential threat to the UAV flight.

[0114] For the embodiments of the present application, the process of image fusion processing on the real-time field of view image includes: performing edge filtering processing on the real-time field of view image to obtain a first-level layer and a second-level layer, the first-level layer being a scene basic contour layer in the real-time field of view image, and the second-level layer being a scene detail contour layer in the real-time field of view image. The first-level layer is divided according to a preset division rule to obtain a plurality of layer sub-regions. The sub-region histogram and the sub-region flatness corresponding to each layer sub-region and the global histogram and the global flatness corresponding to the first-level layer are collected, and the first-level tone curve and the light-darkness enhancement curve corresponding to the first-level layer are determined according to the global histogram and the global flatness. The second-level tone curve and the curve fusion weight corresponding to each sub-region are determined according to the sub-region histogram and the sub-region flatness corresponding to each layer sub-region, the first-level tone curve is fused with the second-level tone curve to obtain a tone fusion curve. The tone fusion curve is fused with the light-darkness enhancement curve based on the curve fusion weight, and each sub-region in the first-level layer is adjusted according to the fused field of view adjustment curve to obtain an adjusted first-level layer. The adjusted first-level layer is fused with the second-level layer to obtain a processed real-time field of view image.

[0115] For the embodiments of the present application, the process of obstacle avoidance flight optimization adjustment of the UAV matrix includes: determining the impact point area of the aerial obstacle with the UAV matrix in the future time period according to the monitored three-dimensional model, and determining the evolving UAV involved in the dynamic adjustment of the UAV matrix based on the impact point area. Determine the first cargo information corresponding to each evolving UAV based on the UAV matrix data, and input the first cargo information into the preset cargo distribution standard model for identification to obtain the first cargo sensitive value corresponding to each cargo in the first cargo information. Determine whether the first cargo sensitive value is greater than the preset sensitive value, if yes, take the cargo corresponding to the first cargo sensitive value as a marker point, measure the boundary distance value of each region boundary of the impact point area, and determine the minimum boundary distance value and the target region boundary corresponding to the minimum boundary distance value according to the boundary distance value. Determine the second cargo information corresponding to the non-evolving UAV located at the target region boundary based on the UAV matrix data, and input the second cargo information into the preset cargo distribution standard model for identification to obtain the second cargo sensitive value corresponding to each cargo in the second cargo information. Determine whether the second cargo sensitive value is greater than the preset sensitive value, if no, calculate the relative distance between each non-evolving UAV in the target region boundary and the marker point, determine the target non-evolving UAV with the closest relative distance to the marker point according to the relative distance, and generate a position replacement instruction based on the UAV corresponding to the target non-evolving UAV and the marker point, control the UAV corresponding to the target non-evolving UAV and the marker point to replace positions, and obtain the UAV matrix after position replacement. The evolving UAV in the UAV matrix after position replacement and the UAV corresponding to the non-target region boundary in the UAV matrix after position replacement are analyzed to obtain a matrix evolution route, the UAV matrix after position replacement is adjusted for obstacle avoidance flight optimization according to the matrix evolution route, and an optimized and adjusted UAV matrix is obtained.

[0116] In addition, when the second cargo sensitive value is greater than the preset sensitive value, the third cargo information corresponding to the non-evolving UAV of each region boundary is determined one by one according to the boundary distance value, the third cargo information is input into the preset cargo distribution standard model for identification, and the third cargo sensitive value corresponding to each cargo in the third cargo information is obtained. Determine whether the third cargo sensitive value is greater than the preset sensitive value, if still greater than the preset sensitive value, determine the center point position of the UAV matrix and the impact center point position of the impact point area, and connect the center point position and the impact center point position in position line to obtain a matrix movement vector line. Generate a matrix obstacle avoidance route according to the movement vector line, and control the UAV matrix to move in the flight line.

[0117] In the embodiments of the present application, the preset cargo delivery standard model is an evaluation model set according to various attributes of the cargo (such as fragility, value, etc.). After the first cargo information, the second cargo information and the third cargo information are input into the model, the model will assign a sensitivity value to each cargo according to the preset rules and algorithms, i.e., the first cargo sensitivity value, the second cargo sensitivity value and the third cargo sensitivity value, for measuring the sensitivity of the cargo to collision or flight state change. The preset sensitivity value is a threshold value set in advance for determining whether the cargo belongs to high-sensitivity cargo. By comparing the first cargo sensitivity value, the second cargo sensitivity value and the third cargo sensitivity value with the preset sensitivity value respectively, it can be determined which cargo is more sensitive to collision or flight adjustment.

[0118] Step S15, controlling the UAV matrix to evolve air obstacle avoidance based on the optimized and adjusted UAV matrix.

[0119] It is worth noting that when it is detected that the UAV matrix flies through the air obstacle, the optimized and adjusted UAV matrix is backtracked according to the matrix evolution route or the flight route of the UAV matrix is backtracked according to the movement vector route, to obtain the backtracked UAV matrix.

[0120] In the embodiments of the present application, the information of the UAV group and the delivery goods carried by each UAV is obtained. Based on the above information, the overall situation of the UAV group and the specific tasks of each UAV can be accurately understood, ensuring that the entire delivery process is orderly carried out, avoiding confusion caused by missing or incorrect information, and improving the initial accuracy and reliability of the delivery task. According to the obtained UAV group information, the application data of each UAV and the UAV application number are determined. By specifying the application data and number, the precise positioning and parameter setting of the individual UAV can be realized, laying the foundation for building an orderly UAV matrix, making the subsequent matrix operation more accurate and efficient. Based on the UAV application number, a UAV matrix is constructed, and the delivery goods information is bound with the UAV matrix to obtain UAV matrix data. The construction of the UAV matrix can integrate the scattered UAVs into an organic whole, and through reasonable matrix arrangement, the cooperative work between the UAVs can be realized, improving the delivery efficiency. Binding the delivery goods information with the matrix makes the position and task of each UAV in the matrix more explicit, avoiding confusion in task allocation. The obtained UAV matrix data provides comprehensive information for subsequent real-time monitoring and adjustment, ensuring that the UAVs can be accurately scheduled according to the actual situation during the delivery process, improving the flexibility and response speed of the entire delivery system. According to the UAV matrix, a reconnaissance UAV layer is determined, and based on the reconnaissance UAV layer, real-time field of view images of the UAV matrix during the goods delivery process are collected. By collecting real-time field of view images, the air situation around the UAV matrix is obtained in real time, including possible obstacles, weather changes and other information. These real-time images provide intuitive basis for subsequent obstacle analysis and obstacle avoidance decision-making, enabling the UAV to accurately perceive the surrounding situation in complex environments, discovering potential dangers in advance, providing strong protection for safe delivery, and effectively reducing the risk in the delivery process. Obstacle analysis is performed on the real-time field of view images and the UAV application data to determine whether there are air obstacles that interfere with the UAV matrix in the real-time field of view images. If there are, static and dynamic analysis is performed on the air obstacles, and based on the analysis results and the UAV matrix data, the UAV matrix is optimized and adjusted for obstacle avoidance flight to obtain the optimized and adjusted UAV matrix. Obstacle analysis can quickly and accurately identify threats, and static and dynamic analysis can further understand the motion state of the obstacles. Combined with the obstacle avoidance optimization adjustment based on the UAV matrix data, the best obstacle avoidance path can be planned for each UAV according to the actual situation, so that the UAV matrix can efficiently adjust the flight state while ensuring the safety of the delivery goods when encountering obstacles, avoiding collision accidents and ensuring the smooth progress of the delivery task, improving the safety and reliability of the delivery. Based on the optimized and adjusted UAV matrix, the UAV matrix is controlled to avoid obstacles in the air. Ensure that the UAV accurately performs the obstacle avoidance action according to the established scheme. Thus, not only can the obstacles be successfully avoided, but also the impact on the delivery goods can be minimized, maintaining the efficiency and stability of the delivery.

[0121] Furthermore, to ensure the efficiency of signal reception by the UAVs, this application collects network communication information used to control the UAV matrix in real time and performs network status detection on the network communication information to determine whether the network status in the network communication information meets the preset network status. If not, it acquires regional network information within a preset range centered on the UAV matrix, determines network channel information based on the regional network information, and determines network channel groups other than the current application channel in the network communication information based on the network channel information. It then performs signal transmission tests on each network channel group to obtain the signal transmission rate corresponding to each network channel. The signal transmission rates are then filtered and combined, and the network channels are reassembled according to the transmission order corresponding to each signal transmission rate in the filtered and combined network channel group information to obtain a spliced ​​network channel for sending control commands to the UAV matrix.

[0122] The above embodiments describe a cooperative dynamic obstacle avoidance method for UAV swarms from the perspective of method flow. The following embodiments describe a cooperative dynamic obstacle avoidance system for UAV swarms from the perspective of virtual modules or virtual units. For details, please refer to the following embodiments.

[0123] This application provides a cooperative dynamic obstacle avoidance system 20 for unmanned aerial vehicle (UAV) swarms, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of a cooperative dynamic obstacle avoidance system for a drone swarm, provided as an embodiment of this application. The system 20 specifically may include:

[0124] Information acquisition module 21 is used to acquire information about the drone swarm and the delivery information carried by each drone in the drone swarm;

[0125] Application determination module 22 is used to determine the application data and application number of each drone based on the drone swarm information;

[0126] The matrix generation module 23 is used to construct a drone matrix based on the drone application number and bind the delivery cargo information to the drone matrix to obtain drone matrix data;

[0127] The image acquisition module 24 is used to determine the reconnaissance drone layer based on the drone matrix, and to acquire real-time field-of-view images of the drone matrix during the cargo delivery process based on the reconnaissance drone layer;

[0128] The obstacle analysis module 25 is configured to perform obstacle analysis on the real-time field-of-view image and the UAV application data, determine whether there is an aerial obstacle interfering with the UAV matrix in the real-time field-of-view image, if there is, perform static and dynamic analysis on the aerial obstacle, and perform obstacle avoidance flight optimization adjustment on the UAV matrix based on the analysis result and the UAV matrix data to obtain an optimized and adjusted UAV matrix.

[0129] The obstacle avoidance optimization module 26 is configured to control the aerial obstacle avoidance evolution of the UAV matrix based on the optimized and adjusted UAV matrix.

[0130] In a possible implementation of the embodiment, when the obstacle analysis module 25 performs obstacle analysis on the real-time field-of-view image and the UAV application data, and determines whether there is an aerial obstacle interfering with the UAV matrix in the real-time field-of-view image, the obstacle analysis module 25 is specifically configured to:

[0131] build a UAV three-dimensional model of the UAV matrix in the flight process according to the UAV application data;

[0132] perform three-dimensional simulation processing on the delivery goods according to the delivery goods information, perform model fitting on the three-dimensional model of the goods and the three-dimensional model of the UAV according to a loading correspondence relationship and a preset proportion after processing to obtain a delivery three-dimensional model;

[0133] perform image fusion processing on the real-time field-of-view image to obtain a processed real-time field-of-view image;

[0134] input the processed real-time field-of-view image into a preset obstacle model for detection to determine whether there is an obstacle in the real-time field-of-view image, if there is, determine a relative position relationship between the obstacle and the investigation UAV layer in the real-time field-of-view image, and determine whether the obstacle reaches a preset flight range of the investigation UAV layer according to the relative position relationship;

[0135] if the obstacle does not reach the preset flight range of the investigation UAV layer, determine an obstacle evolution condition of the relative position relationship at different time nodes, predict and deduce the obstacle evolution condition, determine a future obstacle evolution condition between the obstacle and the investigation UAV layer in a future time period, build an evolution three-dimensional model of the obstacle according to the future obstacle evolution condition, and perform model fitting on the evolution three-dimensional model and the delivery three-dimensional model according to the relative position relationship to obtain a monitoring three-dimensional model;

[0136] determine whether there is an aerial obstacle interfering with the UAV matrix in the real-time field-of-view image according to the monitoring three-dimensional model.

[0137] In another possible implementation of the embodiment, when the obstacle analysis module 25 performs image fusion processing on the real-time field-of-view image to obtain a processed real-time field-of-view image, the obstacle analysis module 25 is specifically configured to:

[0138] The edge filtering processing is performed on the real-time field of view image to obtain a first layer and a second layer, the first layer is a scene basic contour layer in the real-time field of view image, and the second layer is a scene detail contour layer in the real-time field of view image;

[0139] The first layer is divided according to a preset division rule to obtain a plurality of layer sub-regions;

[0140] The sub-region histogram and the sub-region flatness corresponding to each layer sub-region and the global histogram and the global flatness corresponding to the first layer are collected, and the first tone curve and the light and dark enhancement curve corresponding to the first layer are determined according to the global histogram and the global flatness;

[0141] The second tone curve and the curve fusion weight corresponding to each sub-region are determined according to the sub-region histogram and the sub-region flatness corresponding to each sub-region, the first tone curve and the second tone curve are fused to obtain a tone fusion curve;

[0142] The tone fusion curve and the light and dark enhancement curve are fused based on the curve fusion weight, and each sub-region in the first layer is adjusted according to the fused field of view adjustment curve to obtain an adjusted first layer;

[0143] The adjusted first layer and the second layer are fused to obtain a processed real-time field of view image.

[0144] Another possible implementation of the embodiment of the application, the obstacle analysis module 25 in the static and dynamic analysis of the air obstacles, and based on the analysis results and the unmanned aerial vehicle matrix data on the unmanned aerial vehicle matrix obstacle avoidance flight optimization adjustment, get the optimized adjustment of the unmanned aerial vehicle matrix, specific for:

[0145] According to the monitoring three-dimensional model, the impact point area of the air obstacles and the unmanned aerial vehicle matrix in the future time period is determined, and the evolution unmanned aerial vehicle involved in the dynamic adjustment of the unmanned aerial vehicle matrix is determined based on the impact point area;

[0146] Based on the unmanned aerial vehicle matrix data, the first cargo information corresponding to each evolution unmanned aerial vehicle is determined, and the first cargo information is input into the preset cargo distribution standard model for identification to obtain the first cargo sensitive value corresponding to each cargo in the first cargo information;

[0147] It is judged whether the first cargo sensitive value is greater than the preset sensitive value, if yes, the cargo corresponding to the first cargo sensitive value is taken as a marker point, the boundary distance value of each region boundary of the impact point area is measured, and the minimum boundary distance value and the target region boundary corresponding to the minimum boundary distance value are determined according to the boundary distance value;

[0148] determine second cargo information corresponding to the non-evolution UAV located at the target region boundary based on the UAV matrix data, and input the second cargo information into a preset cargo distribution standard model for identification to obtain a second cargo sensitive value corresponding to each cargo in the second cargo information;

[0149] determine whether the second cargo sensitive value is greater than a preset sensitive value, and if not, calculate a relative distance between each non-evolution UAV in the target region boundary and the marker point, determine a target non-evolution UAV with a closest relative distance to the marker point according to the relative distance, and generate a position replacement instruction based on the target non-evolution UAV and the UAV corresponding to the marker point to control the UAV corresponding to the marker point to replace the position of the target non-evolution UAV, thereby obtaining a UAV matrix after position replacement;

[0150] perform interlaced evolution analysis on the evolution UAV in the UAV matrix after position replacement and the UAV outside the target region boundary in the UAV matrix after position replacement, thereby obtaining a matrix evolution route;

[0151] perform obstacle avoidance flight optimization adjustment on the UAV matrix after position replacement according to the matrix evolution route, thereby obtaining an optimized and adjusted UAV matrix.

[0152] In another possible implementation of the embodiment, the obstacle analysis module 25 is specifically configured to, when determining whether the second cargo sensitive value is greater than the preset sensitive value:

[0153] If the second cargo sensitive value is greater than the preset sensitive value, determine third cargo information corresponding to each region boundary non-evolution UAV according to the boundary distance value, input the third cargo information into the preset cargo distribution standard model for identification, and obtain a third cargo sensitive value corresponding to each cargo in the third cargo information;

[0154] determine whether the third cargo sensitive value is greater than the preset sensitive value, and if still greater than the preset sensitive value, determine a center point position of the UAV matrix and a collision center point position of a collision point area, and extend and connect the center point position and the collision center point position to obtain a matrix movement vector line;

[0155] generate a matrix obstacle avoidance route according to the movement vector line to control the UAV matrix to move along a flight route.

[0156] In another possible implementation of the embodiment, the system 20 further includes a matrix backtracking module, wherein,

[0157] The matrix backtracking module is configured to, when detecting that the UAV matrix has flown through an aerial obstacle, perform interlaced backtracking on the optimized and adjusted UAV matrix according to the matrix evolution route or perform flight route backtracking on the UAV matrix according to the movement vector line, thereby obtaining a backtracked UAV matrix.

[0158] In another possible implementation of the embodiment of the application, the system 20 further includes a channel combination module, a channel test module, and a channel recombination module, wherein,

[0159] The channel combination module is configured to collect network communication information for controlling the UAV matrix, and perform network state detection on the network communication information to determine whether a network state in the network communication information meets a preset network state. If not, the channel combination module is configured to acquire regional network information in a preset range centered on the UAV matrix, determine network channel information according to the regional network information, and determine a network channel group other than a current application channel in the network communication information based on the network channel information.

[0160] The channel test module is configured to perform one-by-one signal transmission test on the network channel group to obtain a signal transmission rate corresponding to each network channel.

[0161] The channel recombination module is configured to perform screening and combination analysis on the signal transmission rate, and perform channel group recombination on the network channel according to a transmission order corresponding to each signal transmission rate in the screened and combined network channel group information to obtain a splicing network channel for sending a control instruction to the UAV matrix.

[0162] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described dynamic obstacle avoidance system 20 for a UAV group can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0163] In the embodiment of the application, an electronic device is provided, such as Figure 3 as shown in the figure, Figure 3 which is a structural schematic diagram of an electronic device provided in the embodiment of the application. Figure 3 The electronic device 300 shown in the figure includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 300 can also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of the application.

[0164] The processor 301 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in connection with the disclosure. The processor 301 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0165] The bus 302 can include a path for transmitting information between the above-mentioned components. The bus 302 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 302 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 3 Only one thick line is used in the middle, but it does not mean that there is only one bus or one type of bus.

[0166] The memory 303 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, an optical disk storage (including a compact disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but not limited to this.

[0167] The memory 303 is configured to store application program codes for implementing the solutions of the present application, and the processor 301 is configured to execute the application program codes stored in the memory 303.

[0168] The electronic device includes, but is not limited to, a rugged computer, a monitoring management system, and the like. In a non-restricted occasion of a mobile device, a mobile phone, a notebook computer, a PAD, and the like can also be used. Figure 3 The electronic device shown is only an example, and should not limit the functions and use range of the embodiments of the present application.

[0169] The embodiments of the present application provide a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program is executed on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

[0170] It should be understood that, although each step in the flowchart of the accompanying drawings is shown in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0171] The above is only some of the embodiments of the present application, and it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for cooperative dynamic obstacle avoidance for a swarm of unmanned aerial vehicles, the method comprising: The application comprises the following steps: acquiring drone swarm information and delivery goods information carried by each drone in the drone swarm; determining application data and application number of each drone according to the drone swarm information; constructing a drone matrix based on the drone application number and binding the delivery goods information with the drone matrix to obtain drone matrix data; determining a reconnaissance drone layer according to the drone matrix and collecting real-time visual images of the drone matrix in the goods delivery process based on the reconnaissance drone layer; analyzing the real-time visual images and the drone application data to determine whether there is an aerial obstacle in the real-time visual images that interferes with the drone matrix, and if there is, performing static and dynamic analysis on the aerial obstacle and optimizing and adjusting the flight of the drone matrix based on the analysis results and the drone matrix data to obtain an optimized and adjusted drone matrix; the static and dynamic analysis of the aerial obstacle and the optimization and adjustment of the flight of the drone matrix based on the analysis results and the drone matrix data to obtain the optimized and adjusted drone matrix comprises the following steps: determining the impact point area of the aerial obstacle and the drone matrix in the future time period according to the monitoring three-dimensional model and determining the evolution drone involved in the dynamic adjustment of the drone matrix based on the impact point area; determining the first goods information corresponding to each evolution drone based on the drone matrix data and inputting the first goods information into a preset goods delivery standard model for identification to obtain a first goods sensitivity value corresponding to each goods in the first goods information; judging whether the first goods sensitivity value is greater than a preset sensitivity value, and if so, taking the goods corresponding to the first goods sensitivity value as a marker point, measuring the boundary distance value of each region boundary of the impact point area, and determining the minimum boundary distance value and the target region boundary corresponding to the minimum boundary distance value according to the boundary distance value; determining the second goods information corresponding to the non-evolution drone located at the target region boundary based on the drone matrix data and inputting the second goods information into a preset goods delivery standard model for identification to obtain a second goods sensitivity value corresponding to each goods in the second goods information; judging whether the second goods sensitivity value is greater than a preset sensitivity value, and if not, calculating the relative distance between each non-evolution drone in the target region boundary and the marker point, determining the target non-evolution drone with the closest relative distance to the marker point based on the relative distance, and generating a position replacement instruction based on the drone corresponding to the target non-evolution drone and the marker point to control the drone corresponding to the target non-evolution drone and the marker point to replace positions to obtain a position-replaced drone matrix; performing intercalation evolution analysis on the evolution drone in the position-replaced drone matrix and the drone in the position-replaced drone matrix corresponding to the non-target region boundary to obtain a matrix evolution route; According to the matrix evolution route, the position-replaced UAV matrix is subjected to obstacle avoidance flight optimization adjustment to obtain an optimized and adjusted UAV matrix; The optimized and adjusted UAV matrix is used to control the UAV matrix to evolve in the air to avoid obstacles. 2.The method for cooperative dynamic obstacle avoidance of UAV swarm according to claim 1, wherein, The obstacle analysis on the real-time field of view image and the UAV application data is performed to determine whether there is an aerial obstacle in the real-time field of view image that interferes with the UAV matrix, including: A UAV three-dimensional model of the UAV matrix in the flight process is built according to the UAV application data; According to the delivery goods information, the delivery goods are subjected to three-dimensional simulation processing, and a goods three-dimensional model obtained after the processing is model-fitted with the UAV three-dimensional model according to a loading correspondence and a preset ratio to obtain a delivery three-dimensional model; The real-time field of view image is subjected to image fusion processing to obtain a processed real-time field of view image; The processed real-time field of view image is input into a preset obstacle model for detection to determine whether there is an obstacle in the real-time field of view image, and if there is, a relative position relationship between the obstacle and the investigation UAV layer in the real-time field of view image is determined, and whether the obstacle reaches a preset flight range of the investigation UAV layer is determined according to the relative position relationship; If the obstacle does not reach the preset flight range of the investigation UAV layer, an obstacle evolution condition of the relative position relationship at different time nodes is determined, the obstacle evolution condition is predicted and deduced, a future obstacle evolution condition between the obstacle and the investigation UAV layer in a future time period is determined, an evolution three-dimensional model of the obstacle is built according to the future obstacle evolution condition, and the evolution three-dimensional model is model-fitted with the delivery three-dimensional model according to the relative position relationship to obtain a monitoring three-dimensional model; According to the monitoring three-dimensional model, whether there is an aerial obstacle in the real-time field of view image that interferes with the UAV matrix is determined. 3.The method of claim 2, wherein, The image fusion processing on the real-time field of view image to obtain a processed real-time field of view image includes: The real-time field of view image is subjected to edge filtering processing to obtain a primary layer and a secondary layer, the primary layer is a scene basic outline layer in the real-time field of view image, and the secondary layer is a scene detail outline layer in the real-time field of view image; The primary layer is divided according to a preset division rule to obtain a plurality of layer sub-regions; A sub-region histogram and a sub-region flatness corresponding to each layer sub-region and a global histogram and a global flatness corresponding to the primary layer are collected, and a primary tone curve and a light and dark intensity enhancement curve corresponding to the primary layer are determined according to the global histogram and the global flatness; A secondary tone curve and a curve fusion weight corresponding to each sub-region are determined according to a sub-region histogram and a sub-region flatness corresponding to each layer sub-region, the primary tone curve and the secondary tone curve are subjected to fusion processing to obtain a tone fusion curve; Fusing the tone fusion curve with the light and dark enhancement curve based on the curve fusion weight, and adjusting each sub-region in the primary layer according to the fused visual field adjustment curve to obtain an adjusted primary layer; Fusing the adjusted primary layer with the secondary layer to obtain a processed real-time visual field image.

4. The method of claim 1, wherein, The judgment of whether the second cargo sensitivity value is greater than the preset sensitivity value comprises: If the second cargo sensitivity value is greater than the preset sensitivity value, then according to the boundary distance value, the third cargo information corresponding to the non-evolution unmanned aerial vehicle of each region boundary is determined one by one, the third cargo information is input into a preset cargo distribution standard model for identification, and a third cargo sensitivity value corresponding to each cargo in the third cargo information is obtained; If the third cargo sensitivity value is still greater than the preset sensitivity value, then the center point position of the unmanned aerial vehicle matrix and the impact center point position of the impact point area are determined, and the center point position and the impact center point position are connected by a position line to obtain a matrix movement vector line; According to the movement vector line, a matrix obstacle avoidance route is generated to control the flight route movement of the unmanned aerial vehicle matrix.

5. The method of claim 4, wherein, The method of controlling the unmanned aerial vehicle matrix to avoid obstacles in the air based on the optimized and adjusted unmanned aerial vehicle matrix further comprises: After detecting that the unmanned aerial vehicle matrix flies through the aerial obstacle, the optimized and adjusted unmanned aerial vehicle matrix is backtracked according to the matrix evolution route or the unmanned aerial vehicle matrix is backtracked according to the movement vector line to obtain a backtracked unmanned aerial vehicle matrix.

6. The method of claim 1, wherein, The method of controlling the unmanned aerial vehicle matrix to avoid obstacles in the air based on the optimized and adjusted unmanned aerial vehicle matrix further comprises: Network communication information for controlling the unmanned aerial vehicle matrix is collected, and network state detection is performed on the network communication information to determine whether the network state in the network communication information meets a preset network state. If not, the network information of a region within a preset range is obtained with the unmanned aerial vehicle matrix as the center, network channel information is determined according to the region network information, and a network channel group excluding the current application channel in the network communication information is determined based on the network channel information; Each network channel corresponding signal transmission rate is obtained by testing the network channel group one by one. The signal transmission rates are screened and combined, and the network channels are reorganized according to the transmission order corresponding to each signal transmission rate in the screened and combined network channel group information to obtain a splicing network channel for sending control instructions to the unmanned aerial vehicle matrix.

7. A cooperative dynamic obstacle avoidance system for a swarm of UAVs, characterized in that, It comprises: An information acquisition module is configured to acquire unmanned aerial vehicle group information and distribution cargo information carried by each unmanned aerial vehicle in the unmanned aerial vehicle group; An application determination module is configured to determine unmanned aerial vehicle application data and unmanned aerial vehicle application numbers according to the unmanned aerial vehicle group information; A matrix generation module is configured to construct an unmanned aerial vehicle matrix based on the unmanned aerial vehicle application numbers, bind the distribution cargo information with the unmanned aerial vehicle matrix, and obtain unmanned aerial vehicle matrix data. An image acquisition module is configured to determine a surveillance drone layer according to the drone matrix, and acquire real-time field of view images of the drone matrix in the cargo delivery process based on the surveillance drone layer. An obstacle analysis module is configured to perform obstacle analysis on the real-time field of view images and the drone application data, determine whether there is an aerial obstacle in the real-time field of view images that interferes with the drone matrix, if there is, perform static and dynamic analysis on the aerial obstacle, and based on the analysis result and the drone matrix data, perform obstacle avoidance flight optimization adjustment on the drone matrix to obtain an optimized and adjusted drone matrix. When the obstacle analysis module performs static and dynamic analysis on the aerial obstacle, and based on the analysis result and the drone matrix data, performs obstacle avoidance flight optimization adjustment on the drone matrix to obtain an optimized and adjusted drone matrix, the obstacle analysis module is specifically configured to: determine a collision point area of the aerial obstacle and the drone matrix in a future time period according to the monitoring three-dimensional model, and determine an evolved drone involved in dynamic adjustment in the drone matrix based on the collision point area; determine first cargo information corresponding to each evolved drone based on the drone matrix data, and input the first cargo information into a preset cargo delivery standard model for identification to obtain a first cargo sensitivity value corresponding to each cargo in the first cargo information; determine whether the first cargo sensitivity value is greater than a preset sensitivity value, if yes, take the cargo corresponding to the first cargo sensitivity value as a marker point, measure a boundary distance value of each region boundary of the collision point area, and determine a minimum boundary distance value and a target region boundary corresponding to the minimum boundary distance value according to the boundary distance value; determine second cargo information corresponding to a non-evolved drone located at the target region boundary based on the drone matrix data, and input the second cargo information into a preset cargo delivery standard model for identification to obtain a second cargo sensitivity value corresponding to each cargo in the second cargo information; determine whether the second cargo sensitivity value is greater than a preset sensitivity value, if no, calculate a relative distance between each non-evolved drone in the target region boundary and the marker point, determine a target non-evolved drone closest to the marker point according to the relative distance, and generate a position replacement instruction based on the target non-evolved drone and the drone corresponding to the marker point to control the target non-evolved drone and the drone corresponding to the marker point to replace positions to obtain a position-replaced drone matrix; perform intercalation evolution analysis on the evolved drone in the position-replaced drone matrix and the drone in the position-replaced drone matrix corresponding to a non-target region boundary to obtain a matrix evolution route; perform obstacle avoidance flight optimization adjustment on the position-replaced drone matrix according to the matrix evolution route to obtain an optimized and adjusted drone matrix; An obstacle avoidance optimization module is configured to control the drone matrix to perform aerial obstacle avoidance evolution based on the optimized and adjusted drone matrix.

8. An electronic device, comprising: A computer program stored on a memory and capable of being loaded and executed by a processor to perform the method of any one of claims 1 to 6 for cooperative dynamic obstacle avoidance of a swarm of unmanned aerial vehicles.

9. A computer-readable storage medium, characterized in that, A computer program stored on a memory and capable of being loaded and executed by a processor to perform the method of any one of claims 1 to 6 for cooperative dynamic obstacle avoidance of a swarm of unmanned aerial vehicles.

Images