Unmanned aerial vehicle positioning method and system based on satellite remote sensing technology

Abstract

The invention relates to the technical field of unmanned aerial vehicle positioning, and discloses an unmanned aerial vehicle positioning method based on a satellite remote sensing technology, and the method comprises the steps: recognizing the ground feature data of a pre-flight region of a to-be-positioned unmanned aerial vehicle, carrying out the key attribute marking of a virtual copy, and obtaining a marked virtual copy; identifying a feature map of the to-be-positioned unmanned aerial vehicle, and performing multi-channel processing and deep and shallow feature processing on the feature map to obtain a second processed feature map; performing preliminary positioning on the to-be-positioned unmanned aerial vehicle, collecting first environment data of the current position of the unmanned aerial vehicle, performing copy optimization on the marked virtual copy to obtain a target copy, and constructing a positioning scheme of the unmanned aerial vehicle by using the target copy, the initial position and the second processing feature map; and acquiring second environment data of the unmanned aerial vehicle during continuous flight in real time, performing continuous position correction on the to-be-positioned unmanned aerial vehicle in combination with the positioning scheme to obtain a revised position, and analyzing the real-time position of the to-be-positioned unmanned aerial vehicle in real time based on the revised position. According to the invention, the accuracy of unmanned aerial vehicle positioning can be improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle positioning technology, and in particular to a method and system for unmanned aerial vehicle positioning based on satellite remote sensing technology. Background Art

[0002] Satellite remote sensing technology plays an important role in the field of modern science and technology, and the precise positioning of drones is of key significance in many application scenarios. Developing a drone positioning method based on satellite remote sensing technology can achieve accurate positioning of the drone itself, providing strong guarantees for the safe flight and efficient operation of drones. At the same time, accurate positioning can ensure that the drone knows its own position accurately when performing tasks, better plan flight paths and perform mission operations, and improve the success rate and quality of mission execution.

[0003] At present, the positioning of drones is generally carried out through traditional GPS positioning technology or ground base station positioning technology. However, the positioning accuracy of these methods may be limited in some complex environments. For example, in areas with severe signal shielding such as urban canyons and mountainous areas, the positioning error is large. As the application scenarios of drones are becoming more and more extensive, including field exploration, aerial photography and mapping, and logistics distribution, these scenarios often require higher-precision positioning to ensure the accurate location and safe flight of drones. Traditional positioning methods are difficult to meet the drone's own needs for high-precision positioning when facing complex terrain and changing environments, resulting in inaccurate positioning of drones. Summary of the invention

[0004] The present invention provides a method and system for positioning an unmanned aerial vehicle (UAV) based on satellite remote sensing technology, the main purpose of which is to improve the accuracy of positioning the UAV.

[0005] To achieve the above object, the present invention provides a UAV positioning method based on satellite remote sensing technology, comprising:

[0006] Identify a pre-flight area of ​​the unmanned aerial vehicle to be located, collect ground object data of the pre-flight area, construct a virtual copy of the pre-flight area using the ground object data, identify key data of the ground object data, and use the key data to annotate key attributes of the virtual copy to obtain an annotated virtual copy;

[0007] Identify a drone feature map of the drone to be located, perform multi-channel processing on the drone feature map to obtain a first processed feature map, and perform depth and shallow feature processing on the first processed feature map to obtain a second processed feature map;

[0008] The UAV to be located is communicated and connected with the UAV to be located through a preconfigured satellite remote sensing device, and the UAV to be located is preliminarily located to obtain the initial position of the UAV to be located; the first environmental data of the UAV to be located at the initial position is collected, and the first environmental data is used to optimize the copy of the annotated virtual copy to obtain a target copy, and the target copy, the initial position and the second processing feature map are used to construct a positioning solution for the UAV to be located;

[0009] The second environmental data of the UAV to be located is collected in real time during its continuous flight, and the second environmental data is combined with the positioning scheme to perform continuous position correction on the UAV to be located to obtain a revised position. Based on the revised position, the real-time position of the UAV to be located after the position revision is analyzed in real time.

[0010] Optionally, identifying a pre-flight area of ​​the drone to be located includes:

[0011] Obtaining a pre-flight path of the UAV to be located;

[0012] Based on the pre-flight path, determining the flight target point of the unmanned aerial vehicle to be located;

[0013] Querying the flight mission of the drone to be located;

[0014] Based on the flight mission, determining the flight coverage of the drone to be located;

[0015] Based on the flight target point and the flight coverage range, construct a pre-flight three-dimensional map of the unmanned aerial vehicle to be positioned;

[0016] Identifying flight prohibitions within the flight coverage area;

[0017] Based on the flight prohibition, optimizing the view of the pre-flight three-dimensional image to obtain an optimized view;

[0018] According to the optimized view, a pre-flight area of ​​the UAV to be positioned is determined.

[0019] Optionally, the using the terrain data to construct a virtual copy of the pre-flight area includes:

[0020] Using the ground feature data, construct a three-dimensional model of the pre-flight area;

[0021] Identifying spatial coordinate values ​​in the three-dimensional model;

[0022] Based on the spatial coordinate value, the three-dimensional model is adjusted in geographical height to obtain a model height adjustment value;

[0023] Based on the model height adjustment value, the height of the three-dimensional model is adjusted to obtain a virtual copy.

[0024] Optionally, the using the key data to mark the virtual copy with key attributes to obtain the marked virtual copy includes:

[0025] Performing labeling processing on the key data to obtain label data;

[0026] constructing annotation standards and annotation hierarchies of the virtual replica;

[0027] Based on the label data, the labeling standard and the labeling hierarchy, the virtual copy is labeled with key attributes to obtain a labeled virtual copy.

[0028] Optionally, performing multi-channel processing on the drone feature map to obtain a first processed feature map includes:

[0029] Performing three-channel decomposition on the UAV feature map to obtain a three-channel feature map;

[0030] Extracting the color features of the first channel, the color features of the second channel, and the texture features of the third channel in the three-channel feature map;

[0031] Feature processing is performed on the color features of the first channel, the color features of the second channel, and the texture features of the third channel to obtain a first processed feature map.

[0032] Optionally, performing deep and shallow feature processing on the first processed feature map to obtain a second processed feature map includes:

[0033] Extracting deep features of the first processed feature map;

[0034] Querying shallow features of the first processed feature graph;

[0035] Convert the deep features and the shallow features into feature vectors to obtain a deep vector and a shallow vector;

[0036] Performing vector fusion on the deep vector and the shallow vector to obtain a fused vector;

[0037] Vector compilation is performed on the fused vector to obtain a second processed feature map.

[0038] Optionally, using the target copy, the initial position and the second processed feature map to construct a positioning solution for the unmanned aerial vehicle to be positioned includes:

[0039] Querying the satellite signal strength of the initial position;

[0040] When the satellite signal strength is strong, analyzing the environmental complexity of the initial position;

[0041] Based on the environmental complexity, determining a sensor usage strategy for the drone to be located;

[0042] Based on the sensor usage strategy, construct a first positioning solution for the drone to be positioned;

[0043] When the satellite signal strength is weak, constructing a second positioning solution for the drone to be positioned using the target copy;

[0044] Based on the first positioning solution and the second positioning solution, a positioning solution for the drone to be positioned is determined.

[0045] Optionally, the using the second environmental data in combination with the positioning scheme to continuously correct the position of the unmanned aerial vehicle to be positioned to obtain a revised position includes:

[0046] When the positioning scheme is the corresponding first positioning scheme, using the second environmental data to create a data update frequency scheme for the drone to be positioned;

[0047] Based on the data update frequency scheme, performing a first position correction on the drone to be located to obtain a first corrected position;

[0048] When the positioning scheme is the corresponding second positioning scheme, analyzing the position change state of the unmanned aerial vehicle to be positioned by using the second environmental data;

[0049] Performing a second position correction on the drone to be located based on the position change state to obtain a second corrected position;

[0050] Based on the first corrected position and the second corrected position, the position of the unmanned aerial vehicle to be located is continuously corrected to obtain a revised position.

[0051] Optionally, the using the second environmental data to analyze the position change state of the unmanned aerial vehicle to be located includes:

[0052] Extracting the wind speed and air density of the drone to be located in the second environmental data;

[0053] Query the flight speed of the drone to be located;

[0054] Calculating the real-time aerodynamic force of the UAV to be located based on the wind speed, the flight speed and the air density;

[0055] Based on the real-time aerodynamic force, calculating the three-axis speed of the UAV to be positioned;

[0056] Based on the three-axis speed, calculating the moment position of the drone to be located;

[0057] Based on the momentary position, the position change state of the drone to be located is analyzed.

[0058] In order to solve the above problems, the present invention also provides a UAV positioning system based on satellite remote sensing technology, the system comprising:

[0059] A virtual copy construction module is used to identify a pre-flight area of ​​the UAV to be located, collect ground object data of the pre-flight area, use the ground object data to construct a virtual copy of the pre-flight area, identify key data of the ground object data, and use the key data to annotate key attributes of the virtual copy to obtain an annotated virtual copy;

[0060] A feature map processing module is used to identify the drone feature map of the drone to be located, perform multi-channel processing on the drone feature map to obtain a first processed feature map, and perform depth and shallow feature processing on the first processed feature map to obtain a second processed feature map;

[0061] A positioning solution construction module is used to establish a communication connection with the unmanned aerial vehicle to be positioned through a preconfigured satellite remote sensing device, perform preliminary positioning of the unmanned aerial vehicle to be positioned, and obtain an initial position of the unmanned aerial vehicle to be positioned; collect first environmental data of the unmanned aerial vehicle to be positioned at the initial position, use the first environmental data to perform copy optimization on the annotated virtual copy to obtain a target copy, and use the target copy, the initial position and the second processing feature map to construct a positioning solution for the unmanned aerial vehicle to be positioned;

[0062] The UAV position analysis module is used to collect the second environmental data of the UAV to be located in real time during its continuous flight, and use the second environmental data in combination with the positioning scheme to continuously correct the position of the UAV to be located to obtain a revised position. Based on the revised position, the real-time position of the UAV to be located after the position revision is analyzed in real time.

[0063] The present invention can provide rich details for the virtual copy of the pre-flight area by collecting the ground object data of the pre-flight area, so that the virtual copy is closer to the real environment; the present invention can extract the unique visual features of the drone by identifying the drone feature map of the drone to be located, so as to accurately identify and locate the drone in different environments; the present invention uses a pre-configured satellite remote sensing device to communicate with the drone to be located, and then preliminarily locates the drone to be located, and obtains the initial position of the drone to be located, which can realize real-time data transmission, so that the satellite remote sensing device can obtain relevant information of the drone, and improve the accuracy and timeliness of positioning; the present invention can timely understand the changes in environmental data by real-time collection of the second environmental data of the drone to be located during continuous flight, so as to make more accurate corrections and analysis of the position of the drone; finally, the present invention can analyze the real-time position of the drone to be located after the position is revised based on the revised position, so that the user or the drone itself can understand its current position state, so as to make corresponding decisions and controls. Therefore, the present invention can improve the accuracy of drone positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A schematic diagram of a flow chart of a method for positioning a UAV based on satellite remote sensing technology provided by an embodiment of the present invention;

[0065] Figure 2 A functional module diagram of a UAV positioning system based on satellite remote sensing technology provided by one embodiment of the present invention;

[0066] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0067] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0068] The embodiment of the present application provides a method for positioning a UAV based on satellite remote sensing technology. The execution subject of the method for positioning a UAV based on satellite remote sensing technology includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for positioning a UAV based on satellite remote sensing technology can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0069] Reference Figure 1 FIG. 1 is a flow chart of a method for positioning a UAV based on satellite remote sensing technology according to an embodiment of the present invention. In this embodiment, the method for positioning a UAV based on satellite remote sensing technology includes:

[0070] S1. Identify the pre-flight area of ​​the UAV to be located, collect ground object data of the pre-flight area, use the ground object data to construct a virtual copy of the pre-flight area, identify key data of the ground object data, use the key data to annotate key attributes of the virtual copy, and obtain an annotated virtual copy.

[0071] The embodiment of the present invention can determine the possible activity range of the drone by identifying the pre-flight area of ​​the drone to be located, wherein the pre-flight area refers to the area that no mission will fly through after taking off.

[0072] As an embodiment of the present invention, the identification of a pre-flight area of ​​the UAV to be located includes: obtaining a pre-flight path of the UAV to be located, determining a flight target point of the UAV to be located based on the pre-flight path, querying a flight mission of the UAV to be located, determining a flight coverage range of the UAV to be located based on the flight mission, constructing a pre-flight three-dimensional map of the UAV to be located based on the flight target point and the flight coverage range, identifying flight prohibitions in the flight coverage range, optimizing the view of the pre-flight three-dimensional map based on the flight prohibitions to obtain an optimized view, and determining the pre-flight area of ​​the UAV to be located according to the optimized view.

[0073] Among them, the pre-flight path refers to the flight route planned before the drone takes off, usually including the starting point, the end point and the waypoints in the middle; the flight mission refers to the specific task that the drone needs to perform, such as surveying, inspection, logistics distribution, etc.; the flight coverage range refers to the area that the drone needs to cover when performing the flight mission; the pre-flight three-dimensional map refers to the three-dimensional map constructed before the drone takes off, showing the drone's pre-flight area and surrounding environment; the flight prohibition refers to the regulations to ensure that the drone flies within the legal and compliant scope; the flight target point refers to the specific location that the drone needs to reach during the performance of the flight mission

[0074] Optionally, the pre-flight path and the flight mission can be obtained through a mission planning system, the pre-flight three-dimensional map can be constructed by using a three-dimensional modeling function in GIS software, and the pre-flight three-dimensional map is optimized based on the flight prohibition, and the optimized view can be obtained by using different colors, icons or transparency to distinguish between prohibited areas and flyable areas. For example, the flight prohibition area is marked in red and its transparency is reduced so that the operator can more easily identify it.

[0075] Furthermore, the embodiments of the present invention can provide rich details for the virtual copy of the pre-flight area by collecting the ground object data of the pre-flight area, making the virtual copy closer to the real environment. The ground object data refers to the characteristic information of various natural and man-made objects in the pre-flight area, which can be collected by sensors such as satellite images, aerial photography, ground measurement equipment, etc.

[0076] In the embodiment of the present invention, by using the ground feature data to construct a virtual copy of the pre-flight area, the flight status of the drone can be simulated, which helps the user to determine the location of the drone. The virtual copy refers to a virtual environment that simulates the real pre-flight area constructed by digital technology.

[0077] As an embodiment of the present invention, the use of the terrain data to construct a virtual copy of the pre-flight area includes: using the terrain data to construct a three-dimensional model of the pre-flight area, identifying spatial coordinate values ​​in the three-dimensional model, based on the spatial coordinate values, performing geographic height adjustment on the three-dimensional model to obtain a model height adjustment value, and based on the model height adjustment value, performing height adjustment on the three-dimensional model to obtain a virtual copy.

[0078] Optionally, the use of the terrain data to construct the three-dimensional model of the pre-flight area can be created by combining the terrain data with a Java script, the identification of the spatial coordinate values ​​in the three-dimensional model can be achieved by using global positioning system (GPS) coordinates or a local coordinate system, and the three-dimensional model is geographically adjusted based on the spatial coordinate values ​​to obtain the model height adjustment value by calculating the difference between the spatial coordinate value and its corresponding reference height data, wherein the difference is the height adjustment value. An interpolation algorithm, such as inverse distance weighted interpolation, Kriging interpolation, etc., can be used to interpolate the reference height data to each point of the three-dimensional model to obtain a more accurate height adjustment value.

[0079] Furthermore, the embodiments of the present invention can match and identify the key data of the feature data as key features to improve the positioning accuracy of the drone. The key data refers to specific data with key importance and significant features in the feature data of the pre-flight area, such as tall skyscrapers, landmark buildings with unique shapes, etc. The feature recognition of the feature data, such as height, area, shape complexity, etc., can be analyzed through feature engineering. Generally, buildings with high heights, large areas of water, or features with unique shapes may be identified as key data. A height threshold can be set, and buildings above the threshold are regarded as key data.

[0080] Furthermore, the embodiment of the present invention utilizes the key data to annotate the virtual copy with key attributes, and the annotated virtual copy can help the user quickly identify key features and improve the speed of determining the position of the drone.

[0081] As an embodiment of the present invention, the key data is used to mark the key attributes of the virtual copy to obtain a marked virtual copy, including: labeling the key data to obtain label data, constructing a labeling standard and a labeling hierarchy for the virtual copy, and based on the label data, the labeling standard and the labeling hierarchy, marking the key attributes of the virtual copy to obtain a marked virtual copy.

[0082] Among them, the annotation standard refers to a series of specifications and requirements formulated to ensure the consistency, accuracy and readability of annotations, and the annotation hierarchy refers to the criteria for classifying and organizing different types of key features according to certain logical relationships so as to facilitate rapid retrieval and identification in virtual copies.

[0083] Optionally, the label data can be obtained by adding specific attribute labels such as name, height, color, etc. to the key data through a script constructed by Java. The labeling standard can be obtained by

[0084] The content, format, color coding, etc. of the annotation are formulated. For example, for key buildings, their names, operating status and other information can be annotated. The annotation hierarchy can be classified and annotated by different types of key features. For example, key features can be divided into natural landscapes, historical sites, modern buildings, infrastructure and other categories, and each category is marked with different colors and icons.

[0085] S2. Identify the drone feature map of the drone to be located, perform multi-channel processing on the drone feature map to obtain a first processed feature map, and perform depth and shallow feature processing on the first processed feature map to obtain a second processed feature map.

[0086] The embodiment of the present invention can extract the unique visual features of the drone by identifying the drone feature map of the drone to be located, so as to accurately identify and locate the drone in different environments. The feature map refers to an image representation of the appearance, structure and other features of the drone, which can be recognized by using an image recognition algorithm, such as a convolutional neural network.

[0087] Furthermore, the embodiment of the present invention performs multi-channel processing on the drone feature map to obtain a first processed feature map, which can extract information of the drone feature map from different angles, thereby increasing the richness and diversity of features.

[0088] As an embodiment of the present invention, the multi-channel processing of the UAV feature map to obtain a first processed feature map includes: performing three-channel decomposition on the UAV feature map to obtain a three-channel feature map, extracting color features of the first channel, color features of the second channel, and texture features of the third channel in the three-channel feature map, and performing feature processing on the color features of the first channel, the color features of the second channel, and the texture features of the third channel to obtain a first processed feature map.

[0089] The first channel refers to a color channel, the second channel refers to a texture channel, and the third channel refers to an edge channel.

[0090] Optionally, the drone feature map is subjected to three-channel decomposition, and the obtained three-channel feature map can be decomposed by a filter group, the color feature can be extracted by an RGB channel method, the texture feature can be extracted by a texture filter, and the edge feature can be extracted by using an edge detection algorithm, such as Canny edge detection. The first processed feature map can be obtained by performing feature recognition on the color feature, the color feature and the texture feature, and performing feature translation using a convolutional layer in a deep learning model.

[0091] Furthermore, the embodiment of the present invention performs depth and shallow feature processing on the first processed feature map to obtain a second processed feature map, which can improve the recognition and positioning accuracy of the drone, especially in complex environments, and can better cope with problems such as lighting changes and occlusions.

[0092] As an embodiment of the present invention, the deep and shallow feature processing of the first processed feature map to obtain a second processed feature map includes: extracting deep features of the first processed feature map, querying shallow features of the first processed feature map, converting the deep features and the shallow features into feature vectors to obtain deep vectors and shallow vectors, performing vector fusion on the deep vector and the shallow vector to obtain a fused vector, and performing vector compilation on the fused vector to obtain a second processed feature map.

[0093] The deep features refer to the features obtained by performing multiple convolutions and nonlinear transformations on the original image. They contain high-level semantic information of the image, such as abstract concepts such as the category, shape, and texture of the object.

[0094] Optionally, the deep features can be obtained by inputting the first processed feature map into a pre-trained convolutional neural network and extracting it using the pre-trained convolutional neural network; the converting the deep features and the shallow features into feature vectors to obtain deep vectors and shallow vectors can be achieved through principal component analysis; the vector fusion of the deep vectors and the shallow vectors to obtain a fused vector can be obtained by assigning different weights to the deep vectors and the shallow vectors by weighted averaging, and then weighted summing them; the vector compilation of the fused vector to obtain the second processed feature map can be obtained through SIMD technology processing.

[0095] S3. Communicate and connect with the UAV to be located through a preconfigured satellite remote sensing device, preliminarily locate the UAV to be located, obtain the initial position of the UAV to be located, collect the first environmental data of the UAV to be located at the initial position, use the first environmental data to optimize the marked virtual copy to obtain a target copy, and use the target copy, the initial position and the second processed feature map to construct a positioning solution for the UAV to be located.

[0096] The embodiment of the present invention achieves real-time data transmission by communicating with the unmanned aerial vehicle to be located through the pre-configured satellite remote sensing device, performing preliminary positioning of the unmanned aerial vehicle to be located, and obtaining the initial position of the unmanned aerial vehicle to be located, so as to enable the satellite remote sensing device to obtain relevant information of the unmanned aerial vehicle, thereby improving the accuracy and timeliness of positioning. The satellite remote sensing device is a device system that uses sensors carried by satellites to obtain information on the earth's surface, including optical sensors, data transmission systems, and satellite platforms.

[0097] Optionally, the initial position of the drone to be positioned can be obtained by using a global positioning system (GPS).

[0098] Furthermore, the embodiment of the present invention can provide real-time information about the surrounding environment of the drone when it takes off, including meteorological conditions, topography, obstacle distribution, etc., by collecting the first environmental data of the drone to be positioned at the initial position. The first environmental data refers to environmental data when the drone has not taken off or has just taken off. The first environmental data can be obtained by using meteorological sensors to obtain meteorological information such as temperature, humidity, and air pressure, and using laser radar or cameras to obtain terrain and obstacle information.

[0099] The embodiment of the present invention optimizes the annotated virtual copy by utilizing the first environment data, and obtains a target copy by updating terrain data in the virtual copy, adding obstacle annotations, adjusting meteorological parameters, etc., so that the copy is closer to the actual environment.

[0100] Optionally, the process of using the first environmental data to perform copy optimization on the annotated virtual copy to obtain the target copy is called: fusing the collected first environmental data with the annotated virtual copy, and updating the terrain data in the virtual copy, adding obstacle annotations, adjusting meteorological parameters, etc. For example, if the lidar detects a new building nearby, the model and location information of the building and the environmental information of the area, such as strong airflow, can be added to the virtual copy.

[0101] The embodiment of the present invention can comprehensively utilize multiple information sources to construct a positioning solution for the unmanned aerial vehicle to be positioned by utilizing the target copy, the initial position and the second processed feature map, thereby improving the accuracy and reliability of unmanned aerial vehicle positioning. The positioning solution refers to a positioning method that takes into account real-time flight information of the unmanned aerial vehicle, such as speed information, environmental information, etc.

[0102] As an embodiment of the present invention, the method of constructing a positioning scheme for the unmanned aerial vehicle to be positioned by using the target replica, the initial position and the second processing characteristic graph includes: querying the satellite signal strength of the initial position, analyzing the environmental complexity of the initial position when the satellite signal strength is strong, determining the sensor usage strategy of the unmanned aerial vehicle to be positioned based on the environmental complexity, constructing a first positioning scheme for the unmanned aerial vehicle to be positioned based on the sensor usage strategy, constructing a second positioning scheme for the unmanned aerial vehicle to be positioned using the target replica when the satellite signal strength is weak, and determining the positioning scheme for the unmanned aerial vehicle to be positioned based on the first positioning scheme and the second positioning scheme. The environmental complexity refers to the complexity of the flight environment in which the unmanned aerial vehicle is located in multiple aspects. The first positioning scheme refers to a method of positioning by sensors, and the second positioning scheme refers to a method of positioning without using sensors.

[0103] Optionally, the satellite signal strength can be evaluated by parameter indicators of the satellite signal receiving device (such as signal-to-noise ratio, signal power, etc.), the environmental complexity of the initial position is analyzed by terrain factors, meteorological factors and obstacle factors, and the process of determining the sensor usage strategy of the UAV to be located based on the environmental complexity is: using satellite positioning to locate the UAV to be located in a simple environment, and using a visual sensor to locate it in a complex environment. The process of using the target copy to construct a second positioning scheme for the UAV to be located is: using the target copy in combination with the flight time, flight altitude and flight speed of the UAV to be located to simulate the flight process of the UAV to be located, and using the flight process to determine the position of the UAV to be located.

[0104] S4. Collect in real time the second environmental data of the UAV to be located during its continuous flight, use the second environmental data in combination with the positioning scheme to perform continuous position correction on the UAV to be located to obtain a revised position, and based on the revised position, analyze in real time the real-time position of the UAV to be located after the position revision.

[0105] The embodiment of the present invention can timely understand the changes in environmental data by real-time collection of the second environmental data of the unmanned aerial vehicle to be located during continuous flight, so as to more accurately correct and analyze the position of the unmanned aerial vehicle. The second environmental data refers to the real-time environmental data of the unmanned aerial vehicle during flight, which can be collected by multiple sensors.

[0106] Furthermore, the embodiment of the present invention uses the second environmental data in combination with the positioning scheme to continuously correct the position of the drone to be located, and the revised position obtained can improve the positioning accuracy of the drone, ensuring that the user can always accurately know the location of the drone.

[0107] As an embodiment of the present invention, the method of using the second environmental data in combination with the positioning scheme to continuously correct the position of the drone to be located to obtain a revised position includes: when the positioning scheme is the corresponding first positioning scheme, using the second environmental data to create a data update frequency scheme for the drone to be located, based on the data update frequency scheme, performing a first position correction on the drone to be located to obtain a first corrected position, when the positioning scheme is the corresponding second positioning scheme, using the second environmental data to analyze the position change state of the drone to be located, based on the position change state, performing a second position correction on the drone to be located to obtain a second corrected position, and based on the first corrected position and the second corrected position, continuously correcting the position of the drone to be located to obtain a revised position.

[0108] Among them, the data update frequency scheme refers to a method for collecting the frequency of drone information in real time. It should be noted that high-frequency data updates require more computing resources and power consumption. Therefore, the data update frequency should be reasonably determined based on the hardware performance and endurance of the drone.

[0109] Optionally, the data update frequency scheme can be created according to the speed of environmental changes. For example, if the flight environment changes rapidly, such as under strong winds, turbulence and other meteorological conditions, a higher data update frequency is required to track the position changes of the UAV in real time. Based on the data update frequency scheme, a first position correction is performed on the UAV to be located, and the first corrected position is obtained by using sensors to perform real-time position monitoring, and the position of the UAV is continuously updated.

[0110] Further, in an optional embodiment of the present invention, the use of the second environmental data to analyze the position change state of the unmanned aerial vehicle to be located includes: extracting the wind speed and air density of the unmanned aerial vehicle to be located in the second environmental data, querying the flight speed of the unmanned aerial vehicle to be located, and based on the wind speed, the flight speed and the air density, calculating the real-time aerodynamic force of the unmanned aerial vehicle to be located using the following formula:

[0111]

[0112] Among them, F a represents the real-time aerodynamic force, E represents the air resistance coefficient, ρ represents the air density, Indicates wind speed, represents the flight speed, and H represents the windward area of ​​the UAV to be located;

[0113] Based on the real-time aerodynamic force, the three-axis speed of the UAV to be positioned is calculated using the following formula:

[0114]

[0115] P=(a x +a y +a z

[0116] Among them, P represents the three-axis speed, a x Indicates the x-axis speed of the drone to be located, a y Indicates the y-axis speed of the drone to be located, a z represents the z-axis speed of the drone to be located, m represents the mass of the drone, g represents the acceleration of gravity, δ represents the angle between the flight trajectory of the drone to be located and the horizontal plane, and F a represents the real-time aerodynamic force, F ax represents the real-time aerodynamic force of the UAV to be positioned on the x-axis, F ay represents the real-time aerodynamic force of the UAV to be positioned on the y-axis, F az Indicates the real-time aerodynamic force of the drone to be positioned on the z-axis;

[0117] Based on the three-axis speed, the moment position of the drone to be located is calculated using the following formula:

[0118]

[0119] Among them, τ represents the time position, x t+1 represents the x-axis position of the drone to be located at time t+1, x t represents the x-axis position of the drone to be located at time t, v x represents the acceleration of the drone to be located on the x-axis, t represents time, and a x represents the x-axis speed of the drone to be located, represents the y-axis position of the drone to be located at time t+1, t represents the y-axis position of the drone to be located at time t, v y Indicates the acceleration of the UAV to be positioned on the y-axis, a y Indicates the y-axis speed of the drone to be located, z t+1 represents the z-axis position of the drone to be located at time t+1, z t represents the z-axis position of the drone to be located at time t, v z Indicates the acceleration of the drone to be positioned on the z-axis, a z Indicates the z-axis speed of the drone to be located;

[0120] Based on the momentary position, the position change state of the drone to be located is analyzed.

[0121] Furthermore, the embodiment of the present invention can analyze the real-time position of the drone to be located after the position is revised based on the revised position in real time, so that the user or the drone itself can understand its current position state, so as to make corresponding decisions and controls. For example, if the drone is found to deviate from the predetermined flight path, the flight direction can be adjusted in time, and if the drone is found to be close to a dangerous area, avoidance measures can be taken.

[0122] At the same time, real-time location analysis can also provide a reference for subsequent mission planning and execution. For example, if the drone is performing a search and rescue mission, the search area and search strategy can be determined based on real-time location analysis.

[0123] like Figure 2 , which is a functional module diagram of a UAV positioning system based on satellite remote sensing technology provided by one embodiment of the present invention.

[0124] The UAV positioning system 200 based on satellite remote sensing technology described in the present invention can be installed in an electronic device. According to the functions implemented, the UAV positioning system 200 based on satellite remote sensing technology can include a virtual copy construction module 201, a feature map processing module 202, a positioning solution construction module 203 and a UAV position analysis module 204. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0125] In this embodiment, the functions of each module / unit are as follows:

[0126] The virtual copy construction module 201 is used to identify the pre-flight area of ​​the drone to be located, collect ground object data of the pre-flight area, use the ground object data to construct a virtual copy of the pre-flight area, identify key data of the ground object data, and use the key data to annotate key attributes of the virtual copy to obtain an annotated virtual copy;

[0127] The feature map processing module 202 is used to identify the drone feature map of the drone to be located, perform multi-channel processing on the drone feature map to obtain a first processed feature map, and perform depth and shallow feature processing on the first processed feature map to obtain a second processed feature map;

[0128] The positioning solution construction module 203 is used to communicate with the unmanned aerial vehicle to be positioned through a preconfigured satellite remote sensing device, perform preliminary positioning on the unmanned aerial vehicle to be positioned, and obtain the initial position of the unmanned aerial vehicle to be positioned; collect the first environmental data of the unmanned aerial vehicle to be positioned at the initial position, use the first environmental data to perform copy optimization on the annotated virtual copy to obtain a target copy, and use the target copy, the initial position and the second processing feature map to construct a positioning solution for the unmanned aerial vehicle to be positioned;

[0129] The UAV position analysis module 204 is used to collect the second environmental data of the UAV to be located in real time during its continuous flight, and use the second environmental data in combination with the positioning scheme to continuously correct the position of the UAV to be located to obtain a revised position, and based on the revised position, analyze the real-time position of the UAV to be located after the position is revised.

[0130] In detail, the modules described in the UAV positioning system 200 based on satellite remote sensing technology described in the embodiment of the present invention adopt the same technical means as the UAV positioning method based on satellite remote sensing technology described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0131] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A UAV positioning method based on satellite remote sensing technology, characterized in that: The method comprises: Identify a pre-flight area of ​​the unmanned aerial vehicle to be located, collect ground object data of the pre-flight area, construct a virtual copy of the pre-flight area using the ground object data, identify key data of the ground object data, and use the key data to annotate key attributes of the virtual copy to obtain an annotated virtual copy; Identify a drone feature map of the drone to be located, perform multi-channel processing on the drone feature map to obtain a first processed feature map, and perform depth and shallow feature processing on the first processed feature map to obtain a second processed feature map; The UAV to be located is communicated and connected with the UAV to be located through a preconfigured satellite remote sensing device, and the UAV to be located is preliminarily located to obtain the initial position of the UAV to be located; the first environmental data of the UAV to be located at the initial position is collected, and the first environmental data is used to optimize the copy of the annotated virtual copy to obtain a target copy, and the target copy, the initial position and the second processing feature map are used to construct a positioning solution for the UAV to be located; The second environmental data of the UAV to be located is collected in real time during its continuous flight, and the second environmental data is combined with the positioning scheme to perform continuous position correction on the UAV to be located to obtain a revised position. Based on the revised position, the real-time position of the UAV to be located after the position revision is analyzed in real time.

2. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The step of identifying the pre-flight area of ​​the drone to be located includes: Obtaining a pre-flight path of the UAV to be located; Based on the pre-flight path, determining the flight target point of the unmanned aerial vehicle to be located; Querying the flight mission of the drone to be located; Based on the flight mission, determining the flight coverage of the drone to be located; Based on the flight target point and the flight coverage range, construct a pre-flight three-dimensional map of the unmanned aerial vehicle to be positioned; Identifying flight prohibitions within the flight coverage area; Based on the flight prohibition, optimizing the pre-flight three-dimensional map to obtain an optimized view; According to the optimized view, a pre-flight area of ​​the UAV to be positioned is determined.

3. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The step of constructing a virtual copy of the pre-flight area by using the terrain data includes: Using the ground feature data, construct a three-dimensional model of the pre-flight area; Identifying spatial coordinate values ​​in the three-dimensional model; Based on the spatial coordinate value, the three-dimensional model is adjusted in geographical height to obtain a model height adjustment value; Based on the model height adjustment value, the height of the three-dimensional model is adjusted to obtain a virtual copy.

4. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The step of using the key data to mark the key attributes of the virtual copy to obtain the marked virtual copy includes: Performing labeling processing on the key data to obtain label data; constructing annotation standards and annotation hierarchies of the virtual replica; Based on the label data, the labeling standard and the labeling hierarchy, the virtual copy is labeled with key attributes to obtain a labeled virtual copy.

5. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The performing multi-channel processing on the drone feature graph to obtain a first processed feature graph includes: Performing three-channel decomposition on the UAV feature map to obtain a three-channel feature map; Extracting the color features of the first channel, the color features of the second channel, and the texture features of the third channel in the three-channel feature map; Feature processing is performed on the color features of the first channel, the color features of the second channel, and the texture features of the third channel to obtain a first processed feature map.

6. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The step of performing deep and shallow feature processing on the first processed feature map to obtain a second processed feature map includes: Extracting deep features of the first processed feature map; Querying shallow features of the first processed feature graph; Convert the deep features and the shallow features into feature vectors to obtain a deep vector and a shallow vector; Performing vector fusion on the deep vector and the shallow vector to obtain a fused vector; Vector compilation is performed on the fused vector to obtain a second processed feature map.

7. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The method of constructing a positioning solution for the unmanned aerial vehicle to be positioned by using the target copy, the initial position and the second processed feature map includes: Querying the satellite signal strength of the initial position; When the satellite signal strength is strong, analyzing the environmental complexity of the initial position; Based on the environmental complexity, determining a sensor usage strategy for the drone to be located; Based on the sensor usage strategy, construct a first positioning solution for the drone to be positioned; When the satellite signal strength is weak, constructing a second positioning solution for the drone to be positioned using the target copy; Based on the first positioning solution and the second positioning solution, a positioning solution for the drone to be positioned is determined.

8. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 1, characterized in that: The method of using the second environmental data in combination with the positioning scheme to continuously correct the position of the unmanned aerial vehicle to be positioned to obtain a revised position includes: When the positioning scheme is the corresponding first positioning scheme, using the second environmental data to create a data update frequency scheme for the drone to be positioned; Based on the data update frequency scheme, performing a first position correction on the drone to be located to obtain a first corrected position; When the positioning scheme is the corresponding second positioning scheme, analyzing the position change state of the unmanned aerial vehicle to be positioned by using the second environmental data; Performing a second position correction on the drone to be located based on the position change state to obtain a second corrected position; Based on the first corrected position and the second corrected position, the position of the unmanned aerial vehicle to be located is continuously corrected to obtain a revised position.

9. The method for positioning an unmanned aerial vehicle based on satellite remote sensing technology as claimed in claim 8, characterized in that: The using the second environmental data to analyze the position change state of the unmanned aerial vehicle to be located includes: Extracting the wind speed and air density of the drone to be located in the second environmental data; Query the flight speed of the UAV to be located; Calculating the real-time aerodynamic force of the UAV to be located based on the wind speed, the flight speed and the air density; Based on the real-time aerodynamic force, calculating the three-axis speed of the UAV to be positioned; Based on the three-axis speed, calculating the moment position of the drone to be located; Based on the momentary position, the position change state of the drone to be located is analyzed.

10. An unmanned aerial vehicle positioning system based on satellite remote sensing technology, characterized in that: The system is used to execute a UAV positioning method based on satellite remote sensing technology as described in any one of claims 1 to 9, and comprises: A virtual copy construction module is used to identify a pre-flight area of ​​the UAV to be located, collect ground object data of the pre-flight area, use the ground object data to construct a virtual copy of the pre-flight area, identify key data of the ground object data, and use the key data to annotate key attributes of the virtual copy to obtain an annotated virtual copy; A feature map processing module is used to identify the drone feature map of the drone to be located, perform multi-channel processing on the drone feature map to obtain a first processed feature map, and perform depth and shallow feature processing on the first processed feature map to obtain a second processed feature map; A positioning solution construction module is used to establish a communication connection with the unmanned aerial vehicle to be positioned through a preconfigured satellite remote sensing device, perform preliminary positioning of the unmanned aerial vehicle to be positioned, and obtain an initial position of the unmanned aerial vehicle to be positioned; collect first environmental data of the unmanned aerial vehicle to be positioned at the initial position, use the first environmental data to perform copy optimization on the annotated virtual copy to obtain a target copy, and use the target copy, the initial position and the second processing feature map to construct a positioning solution for the unmanned aerial vehicle to be positioned; The UAV position analysis module is used to collect the second environmental data of the UAV to be located in real time during its continuous flight, and use the second environmental data in combination with the positioning scheme to continuously correct the position of the UAV to be located to obtain a revised position. Based on the revised position, the real-time position of the UAV to be located after the position revision is analyzed in real time.

Images