Precise landing method and system for unmanned aerial vehicle

Abstract

The invention is suitable for the technical field of landing guidance, and particularly relates to a precise landing method and system for an unmanned aerial vehicle, and the method comprises the steps: delimiting a landing range of the unmanned aerial vehicle, selecting a plurality of deployment positions, starting projection equipment which is installed at the deployment positions in advance, and positioning a target landing point in the landing range, collecting attribute data of the target landing point, wherein the attribute data at least comprises height, position and weather data; and integrating the attribute data, generating a two-dimensional code, sending the two-dimensional code to projection equipment, and adjusting the flight attitude of the unmanned aerial vehicle. The deviation degree between the real-time position of the unmanned aerial vehicle and the planned route can be determined by calculating the deviation value, and the attitude of the unmanned aerial vehicle can be adjusted according to the deviation value by constructing the double linked list, so that the deviation is further reduced while the automatic control capability of the unmanned aerial vehicle is improved, and the accuracy of the unmanned aerial vehicle is improved. And the precision of the unmanned aerial vehicle in the landing process is improved.

Description

Technical Field

[0001] The present invention relates to the field of landing guidance technology, and in particular to a method and system for precise landing of an unmanned aerial vehicle. Background Art

[0002] Drone landing usually relies on the collaboration of multiple sensors such as GPS, visual recognition and lidar; among them, GPS provides global location information, visual recognition technology uses cameras to detect landmarks, markers or landing platforms, and lidar is used for distance measurement and precise height control.

[0003] However, during the actual landing process, signal blocking and weather influences may cause deviations in data collection, making it impossible for the sensor to effectively scan and identify the environmental and terrain features of the landing area, resulting in a deviation in the landing position; therefore, "how to transmit the environmental and terrain feature data of the landing area to the drone through a QR code" is a technical problem that the present invention needs to solve. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for accurate landing of a drone, so as to solve the problem of "how to transmit the environmental and terrain feature data of the landing area to the drone through a QR code" raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for accurate landing of a drone, comprising:

[0007] Delineating the landing range of the drone and selecting a number of deployment locations, starting the projection device pre-installed at the deployment location, locating the target landing point within the landing range, and collecting attribute data of the target landing point, wherein the attribute data at least includes: altitude, location and weather data;

[0008] Integrate the attribute data, generate a QR code, and send the QR code to a projection device to adjust the flight attitude of the drone;

[0009] Collect the real-time location data of the drone, create a double linked list, mark the left node and the right node, and build a synchronous update mechanism, upload the real-time location data to the left node and the right node in parallel, embed the target landing point into the middle of the double linked list, generate a planned route, select a number of monitoring points according to a preset step length, create a calibration node corresponding to the monitoring point, and insert the calibration node between the left node and the target landing point;

[0010] Insert a correction node corresponding to the calibration node one by one between the right node and the target landing point, compare the deviation value of the UAV at the monitoring point, build a machine learning model, input the deviation value into the machine learning model, output the attitude adjustment value, and upload the attitude adjustment value to the corresponding correction node. Based on the real-time position data, dynamically update the double linked list and send the double linked list to the UAV.

[0011] Furthermore, the method further comprises:

[0012] When the UAV passes through the monitoring point, the corresponding correction node is found through the calibration node and defined as the target node;

[0013] The attitude adjustment value in the target node is extracted, and the flight attitude of the UAV is adjusted according to the attitude adjustment value.

[0014] Furthermore, the steps of defining the landing range of the drone, selecting a plurality of deployment positions, starting the projection device pre-installed at the deployment positions, and locating the target landing point within the landing range include:

[0015] Collecting factors affecting the landing of drones, wherein the factors at least include: wind speed, congestion and accidents;

[0016] Based on the influencing factors, several backup landing points are selected from the landing range, switching rules are generated, and the target landing point is dynamically adjusted.

[0017] Furthermore, the step of integrating the attribute data, generating a QR code, and sending the QR code to a projection device to adjust the flight attitude of the drone includes:

[0018] Divide the scanning range of the QR code, record the real-time location data when the drone scans the QR code, and obtain the scanning location;

[0019] The scanning position and the target landing point are input into a machine learning model, the target value is output, and the flight attitude of the UAV is adjusted.

[0020] Furthermore, the step of generating a planned route, selecting a number of monitoring points according to a preset step length, and creating calibration nodes corresponding to the monitoring points one by one includes:

[0021] According to a preset numbering rule, the calibration nodes are numbered, and labels generated by the numbers are inserted into the correction nodes;

[0022] Based on the bidirectional linked list, a posture optimized version is generated and sent to a preset terminal for storage.

[0023] Furthermore, the step of constructing a machine learning model, inputting the deviation value into the machine learning model, and outputting a posture adjustment value comprises:

[0024] In the landing range, an edge node is selected, the machine learning model is migrated to the edge node, and a communication link between the drone and the edge node is established;

[0025] A trigger mechanism is constructed and integrated into the edge node, wherein the trigger mechanism is: when the drone is started, a trigger signal is generated, and the trigger signal is sent to the edge node to start the edge node.

[0026] Furthermore, the method further comprises:

[0027] Taking the target landing point as the center and the preset distance as the radius, an electronic fence is constructed. When the drone enters the electronic fence, the attribute data is used to update the QR code;

[0028] Add a color mark to the QR code.

[0029] Furthermore, a demarcation module is used to demarcate the landing range of the drone, select a number of deployment positions, start a projection device pre-installed at the deployment positions, locate a target landing point within the landing range, and collect attribute data of the target landing point, wherein the attribute data at least includes: altitude, position and weather data;

[0030] An adjustment module, used to integrate the attribute data, generate a two-dimensional code, and send the two-dimensional code to a projection device to adjust the flight attitude of the drone;

[0031] The insertion module is used to collect the real-time location data of the drone, create a double linked list, mark the left node and the right node, and build a synchronous update mechanism to upload the real-time location data to the left node and the right node in parallel, embed the target landing point into the middle of the double linked list, generate a planned route, select a number of monitoring points according to a preset step length, create a calibration node corresponding to the monitoring point, and insert the calibration node between the left node and the target landing point;

[0032] The sending module is used to insert a correction node corresponding to the calibration node one by one between the right node and the target landing point, compare the deviation value of the UAV at the monitoring point, build a machine learning model, input the deviation value into the machine learning model, output the attitude adjustment value, and upload the attitude adjustment value to the corresponding correction node, dynamically update the double linked list based on the real-time position data, and send the double linked list to the UAV.

[0033] Furthermore, the delineation module includes:

[0034] A collection unit, used to collect factors affecting the landing of the drone, wherein the factors at least include: wind speed, congestion and accidents;

[0035] A selection unit is used to select a number of backup landing points from the landing range based on the influencing factors, generate switching rules, and dynamically adjust the target landing point.

[0036] Furthermore, the adjustment module includes:

[0037] A division unit is used to divide the scanning range of the QR code, record the real-time position data when the drone scans the QR code, and obtain the scanning position;

[0038] An output unit is used to input the scanning position and the target landing point into a machine learning model, output a target value, and adjust the flight attitude of the UAV.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] By defining the landing range, the flight path can be planned in advance, and by deploying projection equipment within the landing range, the geographical location and real-time weather information of the landing site can be provided to the drone, ensuring that the drone can obtain the latest environmental data, reduce data errors, and greatly improve the accuracy of the drone's landing. By calculating the deviation value, the degree of deviation between the drone's real-time position and the planned route can be determined. By constructing a double linked list, the drone's flight attitude can be adjusted according to the deviation value, thereby further reducing deviations and improving the accuracy of the drone's landing while improving the drone's automated control capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a double linked list structure provided by an embodiment of the present invention;

[0042] Figure 2 A flowchart of a method for accurate landing of a drone provided by an embodiment of the present invention;

[0043] Figure 3 A first sub-flow chart of the method for accurate landing of a drone provided by an embodiment of the present invention;

[0044] Figure 4 A second sub-flow chart of the method for accurate landing of a drone provided in an embodiment of the present invention;

[0045] Figure 5 A third sub-flow chart of the method for accurate landing of a drone provided by an embodiment of the present invention;

[0046] Figure 6 A fourth sub-flow chart of the method for accurate landing of a drone provided in an embodiment of the present invention;

[0047] Figure 7 A block diagram of the components of the UAV precision landing system provided by an embodiment of the present invention;

[0048] Figure 8 A block diagram of the composition of the demarcation module in the UAV precision landing system provided by an embodiment of the present invention;

[0049] Fig. 9 A block diagram of the composition of the demarcation module in the UAV precision landing system provided by an embodiment of the present invention;

[0050] Fig.10 A block diagram of the components of the insertion module in the UAV precision landing system provided by an embodiment of the present invention;

[0051] Fig.11 This is a block diagram of the components of the sending module in the UAV precision landing system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] In Example 1, Figure 1 and Figure 2 The implementation process of the method for accurate landing of a drone provided by an embodiment of the present invention is shown and described in detail below:

[0054] S100: Delineate the landing range of the drone, select a number of deployment locations, start a projection device pre-installed at the deployment locations, locate a target landing point within the landing range, and collect attribute data of the target landing point, wherein the attribute data includes at least: altitude, location and weather data.

[0055] With reference to the starting point, target landing point and mission area of ​​the UAV, the landing range of the UAV is delineated; for example, with the target landing point as the center and a radius of 5 meters, the landing range is constructed; the advantage of constructing the landing range is that it is easier for the UAV to prepare for landing in advance; within the landing range, multiple deployment positions are selected, among which the deployment positions can be determined by the UAV pilot; projection equipment is installed at the deployment position, and the projection equipment is mainly used to project QR codes onto the ground or other display platforms.

[0056] According to the mission requirements, the pilot selects the target landing point within the landing range and determines the attribute data of the target landing point from public data or by using sensor equipment. The attribute data includes altitude, position and weather data. The altitude data helps to determine the relative position of the target landing point between the drone and the ground. The position data is used to provide the drone with the precise coordinates of the target landing point. The weather data includes information such as wind speed, temperature, humidity and air pressure.

[0057] S200: Integrate the attribute data, generate a two-dimensional code, and send the two-dimensional code to a projection device to adjust the flight posture of the drone.

[0058] All attribute data are packaged and the corresponding QR code is generated, which is sent to the projection device. The UAV flight control system adjusts the flight attitude according to weather information, flight altitude and current position deviation. The flight attitude includes pitch angle, roll angle and heading angle, etc. By adjusting the flight attitude, the flight path can be ensured to be accurately aligned with the target landing point.

[0059] S300: Collect the real-time location data of the drone, create a double linked list, mark the left node and the right node, and build a synchronous update mechanism, upload the real-time location data to the left node and the right node in parallel, embed the target landing point into the middle of the double linked list, generate a planned route, select several monitoring points according to the preset step size, create calibration nodes corresponding to the monitoring points one by one, and insert the calibration nodes between the left node and the target landing point.

[0060] Use the GPS or other positioning sensors carried by the drone to collect the real-time location data of the drone; create a double linked list, where the double linked list is not a two-way linked list in the prior art. The double linked list is a symmetrical structure centered on the middle, which is only used to show the data processing process and does not have processing capabilities itself; the leftmost end of the double linked list is the left node, and the rightmost end is the right node (as shown in the attached manual Figure 1 As shown in the figure, the left node and the right node are only used to represent the real-time location data of the drone and have no other meanings; a synchronous update mechanism is embedded in both the left node and the right node, where the synchronous update mechanism is: when the data in the left node changes, the right node will also change synchronously, and vice versa.

[0061] The real-time location data is transferred to the left node and the right node, and the target landing point is embedded in the middle of the double linked list; the starting point, the mission area and the target landing point are integrated to generate a planned route, and monitoring points are selected from the planned route according to the preset step length; for example, the preset step length is set to 1 meter, and monitoring points are selected on the planned route at intervals of 1 meter; the step length is determined by the drone pilot according to the length of the planned route and the complexity of the route; a corresponding calibration node is set for each monitoring point, where the calibration node is mainly used to characterize the monitoring point, not a physical data processing device; the calibration node is inserted between the left node and the target landing point, where the closer the monitoring point is to the target landing point, the closer the corresponding calibration node is to the target landing point.

[0062] S400: Insert a correction node corresponding to the calibration node one by one between the right node and the target landing point, compare the deviation value of the UAV at the monitoring point, build a machine learning model, input the deviation value into the machine learning model, output the attitude adjustment value, and upload the attitude adjustment value to the corresponding correction node, dynamically update the double linked list based on the real-time position data, and send the double linked list to the UAV.

[0063] Create a correction node corresponding to the calibration node one by one, and insert the correction node between the right node and the target landing point; compare the deviation value of the drone at the monitoring point, and the specific comparison process is as follows: map the planned route, monitoring point and real-time position of the drone to the coordinate system, read the coordinates of all monitoring points and real-time position data, traverse the monitoring point closest to the real-time position, calculate the distance between the two, and determine this distance as the deviation value; use the neural network and reinforcement learning algorithm in the existing technology to build a machine learning model, and use the supervised learning algorithm to train the machine learning model to learn how to adjust the attitude under different deviation conditions; input the deviation value into the machine learning model, and output the attitude adjustment value, wherein the attitude adjustment value is displayed in the form of attitude control instructions to guide the drone to adjust the pitch angle, roll angle or yaw angle, etc.; upload the attitude adjustment value to the correction node corresponding to the nearest monitoring point traversed above, and dynamically update the correction node and the double linked list according to the real-time position data of the drone, and send the updated double linked list to the drone, and the drone adjusts the flight attitude in turn according to the attitude adjustment value in the correction node in the double linked list.

[0064] In Example 2, Figure 3 The implementation process of the method for accurate landing of a drone provided by an embodiment of the present invention is shown. The steps of defining the landing range of the drone, selecting a plurality of deployment positions, starting the projection device pre-installed at the deployment positions, and locating the target landing point within the landing range are described in detail as follows:

[0065] S101: Collect influencing factors of drone landing, where the influencing factors at least include: wind speed, congestion and accidents.

[0066] Among the factors that affect the precise landing of drones, wind speed is one of the key factors. The wind speed in this embodiment includes wind direction, wind speed change, turbulence intensity, strong wind and sudden change in wind speed, etc.; it should be noted that congestion in the influencing factors refers to the dynamic environment of the landing area, such as whether there are other drones or aircraft operating nearby, whether there are moving vehicles or personnel on the ground, etc.; accidents refer to whether emergencies occur in the landing area, such as damage to ground facilities, sudden appearance of obstacles, and airspace adjustments caused by emergency tasks.

[0067] S102: Based on the influencing factors, a number of backup landing points are selected from the landing range, a switching rule is generated, and the target landing point is dynamically adjusted.

[0068] If the above-mentioned influencing factors exist, several alternative landing points can be selected from the landing range, and the target landing point can be selected from the alternative landing points according to the switching rules; the switching rules are: determine the priority of each alternative landing point, and define the alternative landing point with the highest priority as the target landing point. The priority is determined by the flight operator.

[0069] In Example 3, Figure 4 The implementation process of the method for accurate landing of a drone provided by an embodiment of the present invention is shown. The steps of integrating the attribute data, generating a QR code, sending the QR code to a projection device, and adjusting the flight attitude of the drone are described in detail as follows:

[0070] S201: Divide the scanning range of the QR code, record the real-time position data when the drone scans the QR code, and obtain the scanning position.

[0071] The area that can scan the QR code, i.e. the scanning range, is determined based on the size of the target landing point, the field of view of the drone camera, the flight altitude, and the size of the QR code. When the drone enters the scanning range, the flight attitude of the drone is adjusted so that it can accurately identify the QR code. The real-time position of the drone when scanning the QR code is recorded and defined as the scanning position.

[0072] S202: Input the scanning position and the target landing point into a machine learning model, output a target value, and adjust the flight attitude of the UAV.

[0073] The scanning position and target landing point are input into the machine learning model, and the target value is output. When the drone completes scanning the QR code, the identified attribute data is used to build a high-precision landing area model, correct the target landing point, and adjust the flight attitude to the target value.

[0074] In Example 4, Figure 5 The implementation process of the method for accurate landing of a drone provided by an embodiment of the present invention is shown. The steps of generating a planned route, selecting a number of monitoring points according to a preset step length, and creating calibration nodes corresponding to the monitoring points are described in detail as follows:

[0075] S301: numbering the calibration nodes according to a preset numbering rule, and inserting labels generated by the numbering into the correction nodes.

[0076] The calibration nodes are numbered according to a preset numbering rule, wherein the numbering rule may be a numerical sequence numbering; and a corresponding relationship between the calibration nodes and the correction nodes is established by using the numbering.

[0077] S302: Generate a posture optimized version based on the bidirectional linked list, and send it to a preset terminal for storage.

[0078] The data in the bidirectional linked list is recorded, and the data is packaged and encrypted to obtain the attitude optimized version. In other words, the attitude adjustment value of the UAV during flight is stored according to the version method; the advantage of this is that after the flight mission is completed, this attitude optimized version can be used to train the machine learning model; the attitude optimized version is sent to the preset terminal for storage, where the preset terminal can be a ground control center, a cloud storage system or an edge computing device, etc.

[0079] In Example 5, Figure 6 The implementation process of the method for accurate landing of a drone provided by an embodiment of the present invention is shown. The steps of constructing a machine learning model, inputting the deviation value into the machine learning model, and outputting the attitude adjustment value are described in detail as follows:

[0080] S401: In the landing range, select an edge node, migrate the machine learning model to the edge node, and establish a communication link between the drone and the edge node.

[0081] Within the landing range, edge nodes are selected, where edge nodes are edge servers, gateway devices or high-performance terminals with strong computing capabilities; the machine learning model is migrated to the edge node to reduce the cloud computing burden of the machine learning model, and a communication link is established between the drone and the edge node so that the data output by the machine learning model can be sent to the drone in a timely manner.

[0082] S402: Construct a trigger mechanism and integrate it into the edge node, wherein the trigger mechanism is: when the drone is started, a trigger signal is generated, and the trigger signal is sent to the edge node to start the edge node.

[0083] A trigger mechanism is embedded in the edge node, and the edge node will only start when the drone is started. The advantage of this is that it can reduce the energy consumption of the edge node and ensure that the drone can be allocated sufficient computing resources; however, in real life, if the edge device needs to process other computing tasks, it can also keep working when the drone is not started.

[0084] In Example 6, different from Example 1, in this embodiment of the present invention, the method further includes:

[0085] Taking the target landing point as the center and the preset distance as the radius, an electronic fence is constructed. When the drone enters the electronic fence, the attribute data is used to update the QR code;

[0086] Add a color mark to the QR code.

[0087] An electronic fence is constructed with the target landing point as the center and the preset distance as the radius, where the preset distance can be pre-determined by the drone operator; the influencing factors of the target landing point may be in real-time change, and the QR code needs to be dynamically updated. However, updating the QR code too early is likely to cause information delays and affect the decision-making efficiency of the drone. Updating the QR code too late will not leave enough time for the collection and analysis of influencing factors. Therefore, by setting up an electronic fence, when the drone enters the electronic fence, it starts to collect real-time data of the influencing factors, and updates the QR code in combination with the attribute data.

[0088] In addition, color tags can be added to the QR code to improve the drone's recognition accuracy of the QR code; for example, by setting different foreground and background colors, the recognition efficiency of the QR code can be improved.

[0089] In Example 7, different from Example 1, in this embodiment of the present invention, the method further includes:

[0090] When the UAV passes through the monitoring point, the corresponding correction node is found through the calibration node and defined as the target node;

[0091] The attitude adjustment value in the target node is extracted, and the flight attitude of the UAV is adjusted according to the attitude adjustment value.

[0092] Calculate the monitoring point closest to the UAV. Since there is a corresponding relationship between the monitoring point and the calibration node, and there is a corresponding relationship between the calibration node and the correction node, the target node is the correction node corresponding to the monitoring point closest to the UAV; use the attitude adjustment value in the correction node to adjust the flight attitude of the UAV.

[0093] Figure 7 The structure diagram of the UAV precision landing system provided by the embodiment of the present invention is shown. The UAV precision landing system 1 includes:

[0094] The demarcation module 11 is used to demarcate the landing range of the UAV, select a number of deployment positions, start the projection device pre-installed at the deployment position, locate the target landing point within the landing range, and collect attribute data of the target landing point, wherein the attribute data at least includes: altitude, position and weather data;

[0095] An adjustment module 12, for integrating the attribute data, generating a two-dimensional code, and sending the two-dimensional code to a projection device to adjust the flight attitude of the drone;

[0096] Insertion module 13, used to collect the real-time location data of the drone, create a bidirectional linked list, mark the left node and the right node, and build a synchronous update mechanism, upload the real-time location data to the left node and the right node in parallel, embed the target landing point into the middle of the bidirectional linked list, generate a planned route, select a number of monitoring points according to a preset step length, create a calibration node corresponding to the monitoring point, and insert the calibration node between the left node and the target landing point;

[0097] The sending module 14 is used to insert a correction node corresponding to the calibration node one by one between the right node and the target landing point, compare the deviation value of the drone at the monitoring point, build a machine learning model, input the deviation value into the machine learning model, output the attitude adjustment value, and upload the attitude adjustment value to the corresponding correction node, dynamically update the bidirectional linked list based on the real-time position data, and send the bidirectional linked list to the drone.

[0098] Figure 8 The structure diagram of the UAV precision landing system provided by the embodiment of the present invention is shown, and the demarcation module 11 includes:

[0099] A collection unit 111 is used to collect factors affecting the landing of the drone, wherein the factors at least include: wind speed, congestion and accidents;

[0100] The selection unit 112 is used to select a number of backup landing points from the landing range based on the influencing factors, generate switching rules, and dynamically adjust the target landing point.

[0101] Fig. 9 The structure diagram of the UAV precision landing system provided by the embodiment of the present invention is shown, and the adjustment module 12 includes:

[0102] The division unit 121 is used to divide the scanning range of the two-dimensional code, record the real-time position data when the drone scans the two-dimensional code, and obtain the scanning position;

[0103] The output unit 122 is used to input the scanning position and the target landing point into the machine learning model, output the target value, and adjust the flight attitude of the UAV.

[0104] Fig.10 The structure diagram of the UAV precision landing system provided by the embodiment of the present invention is shown, and the insertion module 13 includes:

[0105] The correction unit 131 is used to number the calibration nodes according to a preset numbering rule, and insert a label generated by the numbering into the correction node;

[0106] The sending unit 132 is used to generate a posture optimized version according to the bidirectional linked list, and send it to a preset terminal for storage.

[0107] Fig.11 The structure diagram of the UAV precision landing system provided by the embodiment of the present invention is shown, and the sending module 14 includes:

[0108] A building unit 141 is used to select an edge node in the landing range, migrate the machine learning model to the edge node, and build a communication link between the drone and the edge node;

[0109] The start-up unit 142 is used to construct a trigger mechanism and integrate it into the edge node, wherein the trigger mechanism is: when the drone is started, a trigger signal is generated, and the trigger signal is sent to the edge node to start the edge node.

[0110] The demarcation module 11 is mainly used to complete step S100, the adjustment module 12 is mainly used to complete step S200, the insertion module 13 is mainly used to complete step S300, and the sending module 14 is mainly used to complete step S400;

[0111] The collection unit 111 is mainly used to complete step S101, and the selection unit 112 is mainly used to complete step S102;

[0112] The dividing unit 121 is mainly used to complete step S201, and the output unit 122 is mainly used to complete step S202;

[0113] The correction unit 131 is mainly used to complete step S301, and the sending unit 132 is mainly used to complete step S302;

[0114] The building unit 141 is mainly used to complete step S401, and the starting unit 142 is mainly used to complete step S402.

[0115] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0117] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for accurate landing of a drone, characterized in that: The method comprises: Delineating the landing range of the drone and selecting a number of deployment locations, starting the projection device pre-installed at the deployment location, locating the target landing point within the landing range, and collecting attribute data of the target landing point, wherein the attribute data at least includes: altitude, location and weather data; Integrate the attribute data, generate a QR code, and send the QR code to a projection device to adjust the flight attitude of the drone; Collect the real-time location data of the drone, create a double linked list, mark the left node and the right node, and build a synchronous update mechanism, upload the real-time location data to the left node and the right node in parallel, embed the target landing point into the middle of the double linked list, generate a planned route, select a number of monitoring points according to a preset step length, create a calibration node corresponding to the monitoring point, and insert the calibration node between the left node and the target landing point; Insert a correction node corresponding to the calibration node one by one between the right node and the target landing point, compare the deviation value of the UAV at the monitoring point, build a machine learning model, input the deviation value into the machine learning model, output the attitude adjustment value, and upload the attitude adjustment value to the corresponding correction node. Based on the real-time position data, dynamically update the double linked list and send the double linked list to the UAV.

2. The method for accurate landing of a drone according to claim 1, characterized in that: The method further comprises: When the UAV passes through the monitoring point, the corresponding correction node is found through the calibration node and defined as the target node; The attitude adjustment value in the target node is extracted, and the flight attitude of the UAV is adjusted according to the attitude adjustment value.

3. The method for accurate landing of a drone according to claim 1, characterized in that: The steps of defining the landing range of the drone, selecting a plurality of deployment positions, starting the projection device pre-installed at the deployment positions, and locating the target landing point within the landing range include: Collecting factors affecting the landing of drones, wherein the factors at least include: wind speed, congestion and accidents; Based on the influencing factors, several backup landing points are selected from the landing range, switching rules are generated, and the target landing point is dynamically adjusted.

4. The method for accurate landing of a drone according to claim 3, characterized in that: The steps of integrating the attribute data, generating a two-dimensional code, and sending the two-dimensional code to a projection device to adjust the flight attitude of the drone include: Divide the scanning range of the QR code, record the real-time location data when the drone scans the QR code, and obtain the scanning location; The scanning position and the target landing point are input into a machine learning model, the target value is output, and the flight attitude of the UAV is adjusted.

5. The method for accurate landing of a drone according to claim 2, characterized in that: The step of generating a planned route, selecting a number of monitoring points according to a preset step length, and creating calibration nodes corresponding to the monitoring points one by one includes: According to a preset numbering rule, the calibration nodes are numbered, and labels generated by the numbers are inserted into the correction nodes; Based on the bidirectional linked list, a posture optimized version is generated and sent to a preset terminal for storage.

6. The method for accurate landing of a drone according to claim 3, characterized in that: The step of constructing a machine learning model, inputting the deviation value into the machine learning model, and outputting a posture adjustment value comprises: In the landing range, an edge node is selected, the machine learning model is migrated to the edge node, and a communication link between the drone and the edge node is established; A trigger mechanism is constructed and integrated into the edge node, wherein the trigger mechanism is: when the drone is started, a trigger signal is generated, and the trigger signal is sent to the edge node to start the edge node.

7. The method for accurate landing of a drone according to claim 4, characterized in that: The method further comprises: Taking the target landing point as the center and the preset distance as the radius, an electronic fence is constructed. When the drone enters the electronic fence, the attribute data is used to update the QR code; Add a color mark to the QR code.

8. A UAV precision landing system, characterized in that: The system comprises: A demarcation module is used to demarcate the landing range of the UAV, select a number of deployment positions, start the projection device pre-installed at the deployment position, locate the target landing point within the landing range, and collect attribute data of the target landing point, wherein the attribute data at least includes: altitude, position and weather data; An adjustment module, used to integrate the attribute data, generate a two-dimensional code, and send the two-dimensional code to a projection device to adjust the flight attitude of the drone; The insertion module is used to collect the real-time location data of the drone, create a double linked list, mark the left node and the right node, and build a synchronous update mechanism to upload the real-time location data to the left node and the right node in parallel, embed the target landing point into the middle of the double linked list, generate a planned route, select a number of monitoring points according to a preset step length, create a calibration node corresponding to the monitoring point, and insert the calibration node between the left node and the target landing point; The sending module is used to insert a correction node corresponding to the calibration node one by one between the right node and the target landing point, compare the deviation value of the UAV at the monitoring point, build a machine learning model, input the deviation value into the machine learning model, output the attitude adjustment value, and upload the attitude adjustment value to the corresponding correction node, dynamically update the double linked list based on the real-time position data, and send the double linked list to the UAV.

9. The UAV precision landing system according to claim 8, characterized in that: The delineation module comprises: A collection unit, used to collect factors affecting the landing of the drone, wherein the factors at least include: wind speed, congestion and accidents; A selection unit is used to select a number of backup landing points from the landing range based on the influencing factors, generate switching rules, and dynamically adjust the target landing point.

10. The UAV precision landing system according to claim 9, characterized in that: The adjustment module comprises: A division unit is used to divide the scanning range of the QR code, record the real-time position data when the drone scans the QR code, and obtain the scanning position; An output unit is used to input the scanning position and the target landing point into a machine learning model, output a target value, and adjust the flight attitude of the UAV.

Images