Off-line and on-line combined unmanned aerial vehicle ground simulation flight path planning method

Abstract

The invention discloses an off-line and on-line combined unmanned aerial vehicle simulation flight path planning method, which belongs to the field of unmanned aerial vehicle flight path planning, and specifically comprises the following steps: obtaining a series of insertion waypoint longitudes and latitudes through a waypoint interpolation algorithm according to a selected route before a flight task is executed in an off-line mode; then altitude information stored in an elevation database is obtained in an elevation query mode, the target height of each waypoint of the unmanned aerial vehicle is calculated, and then route planning in the whole task process is completed; in the online mode, in the flight mission execution process, a real-time to-ground laser ranging system is adopted to obtain real-time to-ground height information of the unmanned aerial vehicle, after a measurement error is calibrated by combining flight attitude information of the unmanned aerial vehicle, the measurement error is compared with the actual flight height of the unmanned aerial vehicle, and the to-ground height of the unmanned aerial vehicle is calibrated on the basis of a pre-planned route; and the ground-imitating flight precision of the unmanned aerial vehicle is improved. According to the invention, high-precision ground-imitating equal-altitude flight route planning of the long-endurance unmanned aerial vehicle in a complex environment is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle flight path planning, and in particular relates to an unmanned aerial vehicle terrain-simulating flight path planning method combining offline and online operations. Background Art

[0002] UAV terrain-simulating flight is a flight state in which the UAV maintains a constant altitude above the ground during flight. This mode is a common requirement for UAVs to perform tasks. High-precision terrain-simulating flight has significant advantages in many mission scenarios. For example, it can effectively improve the utilization efficiency of pesticides and fertilizers when performing farmland plant protection operations; it can obtain more accurate terrain and geological information when conducting land surveying, geological exploration and other tasks; it can effectively improve the accuracy and stability of animal tracking and identification when conducting ecological monitoring. Based on the above wide range of application scenarios, the development of route planning technology for UAV terrain-simulating flight can effectively improve the versatility and efficiency of UAVs, which is of great significance to the development of UAVs.

[0003] The key issue in the research and development of UAV terrain-following altitude flight path planning technology is how to obtain accurate and effective ground height data. For this purpose, many altitude data acquisition methods have been developed, using different types of distance sensors such as laser radar, millimeter-wave radar, ultrasonic sensor, and binocular ranging sensor to improve data accuracy. However, existing ranging technology is usually accompanied by certain limitations. For example, ultrasonic sensors are greatly affected by the propagation medium, millimeter-wave radar is easily blocked by obstacles, and binocular vision sensors are greatly limited by the computing power of onboard hardware. Therefore, when designing a UAV terrain-following altitude flight algorithm, it is necessary to address the above problems and propose solutions based on actual conditions to ensure high-precision terrain-following altitude flight effects. Summary of the invention

[0004] In view of the above problems, the present invention proposes an offline and online combined unmanned aerial vehicle terrain-imitating flight path planning method, which realizes high-precision isotropic terrain-imitating flight of the unmanned aerial vehicle.

[0005] The offline and online combined unmanned aerial vehicle terrain imitation flight path planning method of the present invention includes an offline mode and an online mode, specifically:

[0006] Offline mode steps:

[0007] 101: Obtain the straight-line distance Dist between two adjacent waypoints according to the latitude and longitude information of each waypoint on the initial route.

[0008] 102: Determine the number n of waypoints to be inserted by specifying the step length.

[0009] 103: According to the number of inserted waypoints n, the longitude and latitude of each waypoint is calculated in turn according to the specified step length Steplen, and the calculation method adopts geometric spherical linear interpolation calculation to achieve equal interval interpolation.

[0010] 104: Perform elevation query based on the longitude and latitude of each waypoint in turn. The elevation data is queried online based on the open-elevation open source database. The method is:

[0011] First, a query URL is generated according to the specified format based on the latitude and longitude of the target waypoint to be queried, and an elevation query request is sent to the cloud, waiting for the return result. When the return is successful, the file is read, and when the return fails, the query is repeated.

[0012] After the query is successful, the information stored in the JSON format file will be read, and the elevation information will be read and saved to a list according to the information format.

[0013] After the query is completed, the read buffer area is cleared, and the waypoint corresponding to the currently saved elevation information is detected based on the number of queried waypoints and the total number of waypoints to see if it is the last waypoint. If so, the query step is completed, otherwise it returns to the first step to generate the elevation query URL for the next target waypoint.

[0014] 105: After obtaining the elevation values ​​of the original route waypoints and the inserted waypoints through step 104, the target height of each waypoint is obtained according to the ground altitude data and the ground altitude of the simulated contour flight target.

[0015] 106: After the target heights of each waypoint obtained in step 105, an altitude check is performed to ensure the safety and effectiveness of the route.

[0016] Online mode steps:

[0017] 201: Install a laser radar distance sensor on the UAV and establish a data link between the UAV flight control system and the laser ranging system, and transmit the UAV altitude above the ground measured by the sensor to the UAV flight control system in real time.

[0018] 202: Calculate the actual altitude of the drone above the ground based on the aircraft attitude angle and the measured altitude of the drone above the ground to eliminate the deviation.

[0019] 203: Setting a calibration threshold for the real-time ground height Hei of the UAV obtained in step 202 and calibrating the flight height of the UAV.

[0020] The advantages of the present invention are:

[0021] 1. The offline and online combined unmanned aerial vehicle terrain imitation flight path planning method of the present invention provides a complete set of unmanned aerial vehicle terrain imitation flight path planning methods, and realizes the high-precision contour terrain imitation flight of the unmanned aerial vehicle;

[0022] 2. The offline and online combined unmanned aerial vehicle terrain-simulating flight path planning method of the present invention sets an elevation interpolation query algorithm and a waypoint switching algorithm to ensure the accuracy and safety of terrain-simulating flight. The elevation interpolation query adds transition waypoints to the route according to the specified step length to improve the fitting degree of the terrain-simulating route and realize the precision control of terrain-simulating flight; the waypoint switching algorithm verifies the elevation difference between each waypoint in combination with the performance limit of the unmanned aerial vehicle, and provides a targeted waypoint switching method for the flight segments that exceed the climbing or descending capacity limit to ensure flight safety.

[0023] 3. It has obvious advantages in meeting the needs of terrain-imitating flight in complex mountain and forest environments. The offline pre-planning of high-precision terrain-imitating flight routes for drones realized through satellite elevation data query makes the drones independent of real-time distance sensors to a certain extent, avoiding the impact on flight safety and accuracy when the distance sensors are affected by complex environments. At the same time, the use of lidar as a real-time distance sensor can effectively avoid the impact of the forest environment, further improve the accuracy of terrain-imitating flight, and make up for the insufficient accuracy of offline planning methods.

[0024] 4. The offline and online combined UAV terrain-simulating flight path planning method of the present invention has a lower structural weight load and airborne equipment requirements, which can greatly reduce the structural weight and UAV cost, and is of great significance in mission requirements with low cost and long flight time. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The offline planning process of the terrain-imitating flight route of the present invention;

[0026] Figure 2 It is the implementation process of elevation online query of the present invention;

[0027] Figure 3 It is the height switching control mode of the present invention;

[0028] Figure 4 This is the implementation process of the real-time laser ranging ground height calibration of the present invention.

[0029] Figure 5 It is a schematic diagram of the tilt angle error calibration method of the UAV of the present invention. DETAILED DESCRIPTION

[0030] The present invention is further described in detail below with reference to the accompanying drawings.

[0031] The present invention discloses a method for planning a terrain-simulating flight path for a drone that combines offline and online operations, including an offline mode and an online mode. In the offline mode, the altitude information of the altitude database is obtained by adopting an altitude query method, and the route in the entire mission process is pre-planned before the drone mission begins. The target altitude of each waypoint of the drone is calculated by the altitude, so as to realize offline pre-planning of the drone's terrain-simulating flight route; while in the online mode, the real-time ground laser ranging system is used to obtain the real-time ground altitude information of the drone, and the drone's current flight altitude information is combined to calibrate the ground altitude of the drone based on the pre-planned route, so as to improve the accuracy of the drone's terrain-simulating flight.

[0032] The specific methods of the above two modes are as follows:

[0033] The offline planning mode of the UAV terrain-simulating flight route first selects waypoints. According to the target mission requirements, a series of waypoints are selected as the initial route, and the waypoints are connected by straight lines. Then, elevation interpolation query is performed based on the selected waypoints. The purpose of this step is to generate a route that fits the fluctuations of the ground altitude by performing waypoint elevation query according to the specified step length to achieve terrain-simulating flight. Figure 1 As shown, the specific steps are as follows:

[0034] 101: According to the latitude and longitude information of each waypoint on the initial route, the straight-line distance Dist between two adjacent waypoints is obtained in turn. The distance calculation can be performed using the Haversine formula. When the earth is approximated as a sphere, the formula can calculate the shortest distance between the two waypoints on the sphere through the longitude and latitude of the two waypoints. The specific calculation method is:

[0035]

[0036] Among them, α is the angle formed by two adjacent waypoints and the center of the sphere; Δφ=φ2-φ1 is the latitude difference between two adjacent waypoints (unit: radians); Δλ=λ2-λ1 is the longitude difference between two adjacent waypoints (unit: radians), and the coefficient R is the radius of the earth, usually 6371km.

[0037] 102: After obtaining the straight-line distance between two adjacent waypoints, the number of waypoints n to be inserted can be determined by specifying the step length:

[0038]

[0039] Among them, n is the number of inserted waypoints, rounded down; Steplen is the specified step value. It can be seen that the calculation of the number of inserted waypoints is limited by the Steplen value. This parameter can effectively adjust the accuracy of terrain-simulating flight. As the step length shortens, the number of inserted waypoints increases, and the degree of fitting between the route and the ground elevation fluctuation continues to increase. However, at the same time, the complexity of the route offline pre-planning calculation will be extended, which will increase the route generation time. In order to avoid too long planning calculations while ensuring a certain accuracy, the specified step length of each segment needs to be adjusted accordingly according to the actual terrain conditions of the segment. For segments with gently undulating terrain, a larger step length is used to save planning time, and for segments with large terrain fluctuations, a smaller step length is used to ensure the accuracy of terrain-simulating flight.

[0040] 103: According to the number of pre-inserted waypoints, the longitude and latitude of each waypoint are calculated in turn according to the specified step length Steplen. In order to achieve equal-interval interpolation, geometric spherical linear interpolation calculation (Geometric Slerp) is required. First, the longitude and latitude are converted to a Cartesian coordinate system. Based on the above assumption that the earth is a perfect sphere, the conversion formula can be obtained as follows:

[0041] x=cosφcosλ

[0042] y=sinφsinλ

[0043] z=sinφ

[0044] At this time, the coordinates of two adjacent waypoints are:

[0045] P1=(cosφ1cosλ1,sinφ1sinλ1,sinφ1)

[0046] P2=(cosφ2cosλ2,sinφ2sinλ2,sinφ2)

[0047] Calculated using spherical linear interpolation method:

[0048] cosθ=P1·P2θ=cos -1 (cosθ)

[0049] Among them, θ is the angle between two adjacent waypoints. If we want to insert n points between the two points, the ratio of each interpolation point is:

[0050]

[0051] Then the Cartesian coordinates of each interpolation point are:

[0052]

[0053] Then, the Cartesian coordinates are converted into longitude and latitude coordinates according to the above conversion formula to obtain the longitude and latitude coordinates of each inserted waypoint.

[0054] 104: Perform elevation query according to the longitude and latitude of each waypoint in turn. Elevation data query needs to rely on a digital elevation database. Usually, this type of map data is stored and published in the form of a digital elevation model. The model represents a physical ground model of ground elevation in the form of a set of ordered numerical arrays. The ground elevation information is stored in grid units, and the terrain features are stored in the format of a discretized rectangular grid array by cutting the map. The direct elevation query and call of the model requires offline storage of huge map information. The online elevation query method based on the open source geographic information database can effectively solve this problem. Through the prescribed URL format combined with the longitude and latitude information, the cloud database can be accessed through the network to obtain the elevation information of the corresponding point. Based on the above method, the present invention proposes an online elevation query process based on the open-elevation open source database:

[0055] like Figure 2 As shown, firstly, a query URL is generated according to the prescribed format based on the latitude and longitude of the target waypoint to be queried, and an elevation query request is sent to the cloud, waiting for the return result. When the return is successful, the file is read, and when the return fails, the query is repeated;

[0056] After the query is successful, the information stored in the JSON format file will be read, and the elevation information will be read and saved to the list according to the information format;

[0057] After the query is completed, the read buffer area is cleared, and the waypoint corresponding to the currently saved elevation information is detected based on the number of queried waypoints and the total number of waypoints to see if it is the last waypoint. If so, the query step is completed, otherwise it returns to the first step to generate the elevation query URL for the next target waypoint.

[0058] 105: After obtaining the elevation values ​​of the original route waypoints and the inserted waypoints through step 104, the target height of each waypoint can be obtained according to the ground altitude data and the ground altitude of the simulated contour flight target:

[0059] THei=GHei+Error

[0060] Among them, THei is the target height, GHei is the ground altitude, and Error is the deviation between the target height and the ground altitude, that is, the height of the target relative to the ground during the terrain-simulating flight, which is manually set according to the aircraft's flight altitude requirements.

[0061] 106: According to the target height of each waypoint obtained in step 105, an altitude check is required to ensure the safety and effectiveness of the route. The altitude check formula is:

[0062] tan -1 θ=(THei i+1 -THei i) / Dist

[0063] Among them, i is the waypoint target ground height, THe i+1 is the target height of the next waypoint, θ is the target pitch angle of the drone. When the target pitch angle of the drone in a certain segment is greater than the maximum pitch angle of the drone, the drone cannot complete the normal straight climb. At this time, the following setting is required in this segment: Figure 3 The control method of the drone circling and then flying forward / flying forward and then circling and rising shown ensures that the drone can safely reach the next target waypoint. For the flight sections where the drone's altitude climbing ability can meet the requirements, the normal control method of slope control is adopted to ensure the terrain-imitating flight effect.

[0064] After completing the above steps, a complete, safe and effective offline planning of the terrain-imitating flight route is realized. By uploading the route to the drone, the drone's terrain-imitating and contour flight can be realized.

[0065] The online planning mode of the UAV terrain-following flight route obtains accurate ground information in real time by installing a ground-following laser radar sensor, and performs corresponding height fine-tuning on the basis of offline route planning according to the flight accuracy requirements to ensure the terrain-following flight accuracy requirements, such as Figure 4 As shown, the specific implementation steps are:

[0066] 201: Install a laser radar distance sensor on the UAV and establish a data link between the UAV flight control system and the laser ranging system, and transmit the UAV altitude above the ground measured by the sensor to the UAV flight control system in real time.

[0067] 202: As Figure 5 As shown in the figure, in the actual mission process, the UAV flight attitude angle will cause a deviation between the measured altitude above the ground and the actual altitude above the ground. The actual altitude above the ground needs to be calculated based on the aircraft attitude angle and the measured altitude of the UAV above the ground to eliminate the deviation. The specific formula is as follows:

[0068] Hei=Hei0·cosθ·cosλ

[0069] Among them, Hei is the actual height of the UAV above the ground, Hei0 is the height of the UAV above the ground measured by the sensor, θ is the pitch angle of the UAV, and λ is the roll angle of the UAV.

[0070] 203: The real-time ground height Hei of the UAV obtained in step 202 is calibrated by setting a calibration threshold and calibrating the flight height of the UAV. The specific formula is as follows:

[0071] Err=THei·η

[0072] ΔH=|THei-Hei|

[0073] Among them, Err is the calibration threshold, ΔH is the difference between the target height and the actual height, η is the error range (unit: %), and THei is the target height above the ground. According to the above formula, when ΔH is greater than the calibration threshold Err, the height of the drone is fine-tuned accordingly, with an amplitude of |ΔH-Err|.

[0074] According to the above steps, the terrain-simulating flight altitude calibration method using real-time lidar ranging can be implemented.

[0075] In summary, the offline and online combined unmanned aerial vehicle terrain-imitating flight path planning method of the present invention can meet the needs of unmanned aerial vehicle terrain-imitating and isotactic flight in complex environments. The offline pre-planning method of elevation interpolation query combined with elevation verification can realize safe and effective path planning before the mission. The real-time ground height information of the ground laser radar can realize the accuracy calibration of terrain-imitating flight in the mission. The planning algorithm combining the two methods can make the unmanned aerial vehicle terrain-imitating flight in complex environments more accurate, reliable and safe.

[0076] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for planning an unmanned aerial vehicle's terrain-simulating flight path by combining offline and online methods, characterized in that: Including offline mode and online mode; Among them, the specific steps of offline mode are: 101: Obtain the straight-line distance Dist between two adjacent waypoints according to the latitude and longitude information of each waypoint on the initial route; 102: Determine the number n of waypoints to be inserted by specifying the step length; 103: According to the number of inserted waypoints n, the longitude and latitude of each waypoint is calculated in turn according to the specified step length Steplen, and the calculation method adopts geometric spherical linear interpolation calculation to achieve equal interval interpolation; 104: Perform elevation query based on the longitude and latitude of each waypoint in turn. The elevation data is queried online based on the open-elevation open source database. The method is: First, generate a query URL in the specified format according to the latitude and longitude of the target waypoint to be queried, and send an elevation query request to the cloud, waiting for the return result. If the return is successful, read the file, and if the return fails, repeat the query; After the query is successful, the information stored in the JSON format file will be read, and the elevation information will be read and saved to the list according to the information format; After the query is completed, the read buffer area is cleared, and the waypoint corresponding to the currently saved elevation information is detected based on the number of queried waypoints and the total number of waypoints to see if it is the last waypoint. If so, the query step is completed, otherwise it returns to the first step to generate the elevation query URL for the next target waypoint. 105: After obtaining the elevation values ​​of the original route waypoints and the inserted waypoints through step 104, the target height of each waypoint is obtained according to the ground altitude data and the ground altitude of the simulated contour flight target; 106: After the target height of each waypoint is obtained in step 105, the safety and effectiveness of the route are ensured by performing an altitude check; The specific steps of online mode are: 201: Install a laser radar distance sensor on the UAV, and establish a data link between the UAV flight control system and the laser ranging system, and transmit the UAV altitude above the ground measured by the sensor to the UAV flight control system in real time; 202: Calculate the actual height of the drone above the ground according to the aircraft attitude angle and the measured height of the drone above the ground to eliminate the deviation; 203: Setting a calibration threshold for the real-time ground height Hei of the UAV obtained in step 202 and calibrating the flight height of the UAV.

2. The method for planning an unmanned aerial vehicle terrain-simulating flight path by combining offline and online operations as claimed in claim 1, characterized in that: In step 101, the straight-line distance between two adjacent waypoints is calculated using the Haversine formula, and the specific calculation method is: Among them, α is the angle formed by two adjacent waypoints and the center of the sphere; Δφ=φ2-φ1 is the latitude difference between two adjacent waypoints; Δλ=λ2-λ1 is the longitude difference between two adjacent waypoints, and the coefficient R is the radius of the earth, usually 6371km.

3. The method for planning an unmanned aerial vehicle terrain-simulating flight path by combining offline and online methods as claimed in claim 1, characterized in that: In step 102, the step length is adjusted accordingly according to the actual terrain conditions of the flight segment. For a flight segment with gently undulating terrain, a large step length is used to save planning time, and for a flight segment with greatly undulating terrain, a small step length is used to ensure the accuracy of the terrain simulation flight.

4. The method for planning an unmanned aerial vehicle terrain-simulating flight path by combining offline and online methods as claimed in claim 1, characterized in that: In step 103, the latitude and longitude of each waypoint are calculated as follows: First, convert the longitude and latitude into the Cartesian coordinate system, and the conversion formula is as follows: x=cosφcosλ y=sinφsinλ z=sinφ At this time, the coordinates of two adjacent waypoints are: P1=(cosφ1cosλ1,sinφ1sinλ1,sinφ1) P2=(cosφ2cosλ2,sinφ2sinλ2,sinφ2) Calculated using spherical linear interpolation method: cosθ=P1·P2 θ=cos -1 (cosθ) Among them, θ is the angle between two adjacent waypoints. If we want to insert n points between the two points, the ratio of each interpolation point is: Then the Cartesian coordinates of each interpolation point are: Then, the Cartesian coordinates are converted into longitude and latitude coordinates according to the above conversion formula to obtain the longitude and latitude coordinates of each inserted waypoint.

5. The method for planning an unmanned aerial vehicle terrain-simulating flight path by combining offline and online methods as claimed in claim 1, characterized in that: In step 106, the elevation verification method is: The elevation calibration formula is: tan -1 θ=(THei i+1 -THei i ) / Dist Where Dist is the straight-line distance between two adjacent waypoints; i is the waypoint target ground height, THei i+1 is the target height of the next waypoint, and θ is the target climb pitch angle of the UAV.

6. A method for planning an unmanned aerial vehicle terrain-simulating flight path by combining offline and online operations as described in claim 5, characterized in that: When the target pitch angle is greater than the maximum pitch angle of the drone, it means that the drone cannot complete normal straight climb in the current flight segment. At this time, the control method of the drone circling and rising before flying forward / flying forward and then circling and rising must be set in the current flight segment to ensure that the drone can safely reach the next target waypoint. For the flight segments where the drone's altitude climbing ability can meet the requirements, the normal control method of slope control is used to ensure the terrain-imitating flight effect.

7. A method for planning an unmanned aerial vehicle terrain-simulating flight path by combining offline and online operations as described in claim 1, characterized in that: The specific method of step 203 is: Err=THei·η ΔH=|THei-Hei| Among them, Err is the calibration threshold, ΔH is the difference between the target height and the actual height, η is the error range, and |THei is the target height above the ground. According to the above formula, when ΔH is greater than the calibration threshold Err, the height of the drone is fine-tuned accordingly with an amplitude of |ΔH-Err|.

Images