Maps including data for determining routes for aircraft during GNSS failures
The camera captures images and divides maps and assigns tracking parameters, solving the problem of unmanned aerial vehicle navigation failure when GNSS signal errors, and realizing accurate position estimation and route planning in areas lacking unique characteristics.
Patent Information
- Application Number
- CN202080079526.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-11-20
- Filing Date
- 2020-09-08
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
GNSS-based navigation systems are difficult to accurately determine the location of unmanned aerial vehicles when signal errors or weak, especially in areas without unique characteristics, such as large bodies of water, deserts or forests, resulting in navigation failure.
The position image is captured by the camera, a map is generated and divided into multiple segments, and tracking parameters are assigned based on the number of features in the image and the uniqueness of the feature in the image, which is used to assist the navigation system to determine the location and route of the unmanned aerial vehicle.
When GNSS signals are lost, the camera navigation system can accurately estimate the location of the unmanned aircraft and plan the route to ensure that the aircraft can safely reach the destination and avoid navigation failure.
Smart Images

Figure CN114867988B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application is based on U.S. application No. 16 / 689,872 filed on November 20, 2019, the contents of which are incorporated herein by reference in their entirety. Technical Field
[0003] The present application relates to the technical field of unmanned aerial vehicles, and more particularly to an unmanned aerial vehicle and a map including data for determining a route for the aircraft during a GNSS failure. Background Art
[0004] An unmanned aerial vehicle (UAV), also known as an autonomous vehicle, is a vehicle that is capable of traveling without the physical presence of a human operator.
[0005] The position and orientation of the UAV should be known for various reasons, for example, in order to maintain the desired flight path of the UAV, to avoid obstacles, to avoid the flight path of commercial aircraft, for accurate delivery of payloads, for accurate coverage of specific ground areas when the UAV is used for telecommunications, etc. In practice, unmanned aerial vehicles (UAVs) usually carry a primary navigation system based on a global navigation satellite system (GNSS) sensor on board to determine the position of the UAV. However, for various reasons, the GNSS-based navigation system may fail, making the UAV inoperable. For example, the GNSS signal may be erroneous or too weak to be used by the UAV. In addition, the GNSS equipment of the UAV may also fail, making the UAV at least unable to complete its mission. Summary of the invention
[0006] Embodiments of the present disclosure provide an unmanned aerial vehicle (UAV) comprising: a propulsion system; a global navigation satellite system (GNSS) sensor; a camera; and a controller comprising logic, the logic responsive to execution of the controller causing the UAV to: in response to detecting a loss of tracking by the GNSS sensor: determine an estimated position of the UAV on a map based on a position image captured by the camera; determine a route to a destination using tracking parameters embedded in the map, wherein the map is divided into a plurality of segments and the tracking parameters indicate an ease of determining the position of the UAV using images captured by the camera with respect to each of the segments; and control the propulsion system so that the UAV follows the route to the destination.
[0007] An embodiment of the present disclosure provides a method for map processing, comprising: obtaining at least one image of an area; determining the number and feature uniqueness of features in the at least one image; generating a map of the area; dividing the map into map segments; and assigning tracking parameters to each of the map segments based on the number and feature uniqueness of features in the at least one image.
[0008] Embodiments of the present disclosure provide a non-transitory computer-readable medium having computer-executable instructions stored thereon, which, in response to execution by a computer, cause the computer to: obtain at least one image of an area; determine the number and feature uniqueness of features in the at least one image; generate a map of the area; divide the map into map segments; and assign tracking parameters to each of the map segments based on the number and feature uniqueness of features in the at least one image. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Many of the attendant advantages of the embodiments of the present disclosure will become more readily appreciated and better understood when reference is made to the following detailed description taken in conjunction with the accompanying drawings, in which:
[0010] Figure 1A is a top isometric view of a UAV according to an embodiment of the present technology;
[0011] Figure 1B is a bottom isometric view of a UAV according to an embodiment of the present technology;
[0012] Figure 1C is a schematic diagram of control electronics for a UAV according to an embodiment of the present technology;
[0013] Figure 2 is a schematic diagram of a UAV in operation according to an embodiment of the present technology;
[0014] Figure 3 is an image captured by a UAV camera according to an embodiment of the present technology;
[0015] Figure 4A A schematic diagram of a three-dimensional map according to an embodiment of the present technology is shown;
[0016] Figure 4B-4D shows a two-dimensional layer of a three-dimensional map according to an embodiment of the present technology;
[0017] Figure 5 is a flowchart of a method for generating and using a three-dimensional map according to an embodiment of the present technology.
[0018] Figure 6 is a flowchart of a method for UAV navigation according to an embodiment of the present technology. DETAILED DESCRIPTION
[0019] The camera-based navigation technology described below can alleviate the single point of failure problem for UAVs that primarily use the Global Navigation Satellite System (GNSS) for navigation.
[0020] Conventional UAVs typically use a primary GNSS-based navigation system to determine the location of the UAV. Some embodiments of the present disclosure relate to using a camera-based navigation system, when the GNSS signal is erroneous or weak, generating a map after determining the location of the UAV and using the map to determine a route for the UAV and navigate to the destination. This camera-based navigation system may be referred to as a backup or guidance navigation system. Camera-based navigation systems may find it difficult to maintain an accurate estimate of the UAV's position in some areas that lack unique features, such as over large bodies of water, deserts, or forests. Therefore, a camera-based navigation system may use a map with tracking parameters (wherein the tracking parameters indicate the ease of determining the location of the UAV using images captured by the camera) in order to plan and follow a route to the destination. A map including tracking parameters can be generated by dividing a map area into segments and assigning tracking parameters to each segment based on features in the map area.
[0021] Figure 1A 1 is a top isometric view of a UAV 100 according to an embodiment of the present technology. The UAV 100 includes a wing 124, a boom 126, and a fuselage 120. In some embodiments, the wing 124 carries a horizontal propulsion unit 112, and the boom 126 carries a vertical propulsion unit 114. In some embodiments, the boom 126 terminates in a rudder 122 for improved yaw control of the UAV 100.
[0022] In operation, power for the propulsion units 112, 114 may be provided by a battery bay of the fuselage 120. In some embodiments, the fuselage 120 also includes an avionics bay and a delivery unit for handling payloads. The fuselage 120 may carry control electronics 130.
[0023] Figure 1B 1 is a bottom isometric view of a UAV 100 according to an embodiment of the present technology. In some embodiments, the UAV 100 carries a camera 136 pointing toward the ground. The camera 136 can be a digital camera operating in the visible spectrum. In other embodiments, the camera 136 can operate in the infrared spectrum when operating, for example, at night, or in the ultraviolet spectrum when operating, for example, in cloudy weather. In some embodiments, the UAV 100 can include multiple cameras 136 configured to operate in different spectrums.
[0024] In some embodiments, the camera 136 is carried by a pivot mechanism 137. In operation, the pivot mechanism 137 can adjust the angle of the camera 136 based on input from the control electronics 130. In response to input from the control electronics 130, the pivot mechanism 137 can orient the camera 136 perpendicular to the ground to provide a common reference angle for images acquired by the camera 136. In other embodiments, the pivot mechanism 137 can orient the camera 136 at other angles.
[0025] Figure 1C 1 is a schematic diagram of the control electronics of the UAV 100 according to an embodiment of the present technology. The control electronics 130 may include a controller 132, a GNSS sensor 134, and a transceiver 138. The controller 132 may communicate with the camera 136 and control the UAV 100. The GNSS sensor 134 may receive a global navigation satellite system signal from a global navigation satellite to determine the location of the UAV 100. In some embodiments, the GNSS sensor 134 may communicate with the controller 132, and the controller 132 may determine the location of the UAV 100. The controller 132 may be a computer, a CPU, a processor, or other control circuit system. In addition, the controller 132 may include a memory including instructions, codes, or logic that, when executed, causes the controller 132 to control the UAV 100. The transceiver 138 may receive and send signals to communicate via direct communication or indirect communication (such as via a cell phone tower or other telecommunications network).
[0026] Figure 2 is a schematic diagram of a UAV 100 in operation according to an embodiment of the present technology. In some embodiments, the UAV 100 acquires images via a camera 136 while flying in a direction 201 (also referred to as a flight direction or flight path) at an altitude of approximately 30-150 m above the ground. During flight, images of a ground area 241 may be acquired using 640×480, 1280×800, or another pixel resolution of the camera 136. The illustrated sample acquired image includes feature points of objects 245 (spire), 243 (outline of a building), 247 (roof of a building), 249 (body of water), etc. In some embodiments, the UAV 100 carries multiple cameras. In many embodiments, the X, Y, Z coordinates of the feature point objects 245, 243, 247, 249 are known and may be included in the land map as auxiliary data.
[0027] As the UAV 100 flies in direction 201, additional images may be acquired and sent to a network transceiver 250 (e.g., a cell phone tower). The network transceiver 250 may forward the images to a base station 255. In some embodiments, the base station 255 includes a computer and / or server capable of processing images acquired by the camera 136 or other information sent by the transceiver 138.
[0028] Figure 3 360 is an example image 360 captured by camera 136 in accordance with an embodiment of the present technology. Controller 132 may process image 360 and identify features such as building 366, intersection 364, and road 362. These features are merely examples, and any number of other features may also be identified, such as bodies of water, fields, forests, landmarks, statues, and other permanent features.
[0029] Figure 4A A schematic diagram of a three-dimensional map 400 according to an embodiment of the present technology is shown. The three-dimensional map 400 includes several two-dimensional layers 402, 404, 406. Each two-dimensional layer 402, 404, 406 can be divided into sections 472, 474, 476, respectively. Each two-dimensional layer 402, 404, 406 can correspond to a UAV 100 located at different heights above the ground. Each section 472, 474, 476 can be associated with a location and can have tracking parameters embedded in the three-dimensional map 400 and assigned to the sections 472, 474, 476. The tracking parameters can be numbers (e.g., numbers between 0-10) or levels (e.g., poor, average, good, very good). The tracking parameters can indicate the ease of tracking the location of the UAV 100 using images captured by the camera 136 when flying over an area associated with the sections 472, 474, 476 of the map 400. To reiterate, the two-dimensional layers 402 , 404 , 406 together form the three-dimensional map 400 .
[0030] The three-dimensional map 400 may include three layers: a first two-dimensional layer 402 for a height below 50 meters; a second two-dimensional layer 404 for a height of 50-100 meters; and a third two-dimensional layer 406 for a height of 100+ meters. The first, second and third two-dimensional layers 402, 404, 406 are divided into first, second and third sections 472, 474, 476, respectively. Each of the first, second and third sections 472, 474, 476 may form a grid in the first, second and third two-dimensional layers 402, 404, 406, respectively. The first section 472 may be about 10m by 10m. The second section 474 may be about 20m by 20m. The third section 476 may be about 40m by 40m. This is just an example, and the three-dimensional map 400 may be divided into any number of two-dimensional layers, and these layers may correspond to any height. The segments 472, 474, 476 of the two-dimensional layers 402, 404, 406 may also be of any size. In some applications, it may be more important to know the exact location of the UAV 100 at a lower altitude. Therefore, in some embodiments, the first segment 472 may be 1 m by 1 m.
[0031] In some embodiments, several three-dimensional maps 400 may be used together to form a four-dimensional map. Each different three-dimensional map 400 forming the four-dimensional map may include images and tracking parameters for different times, such as different times of the year and different times of the day.
[0032] Figure 4B-4D Two-dimensional layers 402, 404, 406 of a three-dimensional map 400 according to an embodiment of the present technology are shown. As shown, each of the two-dimensional layers 402, 404, 406 is associated with the same region of the world. However, because the ease of tracking the location of the UAV 100 can be different depending on the altitude of the UAV 100, different tracking parameters can be assigned to the sections 472, 474, 476 of the two-dimensional layers 402, 404, 406 associated with the same region of the world. In addition, because the features on the ground will appear smaller as the UAV 100 grows in altitude, the sizes of the sections 472, 474, 476 of the two-dimensional layers 402, 404, 406 are different and are associated with the altitude of the UAV 100 for the two-dimensional layers 402, 404, 406.
[0033] For demonstration purposes, the map area in the illustrated two-dimensional layers 402, 404, 406 only includes an area of a few hundred meters. In practice, the two-dimensional layers 402, 404, 406 may cover the entire area and be much larger than the depicted two-dimensional layers 402, 404, 406.
[0034] Figure 55 is a flow chart 500 of a method for generating and using a three-dimensional map 400 according to an embodiment of the present technology. The method can be performed by a base station 255 or another computing system. The method begins at block 510, where the base station 255 obtains one or more images of the area to be mapped. The images can be satellite images, aerial images, or other images taken from above the area. The images can also include street-level images. For some areas, such as urban areas with tall buildings, the images can include images facing the side of the building or taken upward.
[0035] At block 520, the base station 255 can determine the number of features included in the image, and for each feature, determine the uniqueness of the feature. The uniqueness of a feature is related to the ease with which the feature can be uniquely identified. For example, the uniqueness of a building with an uncommon architectural structure is much greater than the uniqueness of a residence with a standard floor plan. For another example, an intersection of three roads is more unique than an intersection of two roads. Temporary features of the image, such as pedestrians and vehicles, can be ignored because they are unlikely to be in the same position when the UAV100 uses a map to navigate. In some embodiments, a combination of features can also be considered to contribute to uniqueness. For example, two buildings with relatively common separate architectural structures may be unique as a pair due to their relative position or orientation to each other. In addition to visible features, tracking parameters can also be derived from image quality metrics such as image contrast.
[0036] At block 530, the base station 255 may generate a map of the area and divide the map into two-dimensional layers 402, 404, 406 and sectors 472, 474, 476. The map layers 402, 404, 406 are two-dimensional because they represent the UAV 100 flying in a two-dimensional plane (if the ground is flat, the two-dimensional layers 402, 404, 406 may be contoured at the changing elevation of the ground), however each two-dimensional layer 402, 404, 406 may include more than one flat image. In some embodiments, as shown in FIG. Figure 4B-4DAs shown, two-dimensional layer 402,404,406 can be satellite image or other overhead image or road map of ground.In other embodiments, two-dimensional layer 402,404,406 can include images taken from some or all sections 472,474,476 of map 400 in many or all directions.This may be especially true for two-dimensional layer 402,404,406 associated with lower altitude, because the view of feature may change significantly according to the angle of observing feature.When using four-dimensional map, each three-dimensional map may include two-dimensional layer 402,404,406 with images taken at different times of day or year, because many areas may appear differently according to the time of day or the time of year.For example, the number of leaves on a tree or the amount of water in a lake or reservoir may change significantly in a year.In addition, some features, such as street lights, may be more useful for navigation at night, and other features, such as natural landmarks, may not be visible at night.
[0037] As described above, the segments 472, 474, 476 associated with the lower height two-dimensional layers 402, 404, 406 may be smaller than the segments 472, 474, 476 associated with the higher height two-dimensional layers 402, 404, 406. Thus, the lower layers may include several times more images than the upper layers.
[0038] At block 540, the base station 255 may assign tracking parameters to each of the segments 472, 474, 476 based on the number of features within the field of view of the UAV 100 at the location and altitude associated with the segment 472, 474, 476 and the feature uniqueness of the features. The tracking parameters may be numerical or other values. If there are a large number of features that can be identified, or if there are specific features that are very unique and easy to identify, the tracking parameters may be higher. If there are very few features, or if it is difficult to determine an accurate location based on the visible features because the feature uniqueness is low, the tracking parameters may be lower.
[0039] As the UAV 100 flies over an area, the tracking parameters may be continuously updated or improved based on data recorded from the UAV 100. The base station 255 may receive information from the UAV 100 and compare the GNSS position with the visually derived position to improve the tracking parameters for a given location. The controller 132 of the UAV 100 may also improve the tracking parameters with or without communicating with the base station 255.
[0040] At block 550, the base station 255 can send the three-dimensional map 400 to the UAV 100. The three-dimensional map 400 can be sent using the network transceiver 250 or by any other means of sending information. The three-dimensional map 400 can be sent when the UAV 100 is on the ground or in the air. The UAV 100 can also send images to the base station 255 while traveling, and the base station 255 can use these images to update the three-dimensional map 400.
[0041] At block 560, the UAV 100 may use the three-dimensional map 400 to determine a route for the UAV 100 based on the tracking parameters embedded in the three-dimensional map 400. Alternatively, the UAV 100 may communicate with the base station 255, and the base station may use the three-dimensional map 400 to determine a route for the UAV 100 based on the tracking parameters of the map. The route determined based on the tracking parameters may be planned to avoid areas of the map where the tracking parameters are low.
[0042] Figure 6 600 is a flow chart of a method for navigation of a UAV 100 according to an embodiment of the present technology. In some embodiments, the method may include additional steps, or may be practiced without all the steps shown in the flow chart. In addition, in some embodiments, the order of the steps listed may be changed.
[0043] The method begins at block 610, where the control electronics 130 detects a loss of GNSS tracking. This may occur when the GNSS sensor 134 detects a loss of GNSS signals, when the controller 132 detects a GNSS sensor 134 failure or some other failure of GNSS tracking. In response, the controller 132 may switch to camera-based navigation.
[0044] At block 620, the controller 132 may control the camera 136 to capture at least one position image, such as image 360. The position image may be taken in any direction. However, typically at least one position image will be taken of the area directly below the UAV 100. In many cases (and depending on local regulations), the UAV 100 will fly between 30m and 150m above the ground. Within this range, the camera 136 may capture images of several blocks in an urban or suburban area (as shown in image 360). In many cases, the amount of information that can be collected from the position images taken at these heights will be sufficient to estimate the position of the UAV 100.
[0045] At block 630, the controller 132 may determine an estimated location of the UAV 100. The controller 132 may analyze features of the location image (such as roads, intersections, buildings, fields, bodies of water, forests, etc.). The controller 132 may then compare these features to features found in the map 400 stored in the control electronics 130. The controller 132 may also use the current destination, flight path, and last tracked GNSS location when determining the estimated location. Estimating the location of the UAV 100 may include extracting features of interest from the location image, matching the features of interest with features on the map, and estimating the location of the UAV 100 based on the matched features of interest. Buildings 366, intersections 364, and roads 362 may be examples of features of interest.
[0046] In some areas (such as rural areas, densely forested areas, etc.), it may be more difficult for UAV 100 to determine the estimated position based on the position image taken at a lower altitude. Therefore, the controller 132 can control the propulsion unit (including the vertical propulsion unit 114 and the horizontal propulsion unit 112) to make the UAV 100 rise to a higher altitude or move to another position where a position image of a wider or more unique area can be taken. The change in altitude may be unnecessary. However, if the estimated position cannot be determined at the current position of the UAV, the controller 132 can control the propulsion units 114 and 112 of the UAV 100 to change the altitude of the UAV or move the UAV to another position to capture a new position image. The controller 132 can also determine several possible estimated positions based on the available information.
[0047] At block 640, the controller 132 may determine a route to a destination. The destination may be the destination that the UAV 100 was heading to before GNSS tracking was lost, or may be a new destination (such as where the UAV 100 departed, a waypoint, or other destination). The destination may be determined based on the estimated location of the UAV 100.
[0048] A three-dimensional map 400 (or a four-dimensional map) can be used to determine a route. Specifically, the tracking parameters of different sections 472, 474, 476 of the two-dimensional layers 402, 404, 406 assigned to the three-dimensional map 400 can be used to determine the route. When a four-dimensional map is used, a three-dimensional map (included in the four-dimensional map) can be selected based on time, date, weather, or other factors. Other factors can also be used to determine the route, including: the length of the route to the destination, the amount of electricity (or fuel) remaining in the UAV 100, the time of day, local regulations, etc. In some embodiments, the route can follow a road to reach the destination.
[0049] It is important for the UAV 100 to be able to track its position as it follows a route. For example, if the UAV 100 travels above a military base or other restricted area, there may be serious consequences. The tracking parameters indicate how easy it is to track the position of the UAV 100 using the camera 136 in a specific area at a specific altitude. Therefore, by using the tracking parameters to plan a route, the UAV 100 can be kept in an area where the position of the UAV 100 can be tracked using the camera 136, thereby avoiding the loss of the UAV 100. Therefore, using the tracking parameters when determining a route provides technical benefits, including ensuring that in the event of a GNSS tracking failure at any point during the route, reliably determining the position using the camera 136 will be available as an emergency navigation option.
[0050] Tracking parameters can be used to plan a route in a variety of ways. For example, a route may include only segments 472, 474, 476 of the map 400 whose tracking parameters meet a particular threshold. To reiterate, a route may not pass through segments 472, 474, 476 of the map 400 that have tracking parameters that do not meet the threshold. In the example where the tracking parameter is a number between 0-10, the threshold may be 3, so that the route will not include flying over any segments 472, 474, 476 where the tracking parameter is 3 or lower. In addition, a route may be determined to have less than a threshold number of segments 472, 474, 476 that do not meet the threshold.
[0051] As another example, the route may be determined using an average tracking parameter along the route. For example, the shortest route to the destination for which the average tracking threshold exceeds a threshold may be determined. As another example, the route may include only a threshold number of segments 472, 474, 476 for which the tracking threshold is below the threshold. For example, the route may include only 5% of the segments 472, 474, 476 for which the tracking parameter is below 4.
[0052] In the event that the controller 132 has determined several potential estimated positions, the determined route may include one or more segments 472, 474, 476 with very high tracking parameters in order to better determine the estimated position of the UAV 100. For example, if there are two potential estimated positions, and each of them has a segment 472, 474, 476 with high tracking parameters to the east, the route may begin with the UAV 100 traveling east to determine a better estimated position of the UAV 100.
[0053] The tracking parameters are specific to one segment 472, 474, 476 of the two-dimensional layers 402, 404, 406, and the two-dimensional layers 402, 404, 406 are specific to a particular altitude. Thus, a route may include a change in altitude in order to travel through the segment 472, 474, 476 at a higher tracking parameter. A route may also include changes in altitude for other reasons, such as local regulations requiring the UAV to fly within a particular altitude range.
[0054] At block 650, the controller 132 controls the propulsion units 112, 114 to cause the UAV to follow the route. At block 660, as the UAV 100 follows the route, the controller 132 controls the camera 136 to capture a confirmation image. Like the position image, the confirmation image can be captured with the camera 136 facing in any direction. However, typically at least one confirmation image will include the area directly below the UAV 100. The controller 132 can process the confirmation image in the same manner as the position image. Image 360 can also be a confirmation image.
[0055] At block 670, the controller 132 uses the confirmation image to identify the transition location and confirm that the transition location is on the route. The controller 132 can use the confirmation image to track the progress on the route and also confirm that the initial estimated location is correct. If the controller 132 is unable to confirm that the transition location is on the route, the controller 132 can return to block 630 and determine a new estimated location.
[0056] Blocks 660 and 670 may be repeated continuously while the UAV 100 is in the air. To reiterate, the controller 132 may use the camera 136 to continuously monitor the progress of the UAV 100 along the route. The controller 132 and camera 136 may perform these functions even when the UAV 100 is traveling using GNSS tracking.
[0057] Alternatively, the operations of flowchart 600 may be performed based on instructions from base station 255. For example, at block 610, base station 255 may communicate with UAV 100 to receive a signal indicating that GNSS tracking is lost. At block 620, base station 255 may send instructions to camera 136 of UAV 100 to capture a position image. At block 630, UAV 100 may send the position image to base station 255, and base station 255 may determine the estimated position of UAV 100. At block 640, base station 255 may then determine a route to the destination. At block 650, base station 255 may send the route to UAV 100, and also send directions to UAV 100 to control the propulsion system to follow the route. At block 660, base station 255 may also send instructions to cause camera 136 to capture a confirmation image and send the confirmation image to base station 255. At block 670, the base station may track the progress of UAV 100 and confirm that the transition location is on the route.
[0058] Many embodiments of the above-mentioned technology can take the form of computer executable or controller executable instructions, including routines stored on non-temporary computer readable media and executed by programmable computers or controllers. It will be appreciated by those skilled in the relevant art that the technology can be practiced on computer / controller systems other than those shown and described above. The technology can be embodied in a special-purpose computer, application-specific integrated circuit (ASIC), controller or data processor that is specially programmed, configured or constructed to execute one or more of the above-mentioned computer executable instructions. In many embodiments, any logic or algorithm described herein can be implemented with software or hardware, or a combination of software and hardware.
[0059] From the foregoing, it will be appreciated that, for purposes of illustration, specific embodiments of the technology have been described herein, but that various modifications may be made without departing from the present disclosure. Furthermore, while various advantages and features associated with specific embodiments have been described in the context of those embodiments, other embodiments may also exhibit such advantages and / or features, and not all embodiments must exhibit such advantages and / or features to fall within the scope of the present technology. Therefore, the present disclosure may encompass other embodiments not expressly shown or described herein.
Claims
1. An unmanned aerial vehicle (UAV), include: Propulsion system; Global Navigation Satellite System GNSS sensors; camera; and A controller comprising logic that, in response to execution of the controller, causes the UAV to: In response to detecting a loss of tracking by the GNSS sensor: determining an estimated position of the UAV on a map based on the position image captured by the camera; determining a route to a destination using tracking parameters embedded in the map, wherein the map is divided into a plurality of segments and each tracking parameter indicates an ease with which a position of the UAV can be determined using one or more images captured by the camera with respect to a respective segment; and The propulsion system is controlled so that the UAV follows the route to the destination.
2. The UAV according to claim 1, in, Determining an estimated position of the UAV on a map based on images captured by the camera includes: An estimated position of the UAV on a map is determined based on images captured by the camera and the last tracked position of the UAV from the GNSS sensor.
3. The UAV according to claim 1, in, The logic also causes the UAV, in response to detecting a loss of tracking by the GNSS sensor, to control the propulsion system to cause the UAV to ascend before the camera captures the position image.
4. The UAV according to claim 1, in, The tracking parameters are numerical values.
5. The UAV according to claim 4, in, Determining the route to the destination using tracking parameters embedded in the map includes determining the route such that the route does not pass through sections of the map having tracking parameters that do not meet a threshold.
6. The UAV according to claim 4, in, The controller also includes logic for determining the route based on an average tracking parameter for a segment of the map in the route.
7. The UAV of claim 4, in, The controller also includes logic for determining the route so that the UAV, while following the route, travels through less than a threshold number of areas associated with segments of the map having tracking parameters that do not meet a threshold.
8. The UAV of claim 1, in, The map is a three-dimensional map, wherein the three-dimensional map includes multiple two-dimensional layers, and wherein different two-dimensional layers of the three-dimensional map correspond to different altitudes of the UAV.
9. The UAV according to claim 8, in, Determining the route includes determining a route that includes changes in altitude of the UAV.
10. The UAV of claim 1, in, Determining an estimated position of the UAV on a map based on the position image captured by the camera includes: extracting features of interest from the location image; matching the feature of interest to features on the map; and The position of the UAV is estimated based on the matched features of interest.
11. The UAV of claim 1, in, Determining the route includes planning the route to preferentially pass through a first section of the map indicated by the tracking parameters where it is believed to be easier to determine the position of the UAV using the image, while avoiding a second section of the map indicated by the tracking parameters where it is believed to be more difficult to determine the position of the UAV using the image.
12. A non-transitory computer readable medium having stored thereon computer executable instructions which, when executed by one or more processors of an unmanned aerial vehicle (UAV), cause the UAV to: In response to detecting a loss of tracking by a Global Navigation Satellite System (GNSS) sensor of the UAV: determining an estimated position of the UAV on a map based on an image of the position captured by a camera of the UAV; A route to a destination is determined using tracking parameters embedded in the map, wherein: The map is divided into a plurality of sectors, and each tracking parameter indicates an ease of determining a position of the UAV using one or more images captured by the camera with respect to a corresponding sector; and A propulsion system of the UAV is controlled so that the UAV follows the route to the destination.
13. The non-transitory computer readable medium of claim 12, in, Determining the route includes planning the route to preferentially pass through a first section of the map indicated by the tracking parameters where it is believed to be easier to determine the position of the UAV using the image, while avoiding a second section of the map indicated by the tracking parameters where it is believed to be more difficult to determine the position of the UAV using the image.
14. The non-transitory computer readable medium according to claim 12 or 13, in, The instructions cause the UAV to: determine the route so that the UAV, while following the route, travels through less than a threshold number of areas associated with segments of the map having tracking parameters that do not meet a threshold.
15. The non-transitory computer readable medium according to claim 12 or 13, in, Determining the route to the destination using tracking parameters embedded in the map includes determining the route such that the route does not pass through sections of the map having tracking parameters that do not meet a threshold.
Citation Information
Patent Citations
Multi-beam landform partitioning method based on statistics of node information content
CN106875486A
Dynamic path planning method for AUV (Autonomous Underwater Vehicle) submarine topography matching
CN108871351A
