A method and system for autonomous return of a swarm of unmanned aerial vehicles in a GPS-denied environment

Abstract

The application discloses a kind of GPS rejection environment under cluster unmanned aerial vehicle autonomous return method and system, it includes the following steps: unmanned aerial vehicle cluster executes flight task;Unmanned aerial vehicle in cluster judges whether GPS signal is normal respectively: if yes, then autonomous return based on GPS signal;If no, then confirm whether the unmanned aerial vehicle of GPS signal anomaly in cluster has collected first return image containing return marker, when first return image is collected, return direction is obtained based on return marker;When first return image is not collected, calculate relative direction according to historical positioning information, and relative direction is used as return direction;GPS signal abnormal unmanned aerial vehicle in cluster flies based on return direction, and when the unmanned aerial vehicle of GPS signal anomaly in cluster collects second return image containing landing marker, landing direction is determined based on landing marker to carry out landing.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method and system for autonomous return of swarm UAVs in a GPS denied environment. Background Technology

[0002] In recent years, with the rapid development of unmanned aerial vehicle (UAV) technology, UAVs have become increasingly widely used in logistics distribution, security inspection, military reconnaissance and other fields due to their flexibility, remote operation and the fact that they do not require on-site operation by personnel.

[0003] Currently, existing unmanned aerial vehicle (UAV) swarms generally rely on satellite navigation signals sent by the Global Positioning System (GPS) for real-time positioning. They achieve positioning and navigation by receiving satellite signals and complete return path planning and landing. They have advantages such as high positioning accuracy, convenient path planning, and wide signal coverage, making them an ideal choice for achieving remote and high-precision operations.

[0004] However, the return stability of such drone swarms is greatly affected by GPS signals. When GPS signals are interfered with, blocked, or completely disabled (GPS denial environment), drones in the swarm often fail to return to the takeoff point accurately, or even get lost or collide. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that the present invention discloses a method and system for autonomous return of swarm unmanned aerial vehicles in a GPS denied environment, which can solve the problem that unmanned aerial vehicles in the existing unmanned aerial vehicle swarm cannot achieve autonomous return and high-precision return when GPS signals are completely denied or interfered with.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment, which includes the following steps: Unmanned aerial vehicle swarms perform flight missions; Each UAV in the cluster determines whether its GPS signal is normal: if so, it autonomously returns to home based on the GPS signal; if not, it checks whether the UAV with abnormal GPS signal in the cluster has collected a first return-to-home image containing the return-to-home marker. If the first return-to-home image has been collected, the return-to-home direction is obtained based on the return-to-home marker; if the first return-to-home image has not been collected, the relative orientation is calculated based on historical positioning information, and the relative orientation is used as the return-to-home direction. The UAVs with abnormal GPS signals in the cluster fly based on the return direction, and when the UAVs with abnormal GPS signals in the cluster acquire a second return image containing landing markers, the landing direction is determined based on the landing markers. Unmanned aerial vehicles in the cluster with abnormal GPS signals landed based on the stated landing direction.

[0007] To address the problem in existing technologies where UAVs in a swarm cannot achieve autonomous and high-precision return-to-home operations when GPS signals are completely denied or interfered with, this invention proposes a novel autonomous return-to-home method for swarm UAVs in GPS-denied environments. This method combines an optimized beacon guidance scheme and a visual recognition scheme with the aforementioned GPS positioning and guidance scheme. Even when GPS signals are completely denied or interfered with, the visual recognition scheme can still identify a first return-to-home image containing return-to-home markers and a second return-to-home image containing landing markers. Based on these first and second return-to-home images, the return-to-home direction and landing direction are determined respectively, enabling autonomous return-to-home and precise landing of UAVs within the swarm. This method has promising prospects and application value.

[0008] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment described in this invention, determining whether the GPS signal is normal specifically involves: The GPS signal of any UAV in the UAV cluster is detected at the first moment. If the GPS signal at the first moment is not interfered with or is not invalid, the GPS signal of the UAV is determined to be normal and the UAV continues to perform the flight mission. If the GPS signal at the first moment is interfered with or is not invalid, the GPS signal of the UAV is determined to be abnormal and the UAV with abnormal GPS signal is marked as abnormal UAV. Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment described in this invention, the following steps are performed on the abnormal unmanned aerial vehicle: Control the abnormal unmanned aerial vehicle to hover in place and wait for a preset time; The GPS signal of the abnormal UAV at the second moment after waiting for the preset time is detected. If the GPS signal at the second moment is not interfered with or fails, the GPS signal of the abnormal UAV is determined to be normal, and the abnormal UAV returns autonomously based on the GPS signal. If the GPS signal of the abnormal UAV at the second moment is interfered with or fails, the GPS signal of the abnormal UAV is determined to be abnormal.

[0009] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment described in this invention, the return marker and / or the landing marker include at least one of the following: a QR code, a barcode, and a light strip with a preset shape.

[0010] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment described in this invention, the return marker and / or the landing marker are infrared light strips of a preset shape.

[0011] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles (UAVs) under GPS denial conditions described in this invention, whether the UAVs in the swarm with abnormal GPS signals have acquired a first return image containing return markers, and if the first return image has been acquired, the return direction is obtained based on the return markers, specifically as follows: The camera on the abnormal unmanned aerial vehicle acquires a first return image containing a return marker, and based on the first return image, the coordinate position of the return marker in the camera coordinate system is obtained; The camera's attitude is calculated based on the camera's onboard parameters on the abnormal unmanned aerial vehicle; Based on the relative positional relationship between the camera and the abnormal UAV, the pose of the abnormal UAV in the camera coordinate system is calculated; Based on the coordinates of the return marker in the camera coordinate system, the pose of the abnormal UAV in the camera coordinate system, and the attitude of the camera, the return direction of the abnormal UAV is calculated.

[0012] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment described in this invention, the historical positioning information includes: The preset return point location information includes the longitude coordinates and latitude coordinates of the return point; The latest GPS positioning information recorded by the abnormal UAV before the GPS signal became abnormal includes the longitude and latitude coordinates of the abnormal UAV.

[0013] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment described in this invention, the relative azimuth is calculated based on historical positioning information, specifically as follows:

[0014] Wherein, ∠a is the relative azimuth angle of the abnormal UAV, FD is the difference in longitude coordinates between the return point and the abnormal UAV, FA is the difference in latitude coordinates between the return point and the abnormal UAV, L1 is the longitude coordinate of the abnormal UAV, L0 is the longitude coordinate of the return point, B1 is the latitude coordinate of the abnormal UAV, and B0 is the latitude coordinate of the return point.

[0015] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles (UAVs) in a GPS-denied environment described in this invention, when the UAV swarm flies based on the return direction, it further includes: The camera of any UAV in the UAV cluster acquires continuous return-to-home images and obtains the number of identical feature points in the continuous return-to-home images. When the number of identical feature points is greater than a preset threshold, the optical flow is determined to be valid, and the abnormal UAV is controlled to fly at a constant speed. When the number of identical feature points is less than the preset threshold, the optical flow is determined to be invalid, and the abnormal UAV is controlled to fly in attitude mode.

[0016] Furthermore, the method for autonomous return of swarm unmanned aerial vehicles in a GPS denied environment described in this invention also includes: the abnormal unmanned aerial vehicle performing hierarchical planning before returning to base. The abnormal unmanned aerial vehicle underwent layered planning before returning to base, specifically as follows: Obtain the ID of the abnormal unmanned aerial vehicle; The layer height is determined by the upper and lower limits of the layer. Based on the abnormal UAV ID and the layer height, the layer position of each abnormal UAV is determined by the integer division method. Control the abnormal unmanned aerial vehicle to descend to the layered position.

[0017] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS denied environment described in this invention, the unmanned aerial vehicles with abnormal GPS signals in the swarm fly based on the return direction, including: the abnormal unmanned aerial vehicle matching the corresponding alternate landing site ID before returning. Before returning, the abnormal unmanned aerial vehicle was matched with the corresponding alternate landing site ID, specifically: Obtain the ID of the abnormal unmanned aerial vehicle; Get the number of alternate landing sites; Based on the abnormal UAV ID and the number of alternate landing sites, the alternate landing site ID corresponding to each abnormal UAV is determined using the integer division method.

[0018] Furthermore, in the method for autonomous return of swarm unmanned aerial vehicles (UAVs) under GPS denial conditions described in this invention, when a UAV with abnormal GPS signal in the swarm acquires a second return image containing landing markers, the landing direction is determined based on the landing markers, including: When the abnormal UAV acquires a second return image containing the alternate landing site ID, the landing direction is determined based on the alternate landing site ID.

[0019] Accordingly, another objective of the present invention is to provide an autonomous return system for swarm unmanned aerial vehicles (UAVs) in a GPS-denied environment, which can be effectively applied to the aforementioned autonomous return method for swarm unmanned aerial vehicles in a GPS-denied environment. When executing the aforementioned autonomous return method for swarm unmanned aerial vehicles in a GPS-denied environment, the system can also ensure that UAVs in the swarm can still achieve autonomous return and precise landing of UAVs in the swarm even when GPS signals are completely denied or interfered with.

[0020] The autonomous return-to-home system for swarm unmanned aerial vehicles in a GPS-denied environment, as designed in this invention, includes: Return markers are placed along the return route of a swarm of unmanned aerial vehicles (UAVs) during a return flight mission. Multiple landing markers are placed at the return destination of the unmanned aerial vehicle swarm during its return flight mission; An unmanned aerial vehicle (UAV) swarm comprises multiple UAVs, each equipped with a GPS positioning device, an image acquisition device, and a controller. The controller is communicatively connected to the GPS positioning device and the image acquisition device. The GPS positioning device acquires GPS signals from the UAVs. The image acquisition device acquires a first return-to-home image containing a return-to-home marker and a second return-to-home image containing a landing marker matching the UAV. The GPS positioning device transmits GPS signals to the controller. The image acquisition device transmits the first and second return-to-home images to the controller, which executes the autonomous return-to-home method for the swarm of UAVs in a GPS-denied environment.

[0021] Furthermore, in the GPS-denied autonomous return system for swarm unmanned aerial vehicles described in this invention, the image acquisition device includes: The first camera is used to capture images in front of the unmanned aerial vehicle to capture the first return image containing the return markers; The second camera is used to capture images below the unmanned aerial vehicle to obtain a second return-to-home image that includes landing markers.

[0022] The beneficial effects of this invention are as follows: The autonomous return method and system for swarm unmanned aerial vehicles (UAVs) in GPS-denied environments designed in this invention, in addition to the traditional GPS positioning-guided return scheme, also incorporates an optimized beacon guidance scheme and a visual recognition scheme. Even when the UAV swarm encounters complete GPS signal denial or interference during its return flight mission, it can still utilize the visual recognition scheme integrated on the UAVs to identify a first return image containing return markers and a second return image containing landing markers. Based on these first and second return images, the return direction and landing direction are determined respectively, enabling autonomous return of the UAV swarm and precise landing of each UAV within the swarm. This invention has good prospects for promotion and application value. Attached Figure Description

[0023] Figure 1 This is a flowchart of one embodiment of the method for autonomous return of swarm unmanned aerial vehicles in a GPS denied environment described in this invention. Figure 2 This is a flowchart illustrating how the autonomous return-to-home system for swarm unmanned aerial vehicles in a GPS-denied environment, as described in this invention, controls a swarm of unmanned aerial vehicles to perform a return-to-home flight mission in one embodiment. Figure 3 The diagram illustrates the return of a swarm of unmanned aerial vehicles (UAVs) in one embodiment of the GPS-denied autonomous return system described in this invention. Figure 4 This is a schematic diagram of an infrared light strip in one embodiment of the method for autonomous return of swarm unmanned aerial vehicles in a GPS denied environment described in this invention. Figure 5 This is a hierarchical planning flowchart of the method for autonomous return of swarm unmanned aerial vehicles in a GPS denied environment according to one embodiment of the present invention. Figure 6 This is a flowchart illustrating the target alternate landing site ID matching process in one embodiment of the GPS denied autonomous return method for swarm unmanned aerial vehicles described in this invention. Detailed Implementation

[0024] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0025] In existing technologies, when GPS signals are interfered with, blocked, or completely disabled (GPS denial environment), unmanned aerial vehicle (UAV) swarms often encounter problems such as UAVs being unable to accurately return to their takeoff point, or even being lost or colliding.

[0026] To at least solve the above problems, such as Figure 1and Figure 2 As shown, this invention designs an autonomous return-to-home system for swarm unmanned aerial vehicles (UAVs) in a GPS-denied environment to control the UAV swarm to perform precise return-to-home maneuvers. This autonomous return-to-home system for swarm UAVs in a GPS-denied environment includes: Return markers are placed along the return route of a swarm of unmanned aerial vehicles (UAVs) during a return flight mission. Multiple landing markers are placed at the return destination of the unmanned aerial vehicle swarm during its return flight mission; An unmanned aerial vehicle (UAV) swarm comprises multiple UAVs, each equipped with a GPS positioning device, an image acquisition device, and a controller. The controller is communicatively connected to the GPS positioning device and the image acquisition device. The GPS positioning device acquires GPS signals from the UAVs. The image acquisition device acquires a first return-to-home image containing a return-to-home marker and a second return-to-home image containing a landing marker matching the UAV. The GPS positioning device transmits GPS signals to the controller. The image acquisition device transmits the first and second return-to-home images to the controller, which executes an autonomous return-to-home method for the swarm of UAVs in a GPS-denied environment.

[0027] In one embodiment of this invention, the image acquisition device mounted on the unmanned aerial vehicle (UAV) may specifically include a first camera and a second camera. The first camera is used to acquire images in front of the UAV to obtain a first return-to-home image containing return-to-home markers, and the second camera is used to acquire images below the UAV to obtain a second return-to-home image containing landing markers. Thus, the two cameras can effectively cooperate, with the first camera used for path guidance and the second camera used for precise landing. Their division of labor is clear, and they have high recognition efficiency.

[0028] Meanwhile, in certain specific embodiments, the aforementioned first camera and / or second camera can be selected as infrared imaging cameras. For example, the first camera can be specifically selected as an infrared imaging camera, and the second camera can be selected as a traditional imaging module. In this case, the first camera can shield visible light and can be paired with markers that emit infrared light to greatly reduce the influence of ambient light, improve visual recognition efficiency, and obtain strong anti-interference capabilities. It can be effectively applied to stable and reliable navigation in nighttime and low-visibility environments, meeting the needs of all-weather operation of unmanned aerial vehicles, and adapting to the return and landing requirements of special scenarios such as nighttime security patrols and nighttime emergency rescues.

[0029] Based on this, in the practical application of the above-mentioned autonomous return system for swarm unmanned aerial vehicles (UAVs) under GPS denial conditions, the present invention also designs a new autonomous return method for swarm unmanned aerial vehicles (UAVs) under GPS denial conditions. This method is implemented through the controllers of each UAV in the UAV swarm within the autonomous return system under GPS denial conditions. The autonomous return method for swarm unmanned aerial vehicles under GPS denial conditions can specifically include the following steps: 100: Unmanned aerial vehicle swarms perform flight missions; 200: Each UAV in the cluster determines whether its GPS signal is normal: if so, it autonomously returns to home based on the GPS signal; if not, it checks whether the UAV with abnormal GPS signal in the cluster has collected a first return-to-home image containing the return-to-home marker. If the first return-to-home image has been collected, the return-to-home direction is obtained based on the return-to-home marker; if the first return-to-home image has not been collected, the relative orientation is calculated based on historical positioning information, and the relative orientation is used as the return-to-home direction. 300: The unmanned aerial vehicle swarm flies based on the return direction, and each unmanned aerial vehicle in the swarm has a matching landing marker. 400: When an unmanned aerial vehicle (UAV) in the UAV cluster acquires a second return image containing a matching landing marker, it determines the landing direction based on the landing marker, and the UAV lands based on the landing direction.

[0030] Therefore, the autonomous return-to-home method for swarm UAVs in GPS-denied environments designed in this invention addresses the problems existing in the return-to-home flight missions of existing UAV swarms. Based on GPS positioning and navigation, it combines visual navigation and beacon guidance schemes to provide gradual path guidance for UAVs in the swarm during their return-to-home flight using return-to-home and landing markers. During the return-to-home flight mission, it can determine whether the GPS signal of the UAV swarm is interfered with or malfunctions, and use cameras on the UAVs in the swarm to capture images of the areas the UAVs have flown over. This results in a first return-to-home image containing the return-to-home markers and a second return-to-home image containing the landing markers. Based on these first and second return-to-home images, the return-to-home direction and landing direction are determined respectively, thereby achieving autonomous return-to-home and precise landing of the UAVs. This method has good prospects for promotion and application value.

[0031] It should be noted that in this invention, in step 100 above, the abnormal unmanned aerial vehicle performs layered planning before returning to base, specifically as follows: Obtain the ID (number 0-m) of the abnormal unmanned aerial vehicle. The layer height H = ab is determined by the upper limit height a and the lower limit height b of the layer. Based on the abnormal UAV ID and the layering height, the layering position of each abnormal UAV is determined by using the integer division method m / H to find the remainder. Control the abnormal unmanned aerial vehicle to descend to the layered position.

[0032] In some embodiments, in steps 102 and 103 above, the preset upper limit height and preset lower limit height can be user-defined inputs or determined based on known positioning information to obtain the lowest height in the UAV formation, which is then used as the upper limit height for layering; and in step 104 above, the height unit is meters, and layering is performed at 1-meter intervals. Figure 3 and Figure 5 In the illustrated embodiment, the target can be divided into three layers; simultaneously, in step 105 above, the remainder is obtained using the integer division method m / H to determine the target layering of the current unmanned aerial vehicle, for example: Assuming the layer height is 2 meters, it can be divided into two layers with a 1-meter interval, defined as layer 1 and layer 2. Currently, there are 10 abnormal UAVs waiting to return (the UAV IDs are numbered 0 to 9 in sequence, so the target layer for UAV 0 is layer 1, because 0 / 2 is divisible and the remainder is 0, which is layer 1; similarly, the target layer for UAV 1 is layer 2, because 1 / 2 is not divisible and the remainder is 1, which is layer 2). UAVs numbered 2 to 9 are assigned their corresponding target layers according to the above logic.

[0033] Accordingly, in this invention, in step 300 above, the abnormal unmanned aerial vehicle matches the corresponding alternate landing site ID before returning to base, specifically as follows: Obtain the ID of the abnormal unmanned aerial vehicle; Get the number of alternate landing sites; Based on the abnormal UAV ID and the number of alternate landing sites, the alternate landing site ID corresponding to each abnormal UAV is determined using the integer division method.

[0034] See Figure 6 The abnormal UAV ID is obtained, with a value from 0 to m. The n landing markers are then assigned IDs, ranging from 0 to (N-1). The remainder can be obtained using the integer division method m / N to determine the corresponding alternate landing site ID for each UAV, and thus the matching landing marker is obtained. The calculation is as follows: Suppose there are two alternate landing sites, with alternate landing site IDs 0 and 1 respectively. The current formation has 10 abnormal UAVs (abnormal UAV IDs 0 to 9). Then, the target alternate landing site ID of abnormal UAV 0 is 0 (because 0 / 2 is divisible and has a remainder of 0). Similarly, the target alternate landing site ID of abnormal UAV 1 is 1 (because 1 / 2 is not divisible and has a remainder of 1). Abnormal UAVs numbered 2 to 9 obtain their respective target alternate landing site IDs according to the above logic.

[0035] In this invention, to avoid misjudgment, in step 200 above, determining whether the GPS signal of the unmanned aerial vehicle cluster is normal can specifically employ the following steps 211-213: 211: Detect the GPS signal of any UAV in the UAV cluster at the first moment. If the GPS signal at the first moment is not interfered with or fails, it is determined that the GPS signal of the UAV is normal and the UAV continues to perform the flight mission. If the GPS signal at the first moment is interfered with or fails, it is determined that the GPS signal of the UAV is abnormal and the UAV with abnormal GPS signal is marked as an abnormal UAV, and proceed to the next step. 212: Control the abnormal unmanned aerial vehicle to hover in place and wait for a preset time; 213: Detect the GPS signal of the abnormal UAV at the second moment after waiting for the preset time. If the GPS signal at the second moment is not interfered with or fails, it is determined that the GPS signal of the abnormal UAV is normal, and the abnormal UAV returns autonomously based on the GPS signal. If the GPS signal of the abnormal UAV at the second moment is interfered with or fails, it is determined that the GPS signal of the abnormal UAV is abnormal.

[0036] The detection of whether the GPS signal is abnormal is based on the message information received by the GPS positioning device on the UAV. The response is timely and the judgment is accurate. In actual application, in order to ensure that the preset waiting time does not interfere with the return flight efficiency of the UAV, it can be controlled to 3-10 seconds according to actual needs.

[0037] In this invention, steps 211-213 are set up so that after judging that the GPS signal is abnormal at the first moment, the GPS signal of the UAV cluster is not immediately judged to be interfered with or malfunctioning. Instead, the UAV cluster is controlled to hover in place and wait for a preset time before the GPS signal of the UAV is detected and judged a second time. This is to avoid misjudgment caused by instantaneous interference, improve the accuracy of judgment, and reduce the risk of false triggering. This makes the autonomous return method of the UAV cluster under GPS denial environment effectively adaptable to complex actual scenarios and improves the applicability of the autonomous return method of the UAV cluster under GPS denial environment.

[0038] If the navigation mode is switched after only one GPS signal anomaly is detected, it is easy to cause the UAV's navigation logic to switch frequently, affecting the return path planning and flight stability. Furthermore, the design of secondary detection can be adapted to the application situation where the UAV is shuttling through complex areas such as tall buildings and jungles, where the GPS signal is prone to temporary obstruction (such as passing through the shadow of tall buildings), thus making the GPS signal detection more in line with the actual working environment and improving practicality.

[0039] Accordingly, it should be noted that in step 200 above, when the UAV cluster is performing a return flight mission, it will use the image acquisition device on the UAV to continuously acquire images of the area the UAV has flown over, in order to acquire a first return image containing return markers and a second return image containing landing markers. In one embodiment, the aforementioned return markers and / or landing markers can be specifically selected from at least one of the following: QR codes, barcodes, and light strips with a preset shape, to reduce deployment costs while improving adaptability; they are easily identifiable and adaptable to the needs of different scenarios.

[0040] See Figure 4 The image shows an infrared light strip diagram of one embodiment of the GPS-denied autonomous return-to-home method for swarm unmanned aerial vehicles (UAVs) in this invention. Two infrared light strips are placed parallel to each other, forming an equal sign. When used as a return-to-home marker, this pattern is placed on the UAV's return path, offering advantages such as fast detection speed and minimal interference. Another embodiment shows two infrared light strips intersecting to form a crosshair cursor pattern. When used as a landing marker, this pattern offers advantages such as a clear landing center and minimal interference.

[0041] In the method for autonomous return-to-home of swarm unmanned aerial vehicles (UAVs) in a GPS-denied environment of the present invention, whether the UAVs in the swarm with abnormal GPS signals have acquired a first return-to-home image containing a return-to-home marker is determined. If the first return-to-home image has been acquired, the return-to-home direction is obtained based on the return-to-home marker, specifically based on the following steps 221-224: 221: The camera on the abnormal unmanned aerial vehicle acquires a first return image containing a return marker, and based on the first return image, obtains the coordinate position of the return marker in the camera coordinate system; 222: The attitude of the camera is calculated based on the onboard parameters of the camera on the abnormal unmanned aerial vehicle; wherein, the onboard parameters may specifically include parameters such as the focal length and shooting angle of the first camera of the image acquisition device, and the attitude of the first camera may be calculated using the PNP algorithm (Perspective-n-Point algorithm). 223: Based on the relative positional relationship between the camera and the abnormal UAV, the pose of the abnormal UAV in the camera coordinate system is calculated; wherein, the relative positional relationship between the first camera and the abnormal UAV may specifically include the mounting angle of the first camera and the attitude angle of the abnormal UAV. 224: Based on the coordinate position of the return marker in the camera coordinate system, the pose of the abnormal UAV in the camera coordinate system, and the attitude of the camera, the return direction of the abnormal UAV is calculated.

[0042] Alternatively, a rotation matrix can be used to calculate the yaw angle of the abnormal UAV. Specifically, first, the pixel coordinates can be transformed to the camera coordinate system, and then the camera coordinate system can be transformed to the body coordinate system. Based on this, the yaw angle is calculated using inverse trigonometric functions, so that the abnormal UAV can fly towards the marker based on the yaw angle.

[0043] In steps 221-224 above, the return direction can be calculated solely based on the first visually acquired return image without relying on GPS signals. By combining the camera attitude with the relative pose of the abnormal UAV, the direction calculation accuracy is higher and the anti-interference is stronger.

[0044] Accordingly, in step 200 above, when it is confirmed that the first return image has not been collected, and the relative bearing is calculated based on historical positioning information to be used as the return direction, the historical positioning information is a crucial parameter, which may specifically include: The preset return point location information includes the longitude coordinates and latitude coordinates of the return point; The latest GPS positioning information recorded by the abnormal UAV before the GPS signal became abnormal includes the longitude and latitude coordinates of the abnormal UAV.

[0045] Therefore, when calculating the relative position based on the above historical positioning information, the following formula can be used:

[0046] Wherein, ∠a is the relative azimuth angle of the abnormal UAV, FD is the difference in longitude coordinates between the return point and the abnormal UAV, FA is the difference in latitude coordinates between the return point and the abnormal UAV, L1 is the longitude coordinate of the abnormal UAV, L0 is the longitude coordinate of the return point, B1 is the latitude coordinate of the abnormal UAV, and B0 is the latitude coordinate of the return point.

[0047] In summary, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment of the present invention, the return direction can be obtained based on GPS signals, or by using a first return image, or by using historical positioning information. The combination of these three methods can ensure that abnormal unmanned aerial vehicles can clearly determine the return direction when they have GPS signals or when the GPS signals are interfered with or denied, so as to carry out return flight based on the aforementioned return direction.

[0048] Accordingly, it should be noted that in the GPS-denied autonomous return-to-home method for swarm unmanned aerial vehicles (UAVs) designed in this invention, in step 300, when the UAV swarm is flying based on the return-to-home direction, it can also select a suitable flight mode to adapt to flight requirements, that is, it can also include: The camera of the abnormal UAV acquires continuous return images and obtains the number of identical feature points in the continuous return images. When the number of identical feature points is greater than a preset threshold, the optical flow is determined to be valid, and the abnormal UAV is controlled to fly at a constant speed. When the number of identical feature points is less than the preset threshold, the optical flow is determined to be invalid, and the abnormal UAV is controlled to fly in attitude mode.

[0049] It should be noted that, for ease of understanding, the aforementioned optical flow refers to: when an unmanned aerial vehicle (UAV) is flying relative to the ground, it uses an optical flow algorithm to perceive its own flight speed and relative position relative to the ground; when the UAV is returning to base, if the optical flow is deemed valid, it can effectively calculate its flight speed and relative position, and then control the UAV to fly at a constant speed; if the optical flow is deemed invalid, the UAV is controlled to fly in attitude mode, which means that the UAV only uses IMU (Inertial Measurement Unit) data for flight, which is well known in the field and will not be elaborated on here.

[0050] Accordingly, in the method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment of the present invention, in step 400, whether the unmanned aerial vehicles with abnormal GPS signals in the swarm have acquired a second return image containing landing markers can also be determined based on the landing markers using a similar process to the calculation of the return direction described above. That is, the landing direction can be obtained specifically based on the following steps 401-404: 401: The camera on the abnormal unmanned aerial vehicle acquires a second return image containing the landing marker, and based on the second return image, obtains the coordinate position of the landing marker in the camera coordinate system; 402: Based on the onboard parameters of the second camera of the image acquisition device on the abnormal unmanned aerial vehicle, the attitude of the second camera of the image acquisition device is calculated; wherein, the onboard parameters may specifically include parameters such as the focal length and shooting angle of the second camera of the image acquisition device, and the attitude of the second camera may be specifically calculated using the PNP algorithm (Perspective-n-Point algorithm). 403: Based on the relative positional relationship between the second camera of the image acquisition device and the abnormal UAV, the pose of the abnormal UAV in the coordinate system of the second camera is calculated; wherein, the relative positional relationship between the second camera and the abnormal UAV may specifically include the installation angle of the second camera and the attitude angle of the abnormal UAV. 404: Based on the coordinate position of the landing marker in the second camera coordinate system, the pose of the abnormal UAV in the second camera coordinate system, and the attitude of the second camera, the landing direction of the abnormal UAV is calculated.

[0051] Alternatively, a rotation matrix can be used to calculate the positional relationship between the anomalous UAV and the landing marker. Specifically, first, the pixel coordinates are transformed to the camera coordinate system, and then the camera coordinate system is transformed to the aircraft coordinate system. Finally, inverse trigonometric functions are used to calculate the positional relationship between the anomalous UAV and the landing marker, thereby guiding the anomalous UAV to decelerate and descend towards the landing marker.

[0052] Therefore, after obtaining the landing direction of the unmanned aerial vehicle (UAV), the UAV can fly based on the landing direction and activate its own visual obstacle avoidance system to decelerate and descend, so as to complete the return landing of the UAV and complete the landing of the entire UAV cluster. The entire landing process is highly accurate and is equipped with visual guidance and beacon guidance schemes for positioning. It can be effectively applied to various scenarios and has good applicability.

[0053] It should be noted that the autonomous return-to-home method designed in this invention solves the problems of existing infrared beacon guidance schemes, which only set up a single beacon, making it difficult for abnormal UAVs to capture signals at long distances, lacking progressive path guidance, and having low navigation recognition efficiency and inability to achieve high-precision landing. By optimizing the collaborative mechanism of visual recognition and beacon guidance, and setting up the aforementioned return-to-home and landing markers for progressive guidance, the method achieves precise guidance for abnormal UAVs from long-distance return to short-distance landing. This can effectively improve navigation recognition efficiency and landing accuracy, thereby meeting the needs of scenarios with high requirements for return-to-home and landing accuracy, such as logistics distribution and precision operations. It has good prospects for promotion and application value.

[0054] Based on this, the present invention overcomes the shortcomings of existing UAV return-to-home navigation technologies and provides a method and system for autonomous return-to-home of swarm UAVs in GPS-denied environments that can adapt to complex scenarios and ensure return-to-home accuracy and reliability. Specifically, it solves the following core technical problems: it addresses the issue that existing UAV swarm return-to-home technologies rely excessively on GPS navigation, and in GPS-denied environments where GPS signals are interfered with, blocked, or completely ineffective, they cannot achieve autonomous return-to-home and accurate landing, and are prone to equipment loss and collision accidents. The present invention achieves autonomous return-to-home navigation without relying on GPS signals, improves the adaptability and safety of UAV swarms in complex environments, and adapts to the application needs of GPS-denied scenarios such as military reconnaissance and inspection of complex urban areas. It has good prospects for promotion and application value.

[0055] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment, characterized in that, Includes the following steps: Unmanned aerial vehicle swarms perform flight missions; Each UAV in the cluster determines whether its GPS signal is normal: if so, it autonomously returns to home based on the GPS signal; if not, it checks whether the UAV with abnormal GPS signal in the cluster has collected a first return-home image containing the return-home marker. If the first return-home image has been collected, the return-home direction is obtained based on the return-home marker. When the first return image is not acquired, the relative orientation is calculated based on historical positioning information, and the relative orientation is used as the return direction. The UAVs with abnormal GPS signals in the cluster fly based on the return direction, and when the UAVs with abnormal GPS signals in the cluster acquire a second return image containing landing markers, the landing direction is determined based on the landing markers. Unmanned aerial vehicles in the cluster with abnormal GPS signals landed based on the stated landing direction.

2. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 1, characterized in that, To determine if the GPS signal is normal, the following steps are taken: The GPS signal of any UAV in the UAV cluster is detected at the first moment. If the GPS signal at the first moment is not interfered with or is not malfunctioning, the GPS signal of the UAV is determined to be normal and the UAV continues to perform its flight mission. If the GPS signal at the first moment is interfered with or is not malfunctioning, the GPS signal of the UAV is determined to be abnormal and the UAV with abnormal GPS signal is marked as an abnormal UAV.

3. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 2, characterized in that, Perform the following steps on the abnormal unmanned aerial vehicle: Control the abnormal unmanned aerial vehicle to hover in place and wait for a preset time; The GPS signal of the abnormal UAV at the second moment after waiting for the preset time is detected. If the GPS signal at the second moment is not interfered with or fails, the GPS signal of the abnormal UAV is determined to be normal, and the abnormal UAV returns autonomously based on the GPS signal. If the GPS signal of the abnormal UAV at the second moment is interfered with or fails, the GPS signal of the abnormal UAV is determined to be abnormal.

4. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 1, characterized in that, The return marker and / or the landing marker include at least one of the following: a QR code, a barcode, and a light strip with a preset shape.

5. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 4, characterized in that, The return marker and / or the landing marker are infrared light strips of a preset shape.

6. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 2, characterized in that, Whether the UAVs in the cluster with abnormal GPS signals have acquired a first return-to-home image containing return-to-home markers, and if the first return-to-home image has been acquired, the return-to-home direction is obtained based on the return-to-home markers, specifically as follows: The camera on the abnormal unmanned aerial vehicle acquires a first return image containing a return marker, and based on the first return image, the coordinate position of the return marker in the camera coordinate system is obtained; The camera's attitude is calculated based on the camera's onboard parameters on the abnormal unmanned aerial vehicle; Based on the relative positional relationship between the camera and the abnormal UAV, the pose of the abnormal UAV in the camera coordinate system is calculated; Based on the coordinates of the return marker in the camera coordinate system, the pose of the abnormal UAV in the camera coordinate system, and the attitude of the camera, the return direction of the abnormal UAV is calculated.

7. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 1, characterized in that, The historical location information includes: The preset return point location information includes the longitude coordinates and latitude coordinates of the return point; The latest GPS positioning information recorded by the abnormal UAV before the GPS signal became abnormal includes the longitude and latitude coordinates of the abnormal UAV.

8. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 7, characterized in that, The relative orientation is calculated based on historical location information, specifically as follows: Wherein, ∠a is the relative azimuth angle of the abnormal UAV, FD is the difference in longitude coordinates between the return point and the abnormal UAV, FA is the difference in latitude coordinates between the return point and the abnormal UAV, L1 is the longitude coordinate of the abnormal UAV, L0 is the longitude coordinate of the return point, B1 is the latitude coordinate of the abnormal UAV, and B0 is the latitude coordinate of the return point.

9. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 1, characterized in that, When the unmanned aerial vehicle swarm flies based on the return direction, it also includes: The camera of the abnormal UAV acquires continuous return images and obtains the number of identical feature points in the continuous return images. When the number of identical feature points is greater than a preset threshold, the optical flow is determined to be valid, and the abnormal UAV is controlled to fly at a constant speed. When the number of identical feature points is less than the preset threshold, the optical flow is determined to be invalid, and the abnormal UAV is controlled to fly in attitude mode.

10. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 1, characterized in that, Also includes: The abnormal unmanned aerial vehicle underwent layered planning before returning to base; The abnormal unmanned aerial vehicle underwent layered planning before returning to base, specifically as follows: Obtain the ID of the abnormal unmanned aerial vehicle; The layer height is determined by the upper and lower limits of the layer. Based on the abnormal UAV ID and the layer height, the layer position of each abnormal UAV is determined by the integer division method. Control the abnormal unmanned aerial vehicle to descend to the layered position.

11. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 1, characterized in that, The unmanned aerial vehicles with abnormal GPS signals in the cluster fly based on the return direction, including: the abnormal unmanned aerial vehicle matches the corresponding alternate landing site ID before returning; Before returning, the abnormal unmanned aerial vehicle was matched with the corresponding alternate landing site ID, specifically: Obtain the ID of the abnormal unmanned aerial vehicle; Get the number of alternate landing sites; Based on the abnormal UAV ID and the number of alternate landing sites, the alternate landing site ID corresponding to each abnormal UAV is determined using the integer division method.

12. The method for autonomous return of swarm unmanned aerial vehicles in a GPS-denied environment according to claim 11, characterized in that, When an unmanned aerial vehicle (UAV) with an abnormal GPS signal in the cluster acquires a second return image containing landing markers, the landing direction is determined based on the landing markers, including: When the abnormal UAV acquires a second return image containing the alternate landing site ID, the landing direction is determined based on the alternate landing site ID.

13. A swarm unmanned aerial vehicle (UAV) autonomous return system under GPS denial conditions, applied to the swarm UAV autonomous return method under GPS denial conditions as described in any one of claims 1-12, characterized in that, include: Return markers are placed along the return route of a swarm of unmanned aerial vehicles (UAVs) during a return flight mission. Multiple landing markers are placed at the return destination of the unmanned aerial vehicle swarm during its return flight mission; An unmanned aerial vehicle (UAV) swarm comprises multiple UAVs, each equipped with a GPS positioning device, an image acquisition device, and a controller. The controller is communicatively connected to the GPS positioning device and the image acquisition device. The GPS positioning device acquires GPS signals from the UAVs. The image acquisition device acquires a first return-to-home image containing a return-to-home marker and a second return-to-home image containing a landing marker matching the UAV. The GPS positioning device transmits GPS signals to the controller. The image acquisition device transmits the first and second return-to-home images to the controller, which executes the autonomous return-to-home method for the swarm of UAVs in a GPS-denied environment.

14. The autonomous return-to-home system for swarm unmanned aerial vehicles in a GPS-denied environment according to claim 13, characterized in that, The image acquisition device includes: The first camera is used to capture images in front of the unmanned aerial vehicle to capture the first return image containing the return marker; The second camera is used to capture images below the unmanned aerial vehicle to obtain a second return-to-home image that includes landing markers.

Images