WO2021146973A1 - 无人机返航的控制方法、设备、可移动平台和存储介质 - Google Patents

无人机返航的控制方法、设备、可移动平台和存储介质 Download PDF

Info

Publication number
WO2021146973A1
WO2021146973A1 PCT/CN2020/073660 CN2020073660W WO2021146973A1 WO 2021146973 A1 WO2021146973 A1 WO 2021146973A1 CN 2020073660 W CN2020073660 W CN 2020073660W WO 2021146973 A1 WO2021146973 A1 WO 2021146973A1
Authority
WO
WIPO (PCT)
Prior art keywords
drone
environment image
obstacle
distance
uav
Prior art date
Application number
PCT/CN2020/073660
Other languages
English (en)
French (fr)
Inventor
刘宝恩
李鑫超
王涛
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to CN202080004128.9A priority Critical patent/CN112639655A/zh
Priority to PCT/CN2020/073660 priority patent/WO2021146973A1/zh
Publication of WO2021146973A1 publication Critical patent/WO2021146973A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft

Definitions

  • the present invention relates to the technical field of image processing, and in particular to a control method, equipment, movable platform and storage medium for the return of an unmanned aerial vehicle.
  • UAV is an unmanned aerial vehicle operated by radio remote control equipment and self-provided program control device. Compared with manned aircraft, it has the characteristics of small size and low cost. It has been widely used in many fields, such as street scene shooting, power inspection, traffic monitoring, post-disaster rescue and so on.
  • the UAV In all stages of UAV flight, obstacles need to be avoided. Especially the return phase of the drone. In the process of returning home, the UAV will usually travel a certain distance in the homeward direction, such as flying upward for a while, and then return home. Take the upward flight as an example. During the upward flight, it is necessary to distinguish whether there are obstacles above the drone. When there is an obstacle, it is necessary to further detect whether the distance between the obstacle and the UAV will affect the ascent of the UAV.
  • the invention provides a control method, equipment, movable platform and storage medium for the return of the drone, which are used to control the return of the drone safely.
  • the first aspect of the present invention is to provide a method for controlling the return of the drone, the method including:
  • the second aspect of the present invention is to provide a movable platform, the movable platform includes: a body, a power system and a control device;
  • the power system is arranged on the body and used to provide power for the movable platform
  • the control device includes a memory and a processor
  • the memory is used to store a computer program
  • the processor is configured to run a computer program stored in the memory to realize:
  • the third aspect of the present invention is to provide a control device for the return of the drone, the device including:
  • Memory used to store computer programs
  • the processor is configured to run a computer program stored in the memory to realize:
  • the fourth aspect of the present invention is to provide a computer-readable storage medium, the storage medium is a computer-readable storage medium, the computer-readable storage medium stores program instructions, and the program instructions are used in the first aspect.
  • the control method, equipment, movable platform and storage medium for the return of the drone acquire the first environment image in the return direction of the drone, and detect whether there are obstacles in the first environment image. If the first environment image contains an obstacle and the obstacle does not affect the normal flight of the drone, then respond to the first flight control instruction to make the drone fly in the first preset mode.
  • the UAV can obtain a second environment image during the flight in the first preset flight mode, and determine the distance between the obstacle and the UAV according to the second environment image. Among them, the first environment image and the second environment image are taken at different locations, but both correspond to the return direction of the drone. If the distance is greater than the preset distance, it indicates that the obstacle is far away from the UAV and will not affect the return home, and then respond to the return home instruction to make the UAV return home.
  • the method provided by the present invention detects whether there is an obstacle in the return direction of the drone. It can initially ensure the safety of the drone's return home. At the same time, when there is an obstacle in the first environment image, the distance between the obstacle and the drone will be further calculated according to the environment image, and the distance can be used to determine whether the drone can return. Since this distance can indicate the precise location of the obstacle, that is, it reflects whether the obstacle is completely in the return direction, so it can further ensure the safety of the return of the UAV.
  • FIG. 1 is a schematic flowchart of a method for controlling the return of a drone according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a pan/tilt configured for drones in different states according to an embodiment of the present invention
  • FIG. 3 is a flowchart of an optional distance measurement method between an obstacle and a UAV according to an embodiment of the present invention
  • FIG. 4 is a schematic diagram of the positional relationship between various parameters in a distance measurement method provided by an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a circular field of view corresponding to an environmental image provided by an embodiment of the present invention.
  • FIG. 6 is a schematic flowchart of another method for controlling the return of a drone according to an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a control device for returning a drone according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a movable platform provided by an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a control device for returning a drone according to an embodiment of the present invention.
  • the automatic return mechanism of the drone may be briefly introduced.
  • the UAV is often in the range of beyond the visual range.
  • the UAV completes the flight mission or encounters the harsh natural environment during the flight, such as the protruding mountain peak, or the ground base station.
  • the communication connection is disconnected, in order to ensure the safety of the drone and avoid damage accidents, the drone often needs to return home automatically. Because there is a stage of flying along the return direction during the automatic return of the UAV. Therefore, determining whether there are obstacles in the return direction of the UAV and whether the location of the obstacles really affects the return of the aircraft has become an important factor in whether the UAV can return home automatically.
  • the control method for the return of the drone provided in the following embodiments can be used to control the return of the drone.
  • the above-mentioned return direction can usually be directly above or diagonally above the UAV.
  • the return direction can also be down or any other direction, and the present invention does not limit the return direction.
  • FIG. 1 is a schematic flowchart of a method for controlling the return of a drone according to an embodiment of the present invention.
  • the main body of the control method for the return of the UAV is the control device.
  • the control device can be implemented as software or a combination of software and hardware.
  • the control device executes the control method of the UAV to return to home, it can realize the return of the UAV.
  • the control device in this embodiment and the following embodiments may specifically be a movable platform, such as a drone.
  • the method may include:
  • S101 Acquire a first environment image in the return direction of the drone.
  • the UAV During the flight of the UAV, its own camera can take pictures of the environment in the flying environment of the UAV.
  • the drone hovering at its current position its own camera can capture the first environment image in the return direction of the drone.
  • the field of view corresponding to the first environmental image may be the field of view of the monocular camera itself.
  • the return direction of the UAV is usually above the UAV.
  • the camera configured by the UAV can also be a monocular camera capable of looking upwards.
  • the upward-looking function of this monocular camera can use the cloud
  • the platform is realized, that is, the monocular camera is placed on a platform that can be lifted upward, and the platform is lifted upward so that the monocular camera can capture the first environment image corresponding to the upper part of the drone.
  • the lifted and non-lifted states of the PTZ can be shown in Figure 2.
  • the drone will recognize the acquired first environment image to determine whether the image contains obstacles and whether the obstacles affect the normal flight of the drone. If the first environment image contains obstacles and the obstacles do not affect the normal flight of the drone, the drone will respond to the first flight control command so that the drone can fly in the first preset mode.
  • the first flight control instruction may be autonomously generated by the UAV, or may be sent to the UAV by the pilot through the control device.
  • the first preset manner may be to make the drone fly in the same direction as the return direction. Assuming that the return direction is above the drone, at this time, the first preset method is to make the drone fly from the current position where the first environment image is taken, for example, from the current position to the first position and the second position. . In practical applications, because there may be obstacles above the drone, there is a small distance difference between the current position, the first position, and the second position, usually only a few centimeters.
  • the first environment image contains obstacles, optionally, it can be done with the aid of a neural network model.
  • the neural network model may be a convolutional neural network (Convolutional Neural Networks, CNN) model.
  • the neural network model can include multiple computing nodes. Each computing node can include a convolution (Conv) layer, batch normalization (BN), and an activation function ReLU.
  • the computing nodes can use skip connection (Skip Connection). ) Way to connect.
  • K ⁇ H ⁇ W can be input into the neural network model, and after the neural network model is processed, the output data of C ⁇ H ⁇ W can be obtained.
  • K can represent the number of input channels, and K can be equal to 4, corresponding to the four channels of red (R, red), green (G, green), blue (B, blue) and depth (D, deep) respectively;
  • H can represent the height of the input image (that is, the first environment image),
  • W can represent the width of the input image, and C can represent the number of categories.
  • the input image when the input image is too large, an input image can be cut into N sub-images.
  • the input data can be N ⁇ K ⁇ H' ⁇ W'
  • the output data can be N ⁇ C ⁇ H' ⁇ W', where H'can represent the height of the sub-image, and W'can represent the width of the sub-image.
  • H'can represent the height of the sub-image
  • W'can represent the width of the sub-image.
  • the feature map may also be obtained in other ways, which is not limited in this application.
  • Using the above-mentioned pre-trained neural network model to process the environment image to obtain a feature map may specifically include the following steps:
  • Step 1 Input the environment image into the neural network model to obtain the model output result of the neural network model.
  • the model output result of the neural network model may include the confidence feature maps output by multiple output channels, and the multiple output channels can correspond to multiple object categories one-to-one, and the pixel values of the confidence feature maps of a single object category are used To characterize the probability that a pixel is an object category.
  • Step 2 According to the model output result of the neural network model, a feature map containing semantic information is obtained.
  • the object category corresponding to the confidence feature map with the largest pixel value at the same pixel location in the multiple confidence feature maps one-to-one corresponding to the multiple output channels may be used as the object category of the pixel location to obtain the feature map.
  • the number of output channels of the neural network model is 4, and the output result of each channel is a confidence feature map, that is, the 4 confidence feature maps are the confidence feature map 1 to the confidence feature map 4, and the confidence The degree characteristic map 1 corresponds to the sky, the confidence characteristic map 2 corresponds to buildings, the confidence characteristic map 3 corresponds to trees, and the confidence characteristic map 4 corresponds to "other". In these categories, except for the sky, the rest can be regarded as obstacles.
  • the pixel value at the pixel location (100, 100) in the confidence feature map 1 is 70
  • the pixel value at the pixel location (100, 100) in the confidence feature map 2 is 50
  • the pixel at the pixel location (100, 100) in the confidence feature map 3 When the value is 20, and the pixel value of the pixel position (100, 100) in the confidence feature map 4 is 20, it can be determined that the pixel position (100, 100) is the sky.
  • the pixel value at the pixel location (100, 80) in the confidence feature map 1 is 20
  • the pixel value at the pixel location (100, 80) in the confidence feature map 2 is 30, and the pixel location in the confidence feature map 3
  • the pixel value of (100,80) is 20
  • the pixel value of pixel position (100,80) in the confidence feature figure 4 is 70
  • the above-mentioned recognition is actually at the pixel level, that is, to identify the category to which each pixel in the first environment image belongs, that is, to identify the category to which each object in the first environment image belongs, and it is also indirect. Determine the position of each object in the first environment image.
  • the UAV responds to the return-to-home instruction and simply returns to the home. If it is recognized that there is an obstacle in the first environment image, it can be further determined whether the obstacle affects the normal flight of the UAV.
  • an optional way to understand whether the obstacle affects the normal flight of the drone can be:
  • the volume of the obstacle in the first environment image is large, it can be considered This large-volume obstacle will cover most of the area above the drone. At this time, it can be considered that this large-volume obstacle will affect the normal flight of the drone, so as to directly control the hover of the drone.
  • the small obstacle in the first environment image When the obstacle in the first environment image is small, it can be considered that the small obstacle will not cover most of the area above the drone, but it may be located directly above the drone. In this case, no one The aircraft may also collide with it during its ascent and return. Therefore, for small obstacles that do not affect the normal flight of the drone, the drone needs to perform subsequent judgment steps to determine whether the obstacle will really affect the return of the drone.
  • the size information of the obstacle in the first environment image can be determined first, and the actual size information of the obstacle can be estimated based on the size information . If the actual size information is greater than the preset size, it is determined that the obstacle affects the normal flight of the UAV. Otherwise, determine that this obstacle does not affect the normal flight of the UAV.
  • S103 Determine the distance between the obstacle and the drone according to the second environment image obtained during the flight of the drone in the first preset mode, where the second environment image corresponds to the return direction of the drone.
  • the UAV can continuously capture the second environment image during the flight in the first preset mode, that is, the process of rising continuously.
  • the field of view corresponding to the second environment image may be the field of view of the monocular camera itself, and the shooting position of each second environment image is different.
  • the UAV can calculate the distance between the obstacle and the UAV based on the second environment images taken at different locations. This distance may specifically include the horizontal distance and the vertical distance between the obstacle and the drone.
  • This distance may specifically include the horizontal distance and the vertical distance between the obstacle and the drone.
  • the calculated horizontal distance between the UAV and the obstacle is greater than the preset distance, then you can respond to the return command and directly control the UAV to return. If the horizontal distance is less than or equal to the preset distance, it indicates that the distance between the drone and the obstacle is small. If it returns home, the drone will collide with the obstacle. Therefore, you can respond to the hovering command at this time. Hover the drone.
  • the first environment image in the return direction of the drone is acquired, and whether there is an obstacle in the first environment image is detected.
  • This detection process can initially ensure the safety of the drone's return home. Then, if the first environment image contains obstacles and the obstacles do not affect the normal flight of the drone, the drone will be controlled to fly in the first preset mode, and the image will be taken during the flight in the first preset mode.
  • the distance between the obstacle and the drone is determined based on the second environment image. If the distance is greater than the preset distance, it indicates that the obstacle is far away from the drone, and it will not affect the return home, so the drone will respond to the return home instruction to make the drone return home. Since this distance can indicate the precise location of the obstacle, that is, it reflects whether the obstacle is completely in the return direction, so it can further ensure the safety of the return of the UAV.
  • the drone needs to accurately determine whether the obstacle above itself affects normal flight, in order to further accurately calculate the distance between the obstacle and the drone to control the drone to return home.
  • the volume of the obstacle cannot accurately reflect whether the obstacle affects the normal flight. Therefore, the method provided in the embodiment shown in FIG. 1 is used to determine whether the obstacle affects the normal flight. The result obtained is obviously not accurate enough. Therefore, there is another more accurate way to understand whether obstacles affect the normal flight of drones:
  • the obstacle in the first environment image is located directly above the drone, it indicates that it will affect the normal flight of the drone. At this point, you can directly control the drone to continue hovering. If the obstacle is not directly above the UAV, it indicates that the obstacle does not affect the normal flight of the UAV. However, because the obstacle may be located diagonally above the UAV, the UAV may also collide with it during the ascent and return. Therefore, at this time, the UAV needs to continue to perform the next steps to determine the obstacle. Will it really affect the return of the drone?
  • another way to optionally determine whether an obstacle affects the normal flight of the drone can be: After obtaining the first environment image, it can determine the category of the object in the image and the objects of different categories in the first environment The position in the image. Among them, the recognition of the object category and the object position can be implemented in the manner provided in step 102 above. At the same time, a preset area can be divided in the first environment image, and this area can be considered as the minimum safe flight area of the drone, and then it is judged whether the object whose object category is an obstacle is in the preset area.
  • the drone responds to the hovering command and the drone is in position. Hovering state. If the obstacle is located in the non-predetermined area, it is determined that the obstacle does not affect the normal flight of the UAV. At this time, the distance between the determined obstacle and the drone can be further determined to determine whether the drone can return home, that is, continue to perform step 103 to step 104 in the embodiment shown in FIG. 1.
  • step 101 of the embodiment shown in FIG. 1 this method is realized by the ascending flight of the drone. At this time, because there are obstacles above the drone, it is more prone to collision between the drone and the obstacle during the ascending flight to obtain the second environment image, which will cause the drone to be damaged.
  • the first flight control command can usually be to control the drone to fly from the first position to the second position in a direction opposite to the return direction.
  • the first flight control command is to make the drone fly from the first position to the second position, and the two positions are separated by a preset distance.
  • step 103 the method for measuring the distance between the obstacle and the UAV, that is, an optional implementation of step 103 can be as shown in Fig. 3:
  • S1032 Determine the distance between the obstacle and the drone according to the second environment image corresponding to the first position and the second position.
  • the drone After the drone responds to the first flight control command, the drone can descend and fly from the current position where the first environmental image is taken to the first position, where the monocular configured on the drone that can look upwards
  • the camera captures an image of the second environment. Then, descend and fly to the second position, and take another second environmental image at the second position, so as to determine the distance between the obstacle and the drone based on the second environmental image taken at different positions.
  • the sensors configured by the drone itself can also be used to determine whether the drone can descend and fly from the first position to the second position.
  • the first posture information of the pan/tilt mounted with the monocular camera is determined when the second environment image is taken; then according to the monocular camera in the second position
  • the second environment image captured by the location determines the second posture information of the pan/tilt when the second environment image is captured.
  • point A represents an obstacle
  • point O represents the center of the drone's fuselage
  • point O 1 represents the first position
  • point O 2 represents the second position.
  • d is the horizontal distance between the obstacle and the drone
  • l 1 is the vertical distance between the obstacle and the drone
  • l 2 is the preset between the first position O 1 and the second position O 2
  • tan ⁇ 1 is the first attitude angle of the gimbal when the UAV is in the first position O 1
  • tan ⁇ 2 is the second attitude angle of the gimbal when the UAV is in the second position O 2.
  • the drone uses the ascending flight in the embodiment shown in FIG. 1 to capture the second environment image, that is, when the drone rises from the position O 2 shown in FIG. 4 to the position O 1 , use the above
  • the horizontal distance d between the obstacle and the drone and the distance l 1 between the position O 1 and the drone can be obtained.
  • the sum of the distance l 1 and the preset distance l 2 between the position O 2 and the position O 1 is calculated, and the sum of this distance is also the vertical distance between the obstacle and the UAV.
  • the second environment images taken at different positions are used to simulate the images taken by the binocular camera, and the binocular distance measurement principle is used to accurately calculate the horizontal distance between the obstacle and the drone, so that According to this horizontal distance, it can be accurately determined whether the UAV can return home, so as to prevent the UAV from being damaged in the process of returning home.
  • the field of view of the first environment image corresponding to the return direction obtained by the drone is also the field of view of the monocular camera. This field of view is small and cannot fully display the drone.
  • the drone can also be controlled to fly in a special way, so as to obtain a first environment image with a larger viewing angle, so as to more comprehensively determine whether there are obstacles in the return direction.
  • the drone can respond to the second flight control command to make itself fly in the second preset mode.
  • the monocular camera configured on the drone can capture the first environment image in the preset field of view in the return direction of the drone.
  • An optional second preset manner may be: the drone responds to the rotation flight control command, so that it rotates and flies once at the current position.
  • the particularity of the flying mode of the UAV will directly lead to the particularity of the field of view corresponding to the first environmental image obtained. That is to say, because the monocular camera on the UAV itself has a certain field of view angle, after rotating and flying, the field of view corresponding to the first environment image obtained is a circular field of view, as shown in Figure 5. And the height of the ring is determined by the field of view of the monocular camera.
  • Another optional second preset manner may be: the drone responds to the rotation flight control command, so that the drone rotates and flies once at the current position. After rotating and flying, a first environment image can be obtained. Then, the drone can also respond to the straight-line flight control command, so that the drone can fly horizontally to a third position that is a preset distance from the current position. Respond to the rotating flight control command to make the UAV rotate and fly one circle in the third position. After rotating and flying in the third position, another first environment image can be obtained.
  • the drone will perform two rotating flights, so that multiple first environmental images can be obtained, and each environmental image corresponds to a circular field of view as shown in FIG. 5.
  • the UAV can respectively identify whether there are obstacles in multiple first environmental images and whether the obstacles affect the normal flight of unmanned people, and comprehensively determine whether the obstacles affect the normal flight according to the recognition results of each first environmental image. Normal flight of man and machine.
  • an environmental image with a circular field of view is obtained.
  • This environmental image can more completely show the distribution of obstacles in the return direction of the UAV, thereby making it more comprehensive Determine whether there are obstacles in the return direction.
  • the image with the circular field of view is used instead of the image with the full field of view. This can also reduce the amount of calculation in the obstacle detection process and improve Detection efficiency.
  • the control method for the drone to return home further includes the following step:
  • S201 Acquire a third environment image in the return direction of the drone taken by a monocular camera configured by the drone, and the field of view corresponding to the third environment image is the field of view of the monocular camera.
  • the monocular camera can first capture the third environmental image of the drone in the return direction.
  • the field of view corresponding to this third environmental image is the same as the field of view of the monocular camera, and the shooting position of the third environmental image can be no The current position of the HMI. Then, the type to which the object contained in the third environment image belongs is recognized. For the identification process of the object category, refer to the related description in step 102 in FIG. 1.
  • the drone can be further controlled to fly in the second preset mode, and the result is shown in Figure 5.
  • the first environmental image of the circular field of view The environmental image of this circular field of view can fully display the distribution of obstacles in the return direction of the drone, and can more accurately determine whether there are obstacles in the return direction, and further According to the methods provided in the foregoing embodiments, it is finally determined whether the drone can return home.
  • This embodiment is combined with the embodiments shown in Figure 1 and Figure 5 for understanding: first obtain a third environment image with a smaller field of view, if the third environment does not contain obstacles, you can directly control the drone to return home .
  • this method requires less calculation and determines the efficiency of returning home. Also higher.
  • the third environment image contains obstacles, it can be further determined whether the obstacles in the first environment image with the circular field of view affect the normal flight of the drone, and then the steps of the embodiment shown in FIG. 1 are further executed. So it is judged that the drone can return home.
  • using the first environment image with a large field of view can more accurately determine that the drone can return home.
  • the unmanned return when it is determined that there is no obstacle in the return direction of the UAV, the unmanned return can be controlled. It is easy to understand that any flight process of the drone requires battery power. Therefore, before controlling the drone to return home, you can also determine the power required during the return process. If the current remaining power is more than the return required When the battery is charged, the drone will be controlled to return home.
  • an alternative way is to first estimate the wind speed information from the current position to the destination of the return home based on the historical wind speed information. Then determine the ground speed information for landing from the current position to the return destination, so as to determine the power required for the drone's return process based on the wind speed information and ground speed information.
  • FIG. 7 is a schematic structural diagram of a control device for the return of a drone provided by an embodiment of the present invention; referring to FIG. 7, this embodiment provides a control device for the return of a drone.
  • the control device can execute the aforementioned control method for the return of the drone; specifically, the control device for the return of the drone includes:
  • the acquiring module 11 is used to acquire the first environment image in the return direction of the drone.
  • the response module 12 is configured to respond to a first flight control instruction if the obstacles in the first environment image do not affect the normal flight of the drone, so that the drone can fly in a first preset manner .
  • the determining module 13 is configured to determine the distance between the obstacle and the drone according to the second environment image obtained during the flight of the drone in the first preset mode, and the second The environment image corresponds to the return direction of the drone.
  • the response module 12 is also configured to respond to the return-to-home instruction if the distance is greater than the preset distance, so that the UAV can return to home
  • the device shown in FIG. 7 can also execute the methods of the embodiments shown in FIGS. 1 to 6.
  • FIGS. 1 to 6 For parts that are not described in detail in this embodiment, reference may be made to the related descriptions of the embodiments shown in FIGS. 1 to 6.
  • the implementation process and technical effects of this technical solution please refer to the description in the embodiment shown in FIG. 1 to FIG. 6, which will not be repeated here.
  • FIG. 8 is a schematic structural diagram of a movable platform provided by an embodiment of the present invention. referring to FIG. 8, an embodiment of the present invention provides a movable platform, and the movable platform is at least one of the following: Aircraft, unmanned ships, unmanned vehicles; specifically, the movable platform includes: a body 21, a power system 22, and a control device 23.
  • the power system 22 is arranged on the body 21 and is used to provide power for the movable platform.
  • the control device 23 includes a memory 231 and a processor 232.
  • the memory is used to store a computer program
  • the processor is configured to run a computer program stored in the memory to realize:
  • the processor 232 is further configured to: if the distance is less than or equal to the preset distance, respond to a hovering instruction to make the drone hover.
  • the first flight control instruction is to control the drone to fly in a direction opposite to the return direction from a first position to a second position that is a preset distance away;
  • the processor 232 is further configured to: when the drone is in the first position and the second position, respectively, acquiring the second environment image captured by a monocular camera configured by the drone;
  • the distance between the obstacle and the drone is determined according to a second environment image corresponding to each of the first position and the second position.
  • the processor 232 is further configured to: according to the second environmental image captured by the monocular camera at the first position, determine that when the second environmental image is captured, the pan/tilt head mounted with the monocular camera First posture information;
  • the movable platform further includes a monocular camera 24 which is arranged on the body 21.
  • the processor 232 is further configured to: respond to a second flight control instruction to make the UAV fly in a second preset mode;
  • the first environment image in the preset field of view in the return direction of the drone, taken by the monocular camera configured by the drone, is acquired.
  • the preset field of view corresponding to the first environment image is a circular field of view.
  • the processor 232 is further configured to: respond to the rotation flight control instruction, so that the UAV rotates and flies once in place at the current position.
  • the processor 232 is further configured to: respond to a straight-line flight control instruction to make the UAV fly horizontally to a third position that is a preset distance from the current position;
  • the UAV In response to the rotation flight control command, the UAV is made to rotate and fly one circle in situ at the third position.
  • the processor 232 is further configured to: determine the category to which the object in the first environment image belongs and the positions of the objects of different categories in the first environment image;
  • the object belonging to the obstacle category is located in the non-predetermined area of the first environment image, it is determined that the obstacle in the first environment image does not affect the normal flight of the drone.
  • the processor 232 is further configured to: if the object belonging to the obstacle category is located in the preset area of the first environment image, respond to the hovering instruction to make the drone hover.
  • the processor 232 is further configured to: acquire a third environment image in the return direction of the drone captured by the monocular camera configured with the drone, and the third environment image corresponds to The field of view is the field of view of the monocular camera;
  • the monocular camera shoots in the return direction of the UAV The first environment image in the preset field of view.
  • the processor 232 is further configured to: if there is no object belonging to the obstacle category in the third environment image, respond to the return-to-home instruction to make the UAV return home.
  • the movable platform shown in FIG. 8 can execute the methods of the embodiments shown in FIGS. 1 to 6.
  • parts that are not described in detail in this embodiment please refer to the related descriptions of the embodiments shown in FIGS. 1 to 6.
  • the implementation process and technical effects of this technical solution please refer to the description in the embodiment shown in FIG. 1 to FIG. 6, which will not be repeated here.
  • the structure of the control device for the return of the drone shown in FIG. 9 can be implemented as an electronic device, and the electronic device can be a drone.
  • the electronic device may include: one or more processors 31 and one or more memories 32.
  • the memory 32 is used to store a program that supports the electronic device to execute the control method for returning the drone provided in the embodiments shown in FIGS. 1 to 6 above.
  • the processor 31 is configured to execute a program stored in the memory 32.
  • the program includes one or more computer instructions, and the following steps can be implemented when one or more computer instructions are executed by the processor 31:
  • the structure of the airspace detection device may further include a communication interface 33 for the electronic device to communicate with other devices or a communication network.
  • the processor 31 is further configured to: if the distance is less than or equal to the preset distance, respond to a hovering instruction to cause the drone to hover.
  • the first flight control instruction is to control the drone to fly in a direction opposite to the return direction from a first position to a second position that is a preset distance away;
  • the processor 31 is further configured to: when the drone is in the first position and the second position, respectively, acquiring the second environment image captured by a monocular camera configured by the drone;
  • the distance between the obstacle and the drone is determined according to a second environment image corresponding to each of the first position and the second position.
  • the processor 31 is further configured to: according to the second environmental image captured by the monocular camera at the first position, determine that when the second environmental image is captured, the pan/tilt head mounted with the monocular camera First posture information;
  • the processor 31 is further configured to: respond to a second flight control instruction to make the UAV fly in a second preset mode;
  • the first environment image in the preset field of view in the return direction of the drone, taken by the monocular camera configured by the drone, is acquired.
  • the preset field of view corresponding to the first environment image is a circular field of view
  • the processor 31 is further configured to: respond to the rotation flight control instruction, so that the UAV rotates and flies once in situ at the current position.
  • the processor 31 is further configured to: respond to a straight-line flight control instruction to make the UAV fly horizontally to a third position that is a preset distance from the current position;
  • the UAV In response to the rotation flight control command, the UAV is made to rotate and fly one circle in situ at the third position.
  • the processor 31 is further configured to: determine the category to which the object in the first environment image belongs and the positions of objects of different categories in the first environment image;
  • the object belonging to the obstacle category is located in the non-predetermined area of the first environment image, it is determined that the obstacle in the first environment image does not affect the normal flight of the drone.
  • the processor 31 is further configured to: if the object belonging to the obstacle category is located in the preset area of the first environment image, respond to the hovering instruction to make the drone hover.
  • the processor 31 is further configured to: acquire a third environment image in the return direction of the drone captured by the monocular camera configured with the drone, and the third environment image corresponds to The field of view is the field of view of the monocular camera;
  • the monocular camera shoots in the return direction of the UAV The first environment image in the preset field of view.
  • the processor 31 is further configured to: if there is no object belonging to the obstacle category in the third environment image, respond to the return-to-home instruction to make the UAV return home.
  • the device shown in FIG. 9 can execute the methods of the embodiments shown in FIGS. 1 to 6.
  • parts that are not described in detail in this embodiment reference may be made to the related descriptions of the embodiments shown in FIGS. 1 to 6.
  • the implementation process and technical effects of this technical solution please refer to the description in the embodiment shown in FIG. 1 to FIG. 6, which will not be repeated here.
  • an embodiment of the present invention provides a computer-readable storage medium.
  • the storage medium is a computer-readable storage medium.
  • the computer-readable storage medium stores program instructions. The control method of man-machine returning home.
  • the related detection device for example: IMU
  • the embodiments of the remote control device described above are merely illustrative.
  • the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components. It can be combined or integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, remote control devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present invention essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer processor (processor) execute all or part of the steps of the method described in each embodiment of the present invention.
  • the aforementioned storage media include: U disk, mobile hard disk, Read-Only Memory (ROM), Random Access Memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

一种无人机返航的控制方法、装置、可移动平台和存储介质。方法包括：获取无人机返航方向上的第一环境图像（S101）；若第一环境图像中包含不影响无人机正常飞行的障碍物，则对第一飞行控制指令进行响应，使无人机按照第一预设方式飞行（S102）；根据无人机以第一预设方式飞行过程中得到的第二环境图像，确定障碍物与无人机之间的距离（S103）；若距离大于预设距离，则对返航指令进行响应，以使无人机返航（S104）。可见，本方法中存在对无人机返航方向上是否存在障碍物的检测过程，通过此检测初步保证无人机的返航安全。在此基础上，还会根据环境图像计算障碍物与无人机之间的距离，通过距离大小来确定无人机能否返航，以进一步保证无人机的返航安全。

Description

无人机返航的控制方法、设备、可移动平台和存储介质 技术领域
本发明涉及图像处理技术领域，尤其涉及一种无人机返航的控制方法、设备、可移动平台和存储介质。
背景技术
无人机是利用无线电遥控设备和自备的程序控制装置操纵的不载人飞行器。与载人飞机相比，它具有体积小、造价低等特点，目前已经广泛使用到众多领域中，比如街景拍摄、电力巡检、交通监视、灾后救援等等。
在无人机飞行的各个阶段，都需要躲避障碍物。特别是无人机的返航阶段。在返航过程中，无人机通常会先沿返航方向一段距离，比如向上飞行一段，再进行返航。以向上飞行为例说明，在向上飞行的过程中，就需要区分出无人机上方是否存在障碍物。当存在障碍物时，还需要进一步检测出障碍物与无人机之间的距离是否会对无人机的上升飞行产生影响。
在现有技术中，通常会默认无人机上方区域不存在障碍物，但这种方式显然是不合理的，在返航过程中为无人机带来了较大的损毁风险。
发明内容
本发明提供了一种无人机返航的控制方法、设备、可移动平台和存储介质，用于控制无人机安全返航。
本发明的第一方面是为了提供一种无人机返航的控制方法，所述方法包括：
获取无人机返航方向上的第一环境图像；
若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
本发明的第二方面是为了提供一种可移动平台，所述可移动平台包括：机体、动力系统以及控制装置；
所述动力系统，设置于所述机体上，用于为所述可移动平台提供动力；
所述控制装置包含存储器和处理器；
所述存储器，用于存储计算机程序；
所述处理器，用于运行所述存储器中存储的计算机程序以实现：
获取无人机返航方向上的第一环境图像；
若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
本发明的第三方面是为了提供一种无人机返航的控制设备，所述设备包括：
存储器，用于存储计算机程序；
处理器，用于运行所述存储器中存储的计算机程序以实现：
获取无人机返航方向上的第一环境图像；
若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
本发明的第四方面是为了提供一种计算机可读存储介质，所述存储介质为计算机可读存储介质，该计算机可读存储介质中存储有程序指令，所述程序指令用于第一方面所述的无人机返航的控制方法。
本发明提供的无人机返航的控制方法、设备、可移动平台和存储介质，获取无人机返航方向上的第一环境图像，并对第一环境图像中是否存在障碍物进行检测。若第一环境图像中包含障碍物且此障碍物不影响无人机正常飞 行，则对第一飞行控制指令进行响应，使无人机按照第一预设方式飞行。无人机在以第一预设飞行方式飞行的过程中可以得到第二环境图像，并根据此第二环境图像确定障碍物与无人机之间的距离。其中，第一环境图像和第二环境图像拍摄的不同的位置，但都对应于无人机的返航方向。若距离大于预设距离，表明障碍物距离无人机较远，不会对返航造成影响，则对返航指令进行响应，以使无人机返航。
根据上述描述可知，相比于现有技术中默认无人机机返航方向上没有障碍物的方式，本发明提供的方法会对无人机返航方向上是否存在障碍物进行检测，此检测处理就能够初步保证无人机的返航安全。同时，当第一环境图像中存在障碍物时，还会进一步根据环境图像计算障碍物与无人机之间的距离，以通过距离的大小来确定无人机能否返航。由于此距离能够表明障碍物的精确位置，也即是反映出障碍物是否完全位于返航方向上，因此，也就能够进一步保证无人机的返航安全。
附图说明
此处所说明的附图用来提供对本申请的进一步理解，构成本申请的一部分，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。在附图中：
图1为本发明实施例提供的一种无人机返航的控制方法的流程示意图；
图2为本发明实施例提供的无人机配置的云台在不同状态下的结构示意图；
图3为本发明实施例提供的一种可选地障碍物与无人机之间距离测量方式的流程图；
图4为本发明实施例提供的距离测量方式中各参数之间的位置关系示意图；
图5为本发明实施例提供的环境图像对应的环形视场的示意图；
图6为本发明实施例提供的另一种无人机返航的控制方法的流程示意图；
图7为本发明实施例提供的一种无人机返航的控制装置的结构示意图；
图8为本发明实施例提供的一种可移动平台的结构示意图；
图9为本发明实施例提供的一种无人机返航的控制设备的结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。
在对本发明实施例提供的无人机返航的控制方法进行详细介绍之前，可以先对无人机的自动返航机制进行简单介绍。
在飞行的过程中，无人机很多时候都处于超视距的范围，当无人机完成飞行任务或者在飞行过程中遇到恶劣的自然环境比如突起的山峰，又或者与地面基站之间的通信连接断开时，为了保证无人机的安全，避免出现损毁事故，无人机往往需要自动返航。由于无人机的自动返航过程中存在一个沿返航方向飞行的阶段。因此，确定无人机在返航方向上是否存在障碍物，以及障碍物所在的位置是否真正对飞机返航造成影响就成为无人机能否自动返航的重要因素。此时，便可以使用下述各实施例提供的无人机返航的控制方法来对无人机的返航进行控制。
其中，需要说明的有，上述的返航方向通常可以为无人机的正上方或者斜上方。随着无人机应用场景的不断扩展，返航方向也可以是下方或者是任何其他方向，本发明并不对返航方向进行限定。
下面结合附图，对本发明的一些实施方式作详细说明。在各实施例之间不冲突的情况下，下述的实施例及实施例中的特征可以相互组合。
图1为本发明实施例提供的一种无人机返航的控制方法的流程示意图。该无人机返航的控制方法的执行主体是控制设备。可以理解的是，该控制设备可以实现为软件、或者软件和硬件的组合。控制设备执行该无人机返航的控制方法则可以实现对无人机返航的。本实施例以及下述各实施例中的控制设备具体来说可以是可移动平台，比如无人机等。具体的，该方法可以包括：
S101，获取无人机返航方向上的第一环境图像。
无人机在飞行过程中，其自身配置的摄像头可以拍得无人机所处飞行环境内的环境图像。当启动自动返航时，无人机悬停于当前位置，其自身配置的摄像头便可以拍摄到无人机返航方向上的第一环境图像。可选地，此第一环境图像对应的视场可以为单目摄像头自身具有的视场。
可选地，无人机的返航方向通常为无人机的上方，相应的，无人机配置的摄像头也可以是能够上视的单目摄像头，此单目摄像头的可上视功能可以借助云台实现，即单目摄像头放置于可以向上抬起的云台上，通过云台的向上抬起从而使得单目摄像头能够拍得对应于无人机上方的第一环境图像。其中，云台的抬起与非抬起状态可以如图2所示。
S102，若第一环境图像中的障碍物不影响无人机正常飞行，则对第一飞行控制指令进行响应，使无人机以第一预设方式飞行。
然后，无人机会对获取到的第一环境图像进行识别，以确定出图像中是否包含障碍物以及障碍物是否影响无人机的正常飞行。若第一环境图像中包含障碍物且障碍物不影响无人机的正常飞行，则无人机会对第一飞行控制指令进行响应，以使无人机按照第一预设方式飞行。
可选地，第一飞行控制指令可以由无人机自主产生，也可以由飞手通过控制设备发送至无人机。可选地，第一预设方式可以为使无人机沿与返航方向相同的方向飞行。假设返航方向为无人机上方，此时，第一预设方式即为使无人机由拍摄第一环境图像的当前位置上升飞行，比如由当前位置依次上升飞行至第一位置、第二位置。在实际应用中，由于无人机上方可能存在障碍物，因此，当前位置、第一位置以及第二位置之间均具有较小的距离差，通常只为几厘米。
对于识别第一环境图像中是否包含障碍物，可选地，可以借助神经网络模型来完成。
具体来说，神经网络模型具体可以为卷积神经网络(Convolutional Neural Networks，CNN)模型。神经网络模型可以包括多个计算节点，每个计算节点中可以包括卷积(Conv)层、批量归一化(Batch Normalization，BN)以及激活函数ReLU，计算节点之间可以采用跳跃连接(Skip Connection)方式连接。
K×H×W的输入数据可以输入神经网络模型，经过神经网络模型处理后， 可以获得C×H×W的输出数据。其中，K可以表示输入通道的个数，K可以等于4，分别对应红(R，red)、绿(G，green)、蓝(B，blue)和深度(D，deep)共四个通道；H可以表示输入图像(即第一环境图像)的高，W可以表示输入图像的宽，C可以表示类别数。
需要说明的是，当输入图像过大时，可以将一个输入图像切割为N个子图像，相应的，输入数据可以为N×K×H’×W’，输出数据可以为N×C×H’×W’，其中，H’可以表示子图像的高，W’可以表示子图像的宽。当然，在其他实施例中，也可以通过其他方式获得特征图，本申请对此不做限定。
利用上述预先训练好的神经网络模型处理环境图像，以得到特征图，具体来说可以包括如下步骤：
步骤1，将环境图像输入神经网络模型，得到神经网络模型的模型输出结果。
其中，神经网络模型的模型输出结果可以包括多个输出通道分别输出的置信度特征图，该多个输出通道可以与多个对象类别一一对应，单个对象类别的置信度特征图的像素值用于表征像素是对象类别的概率。
步骤2，根据神经网络模型的模型输出结果，得到包含语义信息的特征图。
可以将与该多个输出通道一一对应的多个置信度特征图中同一像素位置像素值最大的置信度特征图对应的对象类别，作为像素位置的对象类别，从而得到特征图。
假设，神经网络模型的输出通道的个数为4，每个通道的输出结果是一个置信度特征图，即4个置信度特征图分别为置信度特征图1至置信度特征图4，且置信度特征图1对应天空、置信度特征图2对应建筑物、置信度特征图3对应树木、置信度特征图4对应“其他”。在这几种分类中，除了天空，剩余都可以认为是障碍物。
例如，当置信度特征图1中像素位置(100,100)的像素值是70，置信度特征图2中像素位置(100,100)的像素值是50，置信度特征图3中像素位置(100,100)的像素值是20，置信度特征图4中像素位置(100,100)的像素值是20时，可以确定像素位置(100,100)是天空。
又例如，当置信度特征图1中像素位置(100,80)的像素值是20，置信度特征图2中像素位置(100,80)的像素值是30，置信度特征图3中像素位置(100,80)的像素值是20，置信度特征图4中像素位置(100,80)的像素值是 70时，可以确定像素位置(100,80)是其他，即不是树木、建筑物和树木中的任意一种。
可见，上述这种识别实际上是像素级别的，也即是识别出第一环境图像中每个像素点所属的类别，也即是识别出第一环境图像中各物体所属的类别，同时也是间接确定出各物体在第一环境图像中的位置。
若识别出第一环境图像中没有障碍物，则无人机对返航指令进行响应，直接返航即可。若识别出第一环境图像中存在障碍物，则还可以进一步确定此障碍物是否影响无人机的正常飞行。
承接上述返航方向为无人机上方的假设，则对于障碍物是否影响无人机正常飞行的一种可选地理解方式可以为：当第一环境图像中障碍物的体积较大时，可以认为此大体积的障碍物会覆盖无人机上方的大部分区域，此时，可以认为此大体积的障碍物会影响无人机正常飞行，以直接控制无人机悬停。
当第一环境图像中的障碍物体积较小时，可以认为此小体积的障碍物不会覆盖无人机上方的大部分区域，但其有可能位于无人机的正上方，这样的话，无人机在上升返航过程中也有可能与其发生碰撞。因此，对于不影响无人机正常飞行的小体积障碍物，无人机还需要执行后续判断步骤，以确定此障碍物是否真的会对无人机的返航造成影响。
基于上述理解，对于确定障碍物是否影响无人机正常飞行的方式，可选地，可以先确定出障碍物在第一环境图像中的尺寸信息，并以此尺寸信息估计障碍物的实际尺寸信息。若实际尺寸信息大于预设尺寸，则确定此障碍物影响无人机的正常飞行。否则，确定此障碍物不影响无人机的正常飞行。
S103，根据无人机以第一预设方式飞行过程中得到的第二环境图像，确定障碍物与无人机之间的距离，第二环境图像对应于无人机的返航方向。
S104，若距离大于预设距离，则对返航指令进行响应，使无人机返航。
无人机在以第一预设方式飞行也即是不断上升飞行的过程中可以不断拍得第二环境图像。可选地，第二环境图像对应的视场可以为单目摄像机自身的视场，且每张第二环境图像的拍摄位置不同。此时，无人机可以根据在不同位置拍得的第二环境图像来计算障碍物与无人机之间的距离。此距离具体可以包括障碍物与无人机间的水平距离和垂直距离。距离的详细计算方式可以参见下述如图3～图4所示的实施例。
若计算出的无人机与障碍物之间的水平距离大于预设距离，则此时可以 对返航指令进行响应，直接控制无人机返航。若水平距离小于或等于预设距离，表明无人机与障碍物之间的距离较小，如若返航的话，无人机会与障碍物发生碰撞，因此，此时可以对悬停指令进行响应，以使无人机悬停。
本实施例中，获取无人机返航方向上的第一环境图像，并对第一环境图像中是否存在障碍物进行检测。此检测处理能够初步保证无人机的返航安全。接着，若第一环境图像中包含障碍物且此障碍物不影响无人机正常飞行，则会控制无人机以第一预设方式飞行，得到在以第一预设方式飞行的过程中拍得的第二环境图像，再根据此第二环境图像确定障碍物与无人机之间的距离。若距离大于预设距离，表明障碍物距离无人机较远，其不会对返航造成影响，则对返航指令进行响应，以使无人机返航。由于此距离能够表明障碍物的精确位置，也即是反映出障碍物是否完全位于返航方向上，因此，也就能够进一步保证无人机的返航安全。
根据上述实施例中的描述可知，无人机需要准确确定出自身上方的障碍物是否影响正常飞行，才能进一步准确计算出障碍物与无人机之间的距离，以控制无人机返航。然而障碍物的体积并不能精准地反映出障碍物是否影响正常飞行，因此使用图1所示实施例中提供的方式来确定障碍物是否影响正常飞行，得到的结果显然是不够精确的。因此，对于障碍物是否影响无人机的正常飞行的另一种较为精准的理解方式：
若第一环境图像中的障碍物位于无人机的正上方，则表明其会影响无人机的正常飞行。此时，可以直接控制无人机继续悬停。若障碍物没有位于无人机的正上方，则表明障碍物不影响无人机的正常飞行。但由于障碍物还有可能位于无人机的斜上方，无人机在上升返航过程中也有可能与其发生碰撞，因此，此时，无人机还需要继续执行后续步骤，以确定出此障碍物是否真的会对无人机的返航造成影响。
基于上述理解，另一种可选地确定障碍物是否影响无人机正常飞行的方式可以为：在得到第一环境图像后，可以确定图像中物体所属的类别以及不同类别的物体在第一环境图像中的位置。其中，物体类别以及物体位置的识别可以通过上述步骤102中提供的方式实现。同时还可以在第一环境图像中划分出一个预设区域，此区域可以认为是无人机的最小安全飞行区域，然后，判断物体类别为障碍物的物体是否在此预设区域内。
若障碍物位于此预设区域内，认为障碍物位于无人机正上方，则确定此障碍物影响无人机的正常飞行，此时，无人机对悬停指令进行响应，无人机处于悬停状态。若障碍物位于非预设区域内，则确定此障碍物不影响无人机的正常飞行。此时，可以进一步确定此确定障碍物与无人机之间的距离，以判断无人机是否可以返航，也即是继续执行如图1所示实施例中的步骤103～步骤104。
另外，对于第二环境图像的获取，虽然图1所示实施例步骤101中已经提供了一种方式，但该方式是通过无人机的上升飞行实现的。此时，由于无人机上方是存在障碍物的，因此，在上升飞行获取第二环境图像的过程中就更会容易出现无人机与障碍物发生碰撞，从而导致无人机损毁的情况。
而为了避免此种情况，实际应用中，第一飞行控制指令通常还可以为控制无人机沿与返航方向相反的方向由第一位置飞行至第二位置。继续承接上述返航方向为无人机上方的假设，第一飞行控制指令也即为使无人机由第一位置下降飞行至第二位置，且两位置相距预设距离。
在这种下降飞行的控制指令下，在上述各实施例的基础上，障碍物与无人机之间距离的测量方法，即步骤103一种可选地实现方式可以如图3所示：
S1031,无人机分别处于第一位置和第二位置时，获取无人机配置的单目摄像头拍摄的第二环境图像。
S1032,根据第一位置和第二位置各自对应的第二环境图像确定障碍物与无人机之间的距离。
无人机在响应第一飞行控制指令后，无人机可以由拍得第一环境图像的当前位置下降飞行至第一位置，在第一位置处由无人机上配置的能够上视的单目摄像头拍得第二环境图像。然后，再下降飞行至第二位置，在第二位置处拍得又一张第二环境图像，以根据在不同位置拍得的第二环境图像确定障碍物与无人机之间的距离。其中，容易理解的，无人机的下方也可以存在障碍物，因此，还可以利用无人机自身配置的传感器来确定无人机是否可以由第一位置下降飞行至第二位置。
具体来说，先根据单目摄像头在第一位置拍摄的第二环境图像，确定拍摄第二环境图像时，安装有单目摄像头的云台的第一姿态信息；再根据单目摄像头在第二位置拍摄的第二环境图像，确定拍摄第二环境图像时，云台的 第二姿态信息。
然后，结合图4中所示的各种参数，可以得到以下式子：
Figure PCTCN2020073660-appb-000001
整理后得到：
Figure PCTCN2020073660-appb-000002
在图4中，点A表示障碍物，点O表示无人机的机身中心，点O 1表示第一位置，点O 2表示第二位置。
其中，d为障碍物与无人机之间的水平距离，l 1为障碍物与无人机之间的垂直距离，l 2为第一位置O 1与第二位置O 2之间的预设距离，tanα 1为无人机处于第一位置O 1时云台的第一姿态角，tanα 2为无人机处于第二位置O 2时云台的第二姿态角。
当无人机采用图1所示实施例中的上升飞行的方式拍得第二环境图像时，也即是无人机从图4所示的位置O 2上升飞行至位置O 1时，利用上述方式同样可以得到障碍物与无人机之间的水平距离d以及位置O 1与无人机之间的距离l 1。最终计算距离l 1以及位置O 2与位置O 1之间的预设距离l 2之和，此距离之和也即为障碍物与无人机之间的垂直距离。
本实施例中，利用在不同位置拍得的第二环境图像来模拟双目摄像头拍得的图像，并利用双目测距原理精准地计算出障碍物与无人机之间的水平距离，以便根据此水平距离准确地确定出无人机是否能够返航，从而避免无人机在返航过程中出现损毁的情况。
上述各实施例中，无人机获取到的、对应于返航方向上的第一环境图像具有的视场也即是单目摄像头的视场，此视场较小，并不能全面展示无人机返航方向上障碍物的分布情况。此时，还可以控制无人机以特殊的方式飞行，从而获取到一个视角更大的第一环境图像，以更加全面地判断返航方向上是否存在障碍物。
具体来说，无人机可以对第二飞行控制指令进行响应，以使自身以第二 预设方式飞行。在以此第二预设方式飞行的过程中，无人机上配置的单目摄像头则可以拍得无人机返航方向上预设视场内的第一环境图像。
一种可选地第二预设方式可以为：无人机对旋转飞行控制指令进行响应，以使自身在当前位置原地旋转飞行一周。无人机飞行方式特殊性，便会直接导致得到的第一环境图像对应的视场具有特殊性。也就是说，由于无人机上的单目摄像头本身具有一定的视场角度，因此，在经过旋转飞行后，得到的第一环境图像对应的视场就是一个环形视场，可如图5所示，并且圆环的高度由单目摄像头的视场决定。
另一种可选地第二预设方式可以为：无人机对旋转飞行控制指令进行响应，以使无人机在当前位置原地旋转飞行一周。经过旋转飞行后，便可得到一第一环境图像。然后，无人机还可以对直线飞行控制指令进行响应，以使无人机水平飞行至与当前位置相距预设距离的第三位置。再对旋转飞行控制指令进行响应，使无人机在第三位置原地旋转飞行一周。在第三位置旋转飞行后，便可以得到又一第一环境图像。
使用此种方式，无人机会进行两次旋转飞行，从而可以得到多张第一环境图像，并且每张环境图像均对应于一个如图5所示的环形视场。此时，无人机可以分别识别多张第一环境图像中是否包含障碍物以及障碍物是否影响无人的正常飞行，并根据每张第一环境图像的识别结果来综合确定障碍物是否影响无人机的正常飞行。
本实施例中，通过控制无人机以特殊的飞行方式飞行，从而得到具有环形视场的环境图像，此环境图像能够更完整的展示无人机返航方向上障碍物的分布情况，从而更加全面的确定出返航方向上是否存在障碍物。并且在确定无人机的返航方向上是否存在障碍物时，使用的是具有环形视场的图像，而不是具有全部视场的图像，这样也可以减小障碍物检测过程中的计算量，提高检测效率。
虽然利用上述具有环形视场的第一环境图像能够更加全面地确定出无人机的返航方向上是否具有障碍物，但由于此环形视场远大于单目摄像头自身的视场，因此直接计算此大视场的环境图像中是否包含障碍物，此计算过程仍旧较为复杂。因此，在图5所示实施例的基础上，在以第二预设方式飞行得到具有环形视场的第一环境图像之前，如图6所示，该无人机返航的控制方法 还包括以下步骤：
S201，获取无人机配置的单目摄像头拍摄的无人机返航方向上的第三环境图像，第三环境图像对应的视场为单目摄像头的视场。
S202，识别第三环境图像中物体所属的类别。
S203，若第三环境图像中存在属于障碍物类别的物体，则使无人机在以第二预设方式飞行过程中，单目摄像头拍摄在无人机的返航方向上预设视场内的第一环境图像。
S204，若第三环境图像中不存在属于障碍物类别的物体，则对返航指令进行响应，使无人机返航。
单目摄像头可以先拍得无人机在返航方向上的第三环境图像，此第三环境图像对应的视场与单目摄像头的视场相同，并且此第三环境图像的拍摄位置可以是无人机的当前位置。然后，识别第三环境图像中包含的物体所属的类型。物体类别的识别过程可以参见图1步骤102中的相关描述。
若第三环境图中不存在障碍物，表明无人机的返航方向上没有障碍物，则可以直接控制无人机返航即可。若第三环境图像中存在障碍物，表明无人机在返航方向上存在障碍物。但仅利用第三环境图像还并不能全面地展示无人机返航方向上障碍物的分布情况，此时便可以进一步控制无人机按照第二预设方式飞行，从而得到具有如图5所示的环形视场的第一环境图像，此环形视场的环境图像能够全面展示无人机返航方向上障碍物的分布情况，也就能够更准确的判断出返航方向上是否存在障碍物，并进一步按照上述各实施例中提供的方式最终确定出无人机是否能够返航。
本实施例与图1、图5所示实施例相结合进行理解：先得到一个较小视场的第三环境图像，若此第三环境中不包含障碍物，则可以直接控制无人机返航。此种情况下，相比于图5所示实施例中直接使用较大视场的第一环境图像确定无人机是否能够返航的方式，此种方式的计算量更小，判断能够返航的效率也更高。若此第三环境图像中包含障碍物，便可以进一步确定出具有环形视场的第一环境图像中的障碍物是否影响无人机正常飞行，再进一步执行图1所示实施例的各步骤，从而判断出无人机能够返航。此种情况下，相比于直接使用小视场的第三环境图像，使用大视场的第一环境图像能够更加准确地确定出无人机能够返航。
此外，在上述实施例的基础上，当确定出无人机返航方向上不存在障碍 物时，即可控制无人返航。容易理解的，无人机的任何飞行过程都是需要电池供电的，因此，在控制无人机返航之前，还可以先确定返航过程中所需的电量，若当前的剩余电量多于返航所需电量时，才会控制无人机返航。
而对于返航过程中所需电量的确定，一种可选地方式，可以先根据历史风速信息估计从当前位置降落至返航目的地的风速信息。再确定从当前位置降落至返航目的地的地速信息，以根据风速信息和地速信息确定无人机的返航过程所需的电量。
图7为本发明实施例提供的一种无人机返航的控制装置的结构示意图；参考附图7所示，本实施例提供了一种无人机返航的控制装置，该无人机返航的控制装置可以执行上述的无人机返航的控制方法；具体的，无人机返航的控制装置包括：
获取模块11，用于获取无人机返航方向上的第一环境图像。
响应模块12，用于若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行。
确定模块13，用于根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向。
响应模块12，还用于若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航
图7所示装置还可以执行图1～图6所示实施例的方法，本实施例未详细描述的部分，可参考对图1～图6所示实施例的相关说明。该技术方案的执行过程和技术效果参见图1～图6所示实施例中的描述，在此不再赘述。
图8为本发明实施例提供的一种可移动平台的结构示意图；参考附图8所示，本发明实施例的提供了一种可移动平台，该可移动平台为以下至少之一：无人飞行器、无人船、无人车；具体的，该可移动平台包括：机体21、动力系统22以及控制装置23。
所述动力系统22，设置于所述机体21上，用于为所述可移动平台提供动力。
所述控制装置23包括存储器231和处理器232。
所述存储器，用于存储计算机程序；
处理器，用于运行所述存储器中存储的计算机程序以实现：
获取无人机返航方向上的第一环境图像；
若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
进一步的，处理器232还用于：若所述距离小于或等于所述预设距离，则对悬停指令进行响应，使所述无人机悬停。
进一步的，所述第一飞行控制指令为控制所述无人机沿与返航方向相反的方向由第一位置飞行至相距预设距离的第二位置；
处理器232还用于：所述无人机分别处于所述第一位置和所述第二位置时，获取所述无人机配置的单目摄像头拍摄的所述第二环境图像；
根据所述第一位置和所述第二位置各自对应的第二环境图像确定所述障碍物与所述无人机之间的距离。
进一步的，处理器232还用于：根据所述单目摄像头在所述第一位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，安装有所述单目摄像头的云台的第一姿态信息；
根据所述单目摄像头在所述第二位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，所述云台的第二姿态信息；
根据所述第一姿态信息、所述第二姿态信息以及所述第一位置与所述第二位置之间的预设距离，确定所述障碍物与所述无人机之间的距离。
进一步的，可移动平台还包括单目摄像头24，其设置于机体21上。
处理器232还用于：对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行；
在以所述第二预设方式飞行过程中，获取所述无人机配置的单目摄像头拍摄的在所述无人机返航方向上预设视场内的第一环境图像。
进一步的，所述第一环境图像对应的预设视场为环形视场。
处理器232还用于：对旋转飞行控制指令进行响应，使所述无人机在当前位置原地旋转飞行一周。
进一步的，处理器232还用于：对直线飞行控制指令进行响应，使所述无人机水平飞行至与所述当前位置相距预设距离的第三位置；
对旋转飞行控制指令进行响应，使所述无人机在所述第三位置原地旋转飞行一周。
进一步的，在获取第一环境图像之后，处理器232还用于：确定所述第一环境图像中物体所属的类别以及不同类别的物体在所述第一环境图像中的位置；
若属于障碍物类别的物体位于所述第一环境图像的非预设区域内，则确定所述第一环境图像中的障碍物不影响所述无人机正常飞行。
进一步的，处理器232还用于：若属于障碍物类别的物体位于所述第一环境图像的预设区域内，则对悬停指令进行响应，使所述无人机悬停。
进一步的，在获取第一环境图像之前，处理器232还用于：获取所述无人机配置的单目摄像头拍摄的无人机返航方向上的第三环境图像，所述第三环境图像对应的视场为所述单目摄像头的视场；
识别所述第三环境图像中物体所属的类别；
若所述第三环境图像中存在属于障碍物类别的物体，则使所述无人机在以第二预设方式飞行过程中，所述单目摄像头拍摄在所述无人机的返航方向上预设视场内的第一环境图像。
进一步的，处理器232还用于：若所述第三环境图像中不存在属于障碍物类别的物体，则对返航指令进行响应，使所述无人机返航。
图8所示的可移动平台可以执行图1～图6所示实施例的方法，本实施例未详细描述的部分，可参考对图1～图6所示实施例的相关说明。该技术方案的执行过程和技术效果参见图1～图6所示实施例中的描述，在此不再赘述。
在一个可能的设计中，图9所示无人机返航的控制设备的结构可实现为一电子设备，该电子设备可以是无人机。如图9所示，该电子设备可以包括：一个或多个处理器31和一个或多个存储器32。其中，存储器32用于存储支持电子设备执行上述图1～图6所示实施例中提供的无人机返航的控制方法的程序。处理器31被配置为用于执行存储器32中存储的程序。
具体的，程序包括一条或多条计算机指令，其中，一条或多条计算机指令被处理器31执行时能够实现如下步骤：
获取无人机返航方向上的第一环境图像；
若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
其中，该空域检测设备的结构中还可以包括通信接口33，用于电子设备与其他设备或通信网络通信。
进一步的，处理器31还用于：若所述距离小于或等于所述预设距离，则对悬停指令进行响应，使所述无人机悬停。
进一步的，所述第一飞行控制指令为控制所述无人机沿与返航方向相反的方向由第一位置飞行至相距预设距离的第二位置；
处理器31还用于：所述无人机分别处于所述第一位置和所述第二位置时，获取所述无人机配置的单目摄像头拍摄的所述第二环境图像；
根据所述第一位置和所述第二位置各自对应的第二环境图像确定所述障碍物与所述无人机之间的距离。
进一步的，处理器31还用于：根据所述单目摄像头在所述第一位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，安装有所述单目摄像头的云台的第一姿态信息；
根据所述单目摄像头在所述第二位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，所述云台的第二姿态信息；
根据所述第一姿态信息、所述第二姿态信息以及所述第一位置与所述第二位置之间的预设距离，确定所述障碍物与所述无人机之间的距离。
进一步的，处理器31还用于：对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行；
在以所述第二预设方式飞行过程中，获取所述无人机配置的单目摄像头拍摄的在所述无人机返航方向上预设视场内的第一环境图像。
进一步的，所述第一环境图像对应的预设视场为环形视场；
处理器31还用于：对旋转飞行控制指令进行响应，使所述无人机在当前位置原地旋转飞行一周。
进一步的，处理器31还用于：对直线飞行控制指令进行响应，使所述无人机水平飞行至与所述当前位置相距预设距离的第三位置；
对旋转飞行控制指令进行响应，使所述无人机在所述第三位置原地旋转飞行一周。
进一步的，在获取第一环境图像之后，处理器31还用于：确定所述第一环境图像中物体所属的类别以及不同类别的物体在所述第一环境图像中的位置；
若属于障碍物类别的物体位于所述第一环境图像的非预设区域内，则确定所述第一环境图像中的障碍物不影响所述无人机正常飞行。
进一步的，处理器31还用于：若属于障碍物类别的物体位于所述第一环境图像的预设区域内，则对悬停指令进行响应，使所述无人机悬停。
进一步的，在获取第一环境图像之前，处理器31还用于：获取所述无人机配置的单目摄像头拍摄的无人机返航方向上的第三环境图像，所述第三环境图像对应的视场为所述单目摄像头的视场；
识别所述第三环境图像中物体所属的类别；
若所述第三环境图像中存在属于障碍物类别的物体，则使所述无人机在以第二预设方式飞行过程中，所述单目摄像头拍摄在所述无人机的返航方向上预设视场内的第一环境图像。
进一步的，处理器31还用于：若所述第三环境图像中不存在属于障碍物类别的物体，则对返航指令进行响应，使所述无人机返航。
图9所示设备可以执行图1～图6所示实施例的方法，本实施例未详细描述的部分，可参考对图1～图6所示实施例的相关说明。该技术方案的执行过程和技术效果参见图1～图6所示实施例中的描述，在此不再赘述。
另外，本发明实施例提供了一种计算机可读存储介质，存储介质为计算机可读存储介质，该计算机可读存储介质中存储有程序指令，程序指令用于实现上述图1～图6的无人机返航的控制方法。
以上各个实施例中的技术方案、技术特征在与本相冲突的情况下均可以单独，或者进行组合，只要未超出本领域技术人员的认知范围，均属于本申 请保护范围内的等同实施例。
在本发明所提供的几个实施例中，应该理解到，所揭露的相关检测装置(例如：IMU)和方法，可以通过其它的方式实现。例如，以上所描述的遥控装置实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，遥控装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得计算机处理器(processor)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁盘或者光盘等各种可以存储程序代码的介质。
以上所述仅为本发明的实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。
最后应说明的是：以上各实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述各实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改， 或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。

Claims (35)

  1. 一种无人机返航的控制方法，其特征在于，所述方法包括：
    获取无人机返航方向上的第一环境图像；
    若所述第一环境图像中的障碍物识别结果满足第一条件，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
    根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
    若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    若所述距离小于或等于所述预设距离，则对悬停指令进行响应，使所述无人机悬停；
    或者，所述第一条件为所述第一环境图像中不包含代表障碍物的语义分类。
  3. 根据权利要求1所述的方法，其特征在于，所述第一飞行控制指令为控制所述无人机沿与返航方向相反的方向由第一位置飞行至相距预设距离的第二位置；
    所述根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，包括：
    所述无人机分别处于所述第一位置和所述第二位置时，获取所述无人机配置的单目摄像头拍摄的所述第二环境图像；
    根据所述第一位置和所述第二位置各自对应的第二环境图像确定所述障碍物与所述无人机之间的距离。
  4. 根据权利要求3所述的方法，其特征在于，所述根据所述第一位置和所述第二位置各自对应的第二环境图像确定所述障碍物与所述无人机之间的距离，包括：
    根据所述单目摄像头在所述第一位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，安装有所述单目摄像头的云台的第一姿态信息；
    根据所述单目摄像头在所述第二位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，所述云台的第二姿态信息；
    根据所述第一姿态信息、所述第二姿态信息以及所述第一位置与所述第 二位置之间的预设距离，确定所述障碍物与所述无人机之间的距离。
  5. 根据权利要求1所述的方法，其特征在于，所述获取无人机返航方向上的第一环境图像，包括：
    对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行；
    在以所述第二预设方式飞行过程中，获取所述无人机配置的单目摄像头拍摄的在所述无人机返航方向上预设视场内的第一环境图像。
  6. 根据权利要求5所述的方法，其特征在于，所述对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行，包括：
    对旋转飞行控制指令进行响应，使所述无人机在当前位置原地旋转飞行一周。
  7. 根据权利要求6所述的方法，其特征在于，所述对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行，还包括：
    对直线飞行控制指令进行响应，使所述无人机水平飞行至与所述当前位置相距预设距离的第三位置；
    对旋转飞行控制指令进行响应，使所述无人机在所述第三位置原地旋转飞行一周。
  8. 根据权利要求6或7所述的方法，其特征在于，所述第一环境图像对应的预设视场为环形视场。
  9. 根据权利要求1所述的方法，其特征在于，所述获取无人机返航方向上的第一环境图像之后，所述方法还包括：
    确定所述第一环境图像中物体所属的类别以及不同类别的物体在所述第一环境图像中的位置；
    若属于障碍物类别的物体位于所述第一环境图像的非预设区域内，则确定所述第一环境图像中的障碍物不影响所述无人机正常飞行。
  10. 根据权利要求9所述的方法，其特征在于，所述方法还包括：
    若属于障碍物类别的物体位于所述第一环境图像的预设区域内，则对悬停指令进行响应，使所述无人机悬停。
  11. 根据权利要求5所述的方法，其特征在于，所述获取无人机返航方向上的第一环境图像之前，所述方法还包括：
    获取所述无人机配置的单目摄像头拍摄的无人机返航方向上的第三环境图像，所述第三环境图像对应的视场为所述单目摄像头的视场；
    识别所述第三环境图像中物体所属的类别；
    若所述第三环境图像中存在属于障碍物类别的物体，则使所述无人机在以第二预设方式飞行过程中，所述单目摄像头拍摄在所述无人机的返航方向上预设视场内的第一环境图像。
  12. 根据权利要求11所述的方法，其特征在于，所述方法还包括：
    若所述第三环境图像中不存在属于障碍物类别的物体，则对返航指令进行响应，使所述无人机返航。
  13. 一种可移动平台，其特征在于，所述平台包括：平台本体、动力系统以及控制装置；
    所述动力系统，设置于所述平台本体上，用于为所述可移动平台提供动力；
    所述控制装置包括存储器和处理器；
    所述存储器，用于存储计算机程序；
    所述处理器，用于运行所述存储器中存储的计算机程序以实现：
    获取可移动平台返航方向上的第一环境图像；
    若所述第一环境图像中的障碍物不影响所述可移动平台正常飞行，则对第一飞行控制指令进行响应，使所述可移动平台以第一预设方式飞行；
    根据所述可移动平台以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述可移动平台之间的距离，所述第二环境图像对应于所述无人机的返航方向；
    若所述距离大于预设距离，则对返航指令进行响应，使所述可移动平台返航。
  14. 根据权利要求13所述的平台，其特征在于，所述处理器，还用于：
    若所述距离小于或等于所述预设距离，则对悬停指令进行响应，使所述无人机悬停。
  15. 根据权利要求13所述的平台，其特征在于，所述第一飞行控制指令为控制所述可移动平台沿与返航方向相反的方向由第一位置飞行至相距预设距离的第二位置；
    所述处理器，还用于：
    所述无人机分别处于所述第一位置和所述第二位置时，获取所述可移动平台配置的单目摄像头拍摄的所述第二环境图像；
    根据所述第一位置和所述第二位置各自对应的第二环境图像确定所述障碍物与所述可移动平台之间的距离。
  16. 根据权利要求15所述的平台，其特征在于，所述处理器，还用于：
    根据所述单目摄像头在所述第一位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，安装有所述单目摄像头的云台的第一姿态信息；
    根据所述单目摄像头在所述第二位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，所述云台的第二姿态信息；
    根据所述第一姿态信息、所述第二姿态信息以及所述第一位置与所述第二位置之间的预设距离，确定所述障碍物与所述无人机之间的距离。
  17. 根据权利要求13所述的平台，其特征在于，所述机体包括单目摄像头；所述处理器，还用于：
    对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行；
    在以所述第二预设方式飞行过程中，获取所述无人机配置的单目摄像头拍摄的在所述无人机返航方向上预设视场内的第一环境图像。
  18. 根据权利要求17所述的平台，其特征在于，所述第一环境图像对应的预设视场为环形视场；
    所述处理器，还用于：对旋转飞行控制指令进行响应，使所述无人机在当前位置原地旋转飞行一周。
  19. 根据权利要求18所述的平台，其特征在于，
    所述处理器，还用于：对直线飞行控制指令进行响应，使所述无人机水平飞行至与所述当前位置相距预设距离的第三位置；
    对旋转飞行控制指令进行响应，使所述无人机在所述第三位置原地旋转飞行一周。
  20. 根据权利要求13所述的平台，其特征在于，所述处理器，还用于：
    确定所述第一环境图像中物体所属的类别以及不同类别的物体在所述第一环境图像中的位置；
    若属于障碍物类别的物体位于所述第一环境图像的非预设区域内，则确定所述第一环境图像中的障碍物不影响所述无人机正常飞行。
  21. 根据权利要求20所述的平台，其特征在于，所述处理器，还用于：
    若属于障碍物类别的物体位于所述第一环境图像的预设区域内，则对悬停指令进行响应，使所述无人机悬停。
  22. 根据权利要求17所述的平台，其特征在于，所述处理器，还用于：
    获取所述无人机配置的单目摄像头拍摄的无人机返航方向上的第三环境图像，所述第三环境图像对应的视场为所述单目摄像头的视场；
    识别所述第三环境图像中物体所属的类别；
    若所述第三环境图像中存在属于障碍物类别的物体，则使所述无人机在以第二预设方式飞行过程中，所述单目摄像头拍摄在所述无人机的返航方向上预设视场内的第一环境图像。
  23. 根据权利要求22所述的平台，其特征在于，所述处理器，还用于：
    若所述第三环境图像中不存在属于障碍物类别的物体，则对返航指令进行响应，使所述无人机返航。
  24. 一种无人机返航的控制设备，其特征在于，所述控制设备包括：
    存储器，用于存储计算机程序；
    处理器，用于运行所述存储器中存储的计算机程序以实现：
    获取无人机返航方向上的第一环境图像；
    若所述第一环境图像中的障碍物不影响所述无人机正常飞行，则对第一飞行控制指令进行响应，使所述无人机以第一预设方式飞行；
    根据所述无人机以所述第一预设方式飞行过程中得到的第二环境图像，确定所述障碍物与所述无人机之间的距离，所述第二环境图像对应于所述无人机的返航方向；
    若所述距离大于预设距离，则对返航指令进行响应，使所述无人机返航。
  25. 根据权利要求24所述的设备，其特征在于，所述处理器，还用于：
    若所述距离小于或等于所述预设距离，则对悬停指令进行响应，使所述无人机悬停。
  26. 根据权利要求24所述的设备，其特征在于，所述第一飞行控制指令为控制所述无人机沿与返航方向相反的方向由第一位置飞行至相距预设距离的第二位置；
    所述处理器，还用于：
    所述无人机分别处于所述第一位置和所述第二位置时，获取所述无人机配置的单目摄像头拍摄的所述第二环境图像；
    根据所述第一位置和所述第二位置各自对应的第二环境图像确定所述障碍物与所述无人机之间的距离。
  27. 根据权利要求26所述的设备，其特征在于，所述处理器，还用于：
    根据所述单目摄像头在所述第一位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，安装有所述单目摄像头的云台的第一姿态信息；
    根据所述单目摄像头在所述第二位置拍摄的第二环境图像，确定拍摄所述第二环境图像时，所述云台的第二姿态信息；
    根据所述第一姿态信息、所述第二姿态信息以及所述第一位置与所述第二位置之间的预设距离，确定所述障碍物与所述无人机之间的距离。
  28. 根据权利要求24所述的设备，其特征在于，所述处理器，还用于：
    对第二飞行控制指令进行响应，使所述无人机以第二预设方式飞行；
    在以所述第二预设方式飞行过程中，获取所述无人机配置的单目摄像头拍摄的在所述无人机返航方向上预设视场内的第一环境图像。
  29. 根据权利要求28所述的设备，其特征在于，所述第一环境图像对应的预设视场为环形视场；
    所述处理器，还用于：对旋转飞行控制指令进行响应，使所述无人机在当前位置原地旋转飞行一周。
  30. 根据权利要求29所述的设备，其特征在于，所述处理器，还用于：
    对直线飞行控制指令进行响应，使所述无人机水平飞行至与所述当前位置相距预设距离的第三位置；
    对旋转飞行控制指令进行响应，使所述无人机在所述第三位置原地旋转飞行一周。
  31. 根据权利要求24所述的设备，其特征在于，所述处理器，还用于：
    确定所述第一环境图像中物体所属的类别以及不同类别的物体在所述第一环境图像中的位置；
    若属于障碍物类别的物体位于所述第一环境图像的非预设区域内，则确定所述第一环境图像中的障碍物不影响所述无人机正常飞行。
  32. 根据权利要求31所述的设备，其特征在于，所述处理器，还用于：
    若属于障碍物类别的物体位于所述第一环境图像的预设区域内，则对悬停指令进行响应，使所述无人机悬停。
  33. 根据权利要求28所述的设备，其特征在于，所述处理器，还用于：
    获取所述无人机配置的单目摄像头拍摄的无人机返航方向上的第三环境图像，所述第三环境图像对应的视场为所述单目摄像头的视场；
    识别所述第三环境图像中物体所属的类别；
    若所述第三环境图像中存在属于障碍物类别的物体，则使所述无人机在以第二预设方式飞行过程中，所述单目摄像头拍摄在所述无人机的返航方向上预设视场内的第一环境图像。
  34. 根据权利要求33所述的设备，其特征在于，所述处理器，还用于：
    若所述第三环境图像中不存在属于障碍物类别的物体，则对返航指令进行响应，使所述无人机返航。
  35. 一种计算机可读存储介质，其特征在于，所述存储介质为计算机可读存储介质，该计算机可读存储介质中存储有程序指令，所述程序指令用于实现权利要求1至12中任一项所述的无人机返航的控制方法。
PCT/CN2020/073660 2020-01-21 2020-01-21 无人机返航的控制方法、设备、可移动平台和存储介质 WO2021146973A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202080004128.9A CN112639655A (zh) 2020-01-21 2020-01-21 无人机返航的控制方法、设备、可移动平台和存储介质
PCT/CN2020/073660 WO2021146973A1 (zh) 2020-01-21 2020-01-21 无人机返航的控制方法、设备、可移动平台和存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/073660 WO2021146973A1 (zh) 2020-01-21 2020-01-21 无人机返航的控制方法、设备、可移动平台和存储介质

Publications (1)

Publication Number Publication Date
WO2021146973A1 true WO2021146973A1 (zh) 2021-07-29

Family

ID=75291208

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/073660 WO2021146973A1 (zh) 2020-01-21 2020-01-21 无人机返航的控制方法、设备、可移动平台和存储介质

Country Status (2)

Country Link
CN (1) CN112639655A (zh)
WO (1) WO2021146973A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023272633A1 (zh) * 2021-06-30 2023-01-05 深圳市大疆创新科技有限公司 无人机的控制方法、无人机、飞行系统及存储介质
CN113970314B (zh) * 2021-10-18 2023-12-01 国网辽宁省电力有限公司抚顺供电公司 一种输电线路净空距离无人机航测方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105759836A (zh) * 2016-03-14 2016-07-13 武汉卓拔科技有限公司 一种基于3d摄像头的无人机避障方法及装置
CN106292704A (zh) * 2016-09-07 2017-01-04 四川天辰智创科技有限公司 规避障碍物的方法及装置
CN108334103A (zh) * 2017-12-21 2018-07-27 广州亿航智能技术有限公司 无人机多距离避障方法及避障系统
CN108427438A (zh) * 2018-04-11 2018-08-21 北京木业邦科技有限公司 飞行环境检测方法、装置、电子设备及存储介质
CN109782788A (zh) * 2019-03-26 2019-05-21 台州学院 基于双目视觉的无人机低空避障系统及控制方法
US20190202449A1 (en) * 2018-01-03 2019-07-04 Qualcomm Incorporated Adjustable Object Avoidance Proximity Threshold Based on Classification of Detected Objects

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106814744A (zh) * 2017-03-14 2017-06-09 吉林化工学院 一种无人机飞行控制系统以及方法
CN108700890B (zh) * 2017-06-12 2021-10-29 深圳市大疆创新科技有限公司 无人机返航控制方法、无人机和机器可读存储介质
CN107479575B (zh) * 2017-08-25 2019-12-17 中国地质大学（武汉） 一种多旋翼无人机飞行控制方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105759836A (zh) * 2016-03-14 2016-07-13 武汉卓拔科技有限公司 一种基于3d摄像头的无人机避障方法及装置
CN106292704A (zh) * 2016-09-07 2017-01-04 四川天辰智创科技有限公司 规避障碍物的方法及装置
CN108334103A (zh) * 2017-12-21 2018-07-27 广州亿航智能技术有限公司 无人机多距离避障方法及避障系统
US20190202449A1 (en) * 2018-01-03 2019-07-04 Qualcomm Incorporated Adjustable Object Avoidance Proximity Threshold Based on Classification of Detected Objects
CN108427438A (zh) * 2018-04-11 2018-08-21 北京木业邦科技有限公司 飞行环境检测方法、装置、电子设备及存储介质
CN109782788A (zh) * 2019-03-26 2019-05-21 台州学院 基于双目视觉的无人机低空避障系统及控制方法

Also Published As

Publication number Publication date
CN112639655A (zh) 2021-04-09

Similar Documents

Publication Publication Date Title
US11361665B2 (en) Unmanned aerial vehicle privacy controls
US11604479B2 (en) Methods and system for vision-based landing
US11794890B2 (en) Unmanned aerial vehicle inspection system
US11550315B2 (en) Unmanned aerial vehicle inspection system
US11287835B2 (en) Geo-fiducials for UAV navigation
CN107240063B (zh) 一种面向移动平台的旋翼无人机自主起降方法
EP3901728A1 (en) Methods and system for autonomous landing
EP3249487A2 (en) Uav and uav landing control device and method
KR102254491B1 (ko) 지능형 영상 분석 모듈을 탑재한 자율비행 드론
JP2024053085A (ja) 飛行体制御装置、飛行体制御方法、及びプログラム
WO2017139282A1 (en) Unmanned aerial vehicle privacy controls
CN109885086A (zh) 一种基于复合多边形标志引导的无人机垂直降落方法
WO2021146973A1 (zh) 无人机返航的控制方法、设备、可移动平台和存储介质
WO2020062178A1 (zh) 基于地图识别目标对象的方法与控制终端
WO2021237618A1 (zh) 抓捕辅助方法、地面指挥平台、无人机、系统及存储介质
EP3480118A1 (en) Aerial vehicle landing method, aerial vehicle, and computer readable storage medium
EP3893078A1 (en) Relay point generation method and apparatus, and unmanned aerial vehicle
WO2021129351A1 (zh) 无人机保护方法及装置、无人机
US20220392353A1 (en) Unmanned aerial vehicle privacy controls
WO2021056139A1 (zh) 获取降落位置的方法、设备、无人机、系统及存储介质
KR20190097350A (ko) 드론의 정밀착륙을 위한 방법, 이를 수행하기 위한 기록매체, 및 이를 적용한 드론
JP6481228B1 (ja) 決定装置、制御装置、撮像システム、飛行体、決定方法、及びプログラム
WO2021146972A1 (zh) 空域检测方法、可移动平台、设备和存储介质
WO2021146969A1 (zh) 距离测量方法、可移动平台、设备和存储介质
KR20230016730A (ko) 착륙지점에 드론이 정밀하게 착륙할 수 있도록 유도하는 자동착륙 시스템

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20915519

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20915519

Country of ref document: EP

Kind code of ref document: A1