WO2023173307A1 - 可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质 - Google Patents

可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质 Download PDF

Info

Publication number
WO2023173307A1
WO2023173307A1 PCT/CN2022/081105 CN2022081105W WO2023173307A1 WO 2023173307 A1 WO2023173307 A1 WO 2023173307A1 CN 2022081105 W CN2022081105 W CN 2022081105W WO 2023173307 A1 WO2023173307 A1 WO 2023173307A1
Authority
WO
WIPO (PCT)
Prior art keywords
movable platform
image
blade
detachable
component
Prior art date
Application number
PCT/CN2022/081105
Other languages
English (en)
French (fr)
Inventor
梁湘国
高文良
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 PCT/CN2022/081105 priority Critical patent/WO2023173307A1/zh
Publication of WO2023173307A1 publication Critical patent/WO2023173307A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B64—AIRCRAFT; AVIATION; COSMONAUTICS
    • B64F—GROUND OR AIRCRAFT-CARRIER-DECK INSTALLATIONS SPECIALLY ADAPTED FOR USE IN CONNECTION WITH AIRCRAFT; DESIGNING, MANUFACTURING, ASSEMBLING, CLEANING, MAINTAINING OR REPAIRING AIRCRAFT, NOT OTHERWISE PROVIDED FOR; HANDLING, TRANSPORTING, TESTING OR INSPECTING AIRCRAFT COMPONENTS, NOT OTHERWISE PROVIDED FOR
    • B64F5/00—Designing, manufacturing, assembling, cleaning, maintaining or repairing aircraft, not otherwise provided for; Handling, transporting, testing or inspecting aircraft components, not otherwise provided for
    • B64F5/60—Testing or inspecting aircraft components or systems
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B64—AIRCRAFT; AVIATION; COSMONAUTICS
    • B64C—AEROPLANES; HELICOPTERS
    • B64C27/00—Rotorcraft; Rotors peculiar thereto
    • B64C27/04—Helicopters
    • B64C27/08—Helicopters with two or more rotors

Definitions

  • the present application relates to the technical field of movable platforms. Specifically, it relates to a movable platform and its control method, information prompting method, device, electronic equipment, and computer-readable storage medium.
  • movable platforms are widely used in aerial photography, agriculture, plant protection, transportation, surveying and mapping, disaster relief and other fields.
  • the movable platform can move in space automatically or under user control. How to ensure the safe movement of the movable platform is a technical issue that has been of concern in this field.
  • this application provides a control method, information prompting method, device, movable platform, electronic equipment and computer-readable storage medium for a movable platform to solve the problem of safe movement of the movable platform in related technologies.
  • a first aspect provides an information prompting method for a rotorcraft, wherein a plurality of rotors of the rotorcraft include detachable blades, and the plurality of rotors include different rotors adapted to different blade models;
  • the methods include:
  • a prompt message indicating adjustment of the blade installation status is output.
  • a method for controlling a movable platform includes at least one visual sensor, and at least one detachable component is used to connect with the movable platform; the method includes:
  • the detachable component If the detachable component is not in the safe connection state, output prompt information, the prompt information is used to instruct the reconnection of the detachable component to the safe connection state;
  • the movable platform After determining that the detachable component is in a safe connection state based on the image, the movable platform is controlled to move in space based on the environmental image collected by the visual sensor.
  • an information prompting method for a movable platform includes propeller blades of different types, and the propeller blades are detachably connected to the movable platform; wherein, the propeller blades of different types The connection positions of the leaves and the movable platform are different; the method includes:
  • prompt information is output, and the prompt information is used to instruct the connection position to be reconnected to the propeller blade.
  • a fourth aspect provides an information prompting method for a movable platform, the movable platform including at least one detachable component; the method includes:
  • prompt information is output, and the prompt information is used to instruct the reconnection of the detachable component from the current state to the safe connection state.
  • a control device for a movable platform includes a processor, a memory, and a computer program stored on the memory and executable by the processor.
  • the processor executes the computer program Implement the method described in the first aspect.
  • a sixth aspect provides an information prompting device.
  • the device includes a processor, a memory, and a computer program stored on the memory and executable by the processor.
  • the processor executes the computer program, the first The method described in the aspect, the third aspect or the fourth aspect.
  • a movable platform in a seventh aspect, includes at least one visual sensor and at least one detachable component for connecting with the movable platform; the movable platform also includes a processor, a memory, a storage device, and a processor. a computer program executable by the processor on the memory;
  • An eighth aspect provides an electronic device, the processor, a memory, and a computer program stored on the memory and executable by the processor;
  • a computer-readable storage medium is provided.
  • a number of computer instructions are stored on the computer-readable storage medium.
  • the steps of any one of the methods described in the first to fourth aspects are implemented.
  • the movable platform can automatically detect whether the detachable parts are in a safe connection state without requiring additional user operations.
  • This scheme is based on the images collected by the visual sensor of the movable platform on the detachable parts. to identify whether the detachable part is in a safe connection state, so no additional hardware cost is required; since prompt information for instructing to reconnect the detachable part to the safe connection state can be output, the user discovers in time that the detachable part is not in the safe connection state. In a secure connection state, further ensuring the safe movement of the movable platform.
  • Figure 1A is a schematic architecture diagram of an unmanned aerial system according to an embodiment of the present application.
  • Figure 1B is a schematic diagram of a quad-rotor drone in a stowed state according to an embodiment of the present application.
  • Figure 1C is a schematic diagram of a quad-rotor UAV in an operational state according to an embodiment of the present application.
  • FIG. 1D is a schematic flowchart of an information prompting method for a rotorcraft according to an embodiment of the present application.
  • FIG. 2A is a schematic flowchart of a control method for a movable platform according to an embodiment of the present application.
  • Figure 2B is a schematic diagram of communication between a drone and other devices according to an embodiment of the present application.
  • Figure 2C is a schematic diagram of outputting prompt information according to an embodiment of the present application.
  • Figure 3A is a schematic diagram of a machine learning model according to an embodiment of the present application.
  • Figure 3B is a schematic diagram of six blade installation methods of a quad-rotor UAV according to an embodiment of the present application.
  • Figure 3C is a schematic diagram of another way of outputting prompt information according to an embodiment of the present application.
  • Figure 4 is a schematic flowchart of an information prompting method according to an embodiment of the present application.
  • FIG. 5 is a schematic flowchart of an information prompting method according to another embodiment of the present application.
  • Figure 6 is a structural diagram of a control device for a movable platform according to an embodiment of the present application.
  • Figure 7 is a structural diagram of an information prompting device according to an embodiment of the present application.
  • Figure 8 is a structural diagram of a movable platform according to an embodiment of the present application.
  • Figure 9 is a structural diagram of an electronic device according to an embodiment of the present application.
  • the mobile platform in the embodiment of the present application may include a variety of devices with mobile capabilities, such as drones, vehicles, robots, cleaning equipment, or unmanned ships, and so on.
  • the movable platform can connect a variety of components.
  • FIG. 1A is a schematic architecture diagram of an unmanned aerial system according to an embodiment of the present application.
  • the unmanned aerial system 100 may include a drone 110, a display device 130 and a remote control device 140.
  • the UAV 110 may include a power system 150, a flight control system 160 (referred to as flight control), a frame, and a pan/tilt 120 carried on the frame; the UAV 110 may include a rotor UAV.
  • Drone 110 may communicate wirelessly with remote control device 140 and display device 130 .
  • the frame may include the fuselage and legs (also called landing gear).
  • the fuselage may include a center frame and one or more arms connected to the center frame, and the one or more arms extend radially from the center frame.
  • the tripod is connected to the fuselage and used to support the UAV 110 when it lands.
  • the power system 150 may include one or more electronic speed regulators (electric speed regulators for short) 151, one or more propellers 153, and one or more power motors 152 corresponding to the one or more propellers 153, wherein the power motor 152 Connected between the electronic speed regulator 151 and the propeller 153, the power motor 152 and the propeller 153 can be arranged on the arm of the drone 110, for example, the end of the arm away from the fuselage; the electronic speed regulator 151 is used to receive flight control
  • the system 160 generates a driving signal and provides a driving current to the power motor 152 according to the driving signal to control the rotation speed of the power motor 152 .
  • the power motor 152 is used to drive the propeller to rotate, thereby providing power for the flight of the drone 110.
  • the power enables the drone 110 to achieve movement with one or more degrees of freedom.
  • drone 110 may rotate about one or more axes of rotation.
  • the above-mentioned rotation axis may include a roll axis (Roll), a yaw axis (Yaw), and a pitch axis (pitch).
  • the motor 152 may be a DC motor or an AC motor.
  • the motor 152 may be a brushless motor or a brushed motor.
  • Flight control system 160 may include flight controller 161 and sensing system 162 .
  • the sensing system 162 is used to measure the attitude information of the UAV, that is, the position information and status information of the UAV 110 in space, such as three-dimensional position, three-dimensional angle, three-dimensional speed, three-dimensional acceleration, three-dimensional angular velocity, etc.; the sensing system 162
  • Other information can also be collected, such as positioning information, or information about the scenery in the space where the drone is located, such as depth information or heat information, etc.
  • the sensing system 162 may include, for example, at least one of a gyroscope, an ultrasonic sensor, an electronic compass, an inertial measurement unit (IMU), a visual sensor, a heat meter, a global navigation satellite system, a barometer, and other sensors.
  • the global navigation satellite system may be the Global Positioning System (GPS).
  • the flight controller 161 is used to control the flight of the UAV 110.
  • the flight of the UAV 110 can be controlled based on the attitude information measured by the sensing system 162. It should be understood that the flight controller 161 can control the UAV 110 according to pre-programmed instructions, or can also control the UAV 110 by responding to one or more remote control signals from the remote control device 140 .
  • the pan/tilt 120 may include a motor 122 .
  • the pan/tilt is used to carry various devices such as the shooting device 123 and so on.
  • the flight controller 161 can control the movement of the gimbal 120 through the motor 122 .
  • the pan/tilt 120 may also include a controller for controlling the movement of the pan/tilt 120 by controlling the motor 122 .
  • the gimbal 120 can be independent of the drone 110 or can be a part of the drone 110 .
  • the motor 122 may be a DC motor or an AC motor.
  • the motor 122 may be a brushless motor or a brushed motor.
  • there are many ways to realize the position of the gimbal for example, it can be located on the top of the drone, or it can be located on the bottom of the drone, etc.
  • the photographing device 123 may be, for example, a camera or video camera or other device used to capture images.
  • the photographing device 123 may communicate with the flight controller and perform photography under the control of the flight controller.
  • the photographing device 123 of this embodiment at least includes a photosensitive element, such as a complementary metal oxide semiconductor (Complementary Metal Oxide Semiconductor, CMOS) sensor or a charge-coupled device (Charge-coupled Device, CCD) sensor. It can be understood that the shooting device 123 can also be directly fixed on the drone 110, so the pan/tilt 120 can be omitted.
  • CMOS Complementary Metal Oxide Semiconductor
  • CCD charge-coupled Device
  • the display device 130 is located at the ground end of the unmanned aerial system 100, can communicate with the UAV 110 through wireless means, and can be used to display attitude information of the UAV 110.
  • the image captured by the photographing device 123 may also be displayed on the display device 130 .
  • the display device 130 may be an independent device or may be integrated into the remote control device 140 .
  • the remote control device 140 is located at the ground end of the unmanned aerial system 100 and can communicate with the UAV 110 through wireless means for remote control of the UAV 110 .
  • connection status of some components may affect the safe movement of the movable platform.
  • the connection status may affect the safe movement of the movable platform.
  • these parts due to the detachable characteristics of some detachable parts, due to various reasons, such as user disassembly, user error installation or loose parts, etc., these parts have multiple different connection states with the movable platform, and some connections The status may affect the safe movement of the movable platform.
  • Figure 1B is a schematic diagram of a four-rotor drone shown in this application according to an exemplary embodiment.
  • the drone 110 is in a non-working storage state. state, the UAV in Figure 1B includes a fuselage 170 and four arms. Taking the arm 171 as an example, the arm 171 is foldable and connected to the fuselage 170, and the arm 171 is not unfolded; the UAV in Figure 1B
  • the machine 110 also includes four propellers. Taking the propeller 153 as an example, the blades of the propeller 153 are also foldable, and the blades of the propeller 153 are not unfolded.
  • FIG. 1C it is another schematic diagram of a four-rotor UAV according to an exemplary embodiment of the present application.
  • the UAV 110 is in an operable state, with four arms and four propellers. In expanded state.
  • Detachable components may include components related to the movement of the movable platform and components related to the normal performance of tasks by the movable platform. Due to the detachable characteristics of the detachable parts, the detachable parts and the movable platform may have multiple connection states. Only when the detachable parts are safely connected can the movable platform be in normal working condition and the movable platform guaranteed. safe movement.
  • the manufacturer of the movable platform usually provides an instruction manual for the product. The user reads the instruction manual to understand the safe connection status of the detachable parts, and then the user operates the detachable parts to operate the detachable parts to the safe connection status.
  • This method requires the user to consult the operating instructions by himself, which brings great trouble to the user; moreover, there may be situations where the user fails to operate the detachable parts to a safe connection state, for example, the user forgets the safe connection state of the detachable parts, user operation is not in place, or the user hands the movable platform to other users who are not familiar with the movable platform, etc.; in addition, the movable platform may be connected to multiple detachable parts, and these multiple detachable parts are very similar in appearance, and the user It is difficult to distinguish, but requires connection at different positions of the movable platform; these situations may cause the detachable parts to fail to be in a safe connection state, which will affect the subsequent work of the movable platform and bring greater trouble to the movement of the movable platform. Big safety hazard.
  • this specification shows an information prompting method for a rotorcraft according to an exemplary embodiment.
  • the plurality of rotors of the rotorcraft include detachable blades, and the plurality of rotors include Adapt to different rotors with different blade models; the method includes:
  • step 1110 obtain multiple preset blade models of the rotors
  • step 1112 identify the blade model installed on each rotor of the rotorcraft based on the image data
  • step 1114 if the identified blade model does not match the preset blade model, a prompt message indicating adjustment of the blade installation status is output.
  • the solution of this embodiment can be executed by a rotorcraft or a ground controller connected to the rotorcraft, where the image data can be collected by an image sensor on the rotorcraft or an image sensor of a ground controller connected to the rotorcraft.
  • the rotorcraft or electronic device of this embodiment can acquire image data collected by any of the above image sensors.
  • the number of rotors of rotor drones includes two rotors, four rotors, six rotors, eight rotors, etc.
  • a power motor drives a propeller to rotate.
  • the pitch, roll and yaw movements of the aircraft can be achieved.
  • the flight control system in the aircraft can adjust the speed of the four motors to change the propeller speed to achieve changes in lift, thereby controlling the attitude and position of the aircraft.
  • air resistance will form a counter-torque opposite to the direction of rotation.
  • propeller blades in rotorcraft are usually designed in pairs, including forward propeller blades and reverse propeller blades.
  • the forward propeller blades and reverse propeller blades need to be installed in the correct position in the drone, such as forward propeller blades.
  • the number of propeller blades is usually 2 or more, the length of the propeller is also different, and different drones are designed with different types of propellers.
  • each rotor can be pre-set with a corresponding preset blade model. Based on this, the blade model installed on each rotor of the rotorcraft is identified through image data. If the identified blade If the blade model does not match the preset blade model, a prompt message indicating adjustment of the blade installation status is output, thereby prompting the user so that the user can adjust the blade installation status to ensure the safety of the rotorcraft. move.
  • Blade models may indicate different blade configurations.
  • the forward propeller or the reverse propeller of a multi-rotor aircraft under normal installation, the forward propeller and the reverse propeller rotate in opposite directions to partially offset the torque in the horizontal direction).
  • blade configurations adapted to different flight conditions such as plain blades, silent blades, plateau blades, etc.
  • plateau blades may have longer blade lengths to provide higher lift at the same rotational speed and consume greater power per unit time.
  • Silent blades may not be much different in size from plain blades, but the ends of the blades are treated with special materials and shapes to reduce wind noise caused by the rotation of the blades cutting the air.
  • multi-rotor aircraft are often packaged with a variety of different types of blades.
  • the blades are detachably connected to the corresponding rotors, and users can manually remove or install the blades.
  • manual operations will inevitably introduce assembly errors.
  • plateau blades and plain blades are installed on the fuselage, the lift generated by the blades will be inconsistent, which will seriously affect the flight safety of the aircraft.
  • plain blades and silent blades are installed on the fuselage at the same time. This installation may have a controllable impact on the stability of the aircraft, but it may not meet the user's operational needs for noise reduction.
  • the blade model may include appearance information that characterizes the appearance of the blade.
  • the appearance information may be any of the following: shape information, length information, blade number information, color information, material information, or size information. Etc. Based on this, through the recognition of appearance information, it can be accurately determined whether the blades connected to each rotor are correct.
  • the rotorcraft includes one or more image sensors, the observation range of the one or more image sensors covers multiple rotors of the rotorcraft, and the image data is based on the one or more image sensors. collected by the image sensor.
  • the image sensor mounted on the rotorcraft can be used to collect data.
  • the observation range of the image sensor simultaneously covers the surrounding environment of the rotorcraft, and the image sensor is used to observe the surrounding environment when the rotor is in a flight state, so that multiple The rotorcraft adjusts one or more rotor operating parameters based on the observed information.
  • the rotorcraft's existing visual sensor for obstacle avoidance can be used for image collection, and costs can be reduced by reusing the visual sensor.
  • the prompt information includes image prompt information displayed on the ground controller of the rotorcraft; the image prompt information includes: rotor identification indicating a plurality of the rotors, and indicating the rotor adaptation.
  • the ground controller of the rotorcraft may include a mobile terminal or a remote control, etc.
  • the ground controller may include a display, and the ground controller may display image prompt information, since the image prompt information includes indicating the rotors of multiple rotors. Identifies and indicates the first blade mark of the preset blade model that the rotor is adapted to, so that the user can intuitively check each rotor and its adapted preset blade model, so that the user knows how to install it correctly. rotor.
  • the image prompt information may also include: the second blade identification of the blade model currently installed on any of the rotors. Based on this, the user can intuitively check which rotor is currently installed on the rotorcraft. The incorrect blade is removed so that the user can clearly identify which rotor to reinstall with the appropriate blade.
  • the image prompt information includes: an adjustment guideline indicating the relationship between the first blade identification of the preset blade model adapted to the rotor and the corresponding rotor. Based on this, for installation For the wrong rotor, the adjustment guide mark can indicate to the user the preset blade model that the rotor is suitable for, allowing the user to quickly and correctly adjust the blades of the rotor.
  • the image prompt information includes: a fuselage logo indicating the fuselage of the rotorcraft; a display position relationship between the rotor logo and the fuselage logo, and the position of the rotor on the rotorcraft.
  • the position distribution of the fuselage is consistent.
  • the image can be displayed according to the actual position relationship between the rotorcraft and the rotor, so that the user can compare each rotor in the rotorcraft through the image prompt information to facilitate the user to adjust the blades.
  • this specification also provides an embodiment of a control method for a movable platform, as shown in Figure 2A, which is a flow chart of the control method for the movable platform, including the following steps:
  • step 202 images are collected based on the visual sensor of the detachable component of the movable platform.
  • step 204 it is determined whether the detachable component is in a safe connection state based on the collected image.
  • step 206 if the detachable component is not in the safe connection state, prompt information is output, and the prompt information is used to instruct the reconnection of the detachable component to the safe connection state.
  • step 208 after it is determined that the detachable component is in a safe connection state based on the image, the movable platform is controlled to move in space based on the environmental image collected by the visual sensor.
  • the movable platform can automatically detect whether the detachable parts are in a safe connection state without requiring additional user operations.
  • This solution is based on the image collected by the visual sensor of the movable platform to identify whether the detachable parts are in a safe connection state.
  • the secure connection state no additional hardware cost is required; since prompt information for instructing to reconnect the detachable component to the secure connection state can be output, the user can be reminded, further ensuring the safe movement of the movable platform.
  • the execution timing of the solution in this embodiment can be set as needed in actual applications.
  • the step of collecting images of the detachable component of the movable platform based on the visual sensor is performed based on messages sent by other devices communicating with the movable platform; and/or, It is executed based on the movement status information of the movable platform; and/or it is executed based on the trigger signal generated after the movable platform is powered on.
  • the movable platform may be executed before the start of the job task.
  • the trigger signal may be generated after detecting the start of the job task to automatically trigger execution of the solution of this embodiment.
  • it may also be executed based on a message sent by other devices communicating with the movable platform.
  • the message may include a message instructing the movable platform to start the operation.
  • the movable platform determines through the message that the user instructs the movable platform to start the operation. , execute the solution of this embodiment.
  • it can also be performed based on the movement status information of the movable platform.
  • the implementation of the solution in this embodiment can be performed when the movable platform is not in a moving state.
  • the movement status information can be obtained through the inertial measurement sensor of the movable platform. Determined by the collected data. Or, it is executed based on a trigger signal generated after the movable platform is turned on.
  • the trigger signal may be generated after the movable platform detects that certain physical keys are triggered, such as the power button.
  • the trigger signal may be generated each time. It is generated after the power button is triggered, or it can be generated according to the set period.
  • other detection methods can be used to determine the execution timing of the solution of this embodiment.
  • the movable platform can detect that it is in certain scenarios before executing it.
  • the available data can be collected through visual sensors, lidar or ranging sensors.
  • Environmental observation data of the mobile platform Through the environmental observation data, it is detected that the mobile platform is in a safe environment, etc.
  • the movable platform includes at least one vision sensor. Under normal working conditions of the movable platform, the movable platform can move under the control of other devices that communicate with the movable platform, or the movable platform can automatically control the movement.
  • the visual sensor may include an image sensor for obstacle avoidance, and the images collected by the visual sensor for obstacle avoidance may be used to obtain depth information.
  • the visual sensor may measure the field of view. Images of the environment within the environment are collected. Based on the collected images, the depth information of the scenery within the field of view can be determined through the visual positioning algorithm. Using the determined depth information, the movable platform can control itself to move safely in space.
  • the image collected by the visual sensor is used by the movable platform to calculate the depth information, and the image is not displayed to the user; optionally, in some scenarios, the image collected by the visual sensor can also be displayed to the user as needed.
  • the visual sensor may also include image sensors with other functions, such as sensors used to collect images displayed to the user, etc. This embodiment is not limited to this.
  • the movable platform includes one or more vision sensors, and the vision sensors may be fixedly or non-fixedly disposed on the movable platform.
  • one or more visual sensors can be selected to collect images as needed. For example, the selection may be based on the field of view of the vision sensor and/or the connection position of the detachable component that needs to be identified on the movable platform. For example, for a detachable component that needs to be identified, a visual sensor whose field of view covers the connection position between the detachable component and the movable platform is selected.
  • the visual sensor is fixedly disposed on the movable platform, and the field of view of the visual sensor covers the connection position of the movable platform for connecting the detachable components.
  • the visual sensor is mounted on the movable platform through a pan/tilt; the image is made by controlling the pan/tilt to make the visual sensor face the movable platform for connecting the detachable platform. Collected after the connection position of the components.
  • the collected image may be an image collected by one visual sensor, or may be an image of multiple visual sensors.
  • it can be one or more images collected by the same visual sensor, or it can be an image collected by multiple sensors respectively, or it can be multiple images collected by different sensors among multiple sensors, etc.
  • the collected image may include all or part of the movable platform, that is, the image content may include the entire movable platform, or may be only a part of the movable platform.
  • the collected images are not limited to whether they include detachable parts.
  • the detachable parts are disassembled and not connected to the movable platform, so the detachable parts are not included in the collected images; or the detachable parts are not securely connected to the movable platform, so the collected images Including at least some detachable parts, such as all or part of the detachable parts.
  • the collected images include all the detachable parts, or the connection with the movable platform, etc. In practical applications, it can be configured as needed, as long as the collected images can be used to detect whether the detachable parts are safely connected to the movable platform.
  • the detachable components are detachably connected to the movable platform, and may include components related to the movement of the movable platform and components related to the normal performance of tasks by the movable platform.
  • the mobile platform includes a drone.
  • the detachable parts include any of the following: machine arms, propeller blades, landing gear, gimbal, gimbal protection parts, antennas or battery components, etc.
  • the movable platform includes a cleaning robot.
  • the detachable parts include any of the following: cleaning parts, protection parts of the cleaning parts, driving parts or protection parts of the cleaning robot, etc.
  • the movable platform includes a vehicle.
  • the detachable parts include any of the following: doors, windows, wheels, rear mirrors, lights, hoods, antennas, wipers or sensors, etc.
  • step 204 it is determined whether the detachable component is in a safe connection state based on the collected image; there are many implementation methods in practical applications.
  • the safe connection status of the detachable parts can be set in advance, and the collected images can be compared with the preset safe connection status of the detachable parts through image recognition, and based on the comparison results, it can be determined whether the detachable parts are in a safe connection status.
  • image recognition can be performed on one or more pixel areas in the image that represent the connection position of the detachable component and the movable platform to analyze whether the detachable component in the pixel area is in a safe connection state.
  • the collected images are matched with images that indicate that the detachable parts are in a safe connection state, and based on the image matching results, it can be determined whether the detachable parts are safely connected; or, it can also be implemented through a machine learning model, etc.
  • determining whether the detachable component is in a safe connection state based on the collected image includes any of the following: identifying whether the connection position of the movable platform in the image for connecting the detachable component is There are components connected; identify the type of the component connected to the connection position in the image, and determine whether the type of the recognized component matches the preset type; or, identify the component connected to the connection position in the image and the Connect the relative position relationship of the position and determine whether the recognized relative position relationship matches the preset relative position relationship.
  • one or more combinations of the above implementation methods may be used as needed. For example, if the detachable component is not installed, by identifying whether the connection position of the movable platform for connecting the detachable component in the image is connected with the component, if it is not installed, it can be determined that the detachable component is not in a safe connection. state. In some examples, the user may have installed the wrong detachable component. If it is determined that it is installed on the movable platform, the type of component connected to the connection position in the image can be further identified, and it can be determined whether the type of the recognized component is consistent with the preset one. Type match.
  • the user may also install the correct detachable parts, but the detachable parts are not installed in the correct position. Therefore, the user can also identify the parts connected to the connection position in the image and the connection position.
  • the relative position relationship is determined to determine whether the recognized relative position relationship matches the preset relative position relationship.
  • a rotorcraft is used as an example to illustrate an embodiment of determining whether the detachable parts are in a safe connection state based on the collected images.
  • the rotorcraft includes a fuselage, in which the arms are detachably connected to the fuselage, that is, the arms can be detached, and the arms are movably connected when connected to the drone, and the movably connected It means that the arms can be folded when the rotorcraft is not in operation, but the arms are unfolded during flight. Therefore, as the arm is a detachable part, whether it is in a safe state will affect the safe flight of the rotorcraft.
  • the propeller blades in the rotorcraft also affect the safe flight of the drone.
  • the number of rotors of rotor drones includes two rotors, four rotors, six rotors, eight rotors, etc.
  • a power motor drives a propeller to rotate.
  • the pitch, roll and yaw movements of the aircraft can be achieved.
  • the flight control system in the aircraft can adjust the speed of the four motors to change the propeller speed to achieve changes in lift, thereby controlling the attitude and position of the aircraft.
  • air resistance will form a counter-torque opposite to the direction of rotation.
  • propeller blades in rotorcraft are usually designed in pairs, including forward propeller blades and reverse propeller blades.
  • the forward propeller blades and reverse propeller blades need to be installed in the correct position in the drone, such as forward propeller blades.
  • the number of propeller blades is usually 2 or more, the length of the propeller is also different, and different drones are designed with different types of propellers.
  • the detachable parts include different types of propeller blades, wherein the connection positions of different types of propeller blades to the movable platform are different; and it is determined based on the collected images whether the detachable parts are in
  • the safe connection state includes: based on the collected images, determining whether the type of propeller blade connected to the connection position of the movable platform matches the preset propeller blade type corresponding to the connection position.
  • the movable platform can be connected to multiple propeller blades. Since propeller blades have different types, in this embodiment, each connection position between the movable platform and the propeller blades can be provided with a corresponding preset propeller. Based on this, by judging whether the type of propeller blade connected to the connection position of the movable platform matches the preset propeller blade type corresponding to the connection position, the movable platform can be accurately identified Whether each propeller blade is installed in the correct connection position.
  • the appearance information of the propeller blades at the connection position of the movable platform can be recognized through images to determine whether the propeller blades at the connection position are correct.
  • the type of propeller blades includes: characterization Appearance information about the appearance of the propeller blades.
  • the appearance information can be any one of the following: shape information, length information, blade number information, color information, material information or size information, etc. Based on this, through the identification of appearance information, it can be accurately It can be accurately judged whether the propeller blades connected to each connection position of the movable platform are correct.
  • machine learning models can also be used to identify whether detachable parts are securely connected.
  • determining whether the detachable component is in a safe connection state based on the collected image includes: using a machine learning model to determine whether the detachable component is in a safe connection state based on the collected image.
  • the training process of the model can be as follows: first represent a model through modeling, then evaluate the model by constructing an evaluation function, and finally optimize the evaluation function based on sample data and optimization methods to adjust the model to the optimum.
  • modeling is to transform actual problems into problems that computers can understand, that is, to transform actual problems into ways that computers can represent.
  • Modeling generally refers to the process of estimating the objective function of the model based on a large number of sample data.
  • evaluation is an indicator used to indicate the quality of the model. This will involve some evaluation indicators and the design of some evaluation functions. There will be targeted evaluation indicators in machine learning. For example, after the modeling is completed, a loss function needs to be designed for the model to evaluate the output error of the model.
  • the goal of optimization is the evaluation function. That is, the optimization method is used to optimize the evaluation function and find the model with the highest evaluation. For example, the minimum value of the output error of the loss function (optimal solution) can be found through optimization methods such as the gradient descent method, and the parameters of the model can be adjusted to the optimum.
  • model training can be supervised training, semi-supervised training or unsupervised training, etc.
  • a supervised training method can be used to increase the training speed, and the sample data can be marked with real values.
  • the machine learning model is trained through a sample image set, and the sample image set includes: A sample image in which the detachable parts are not in a safe connection state and a sample image marked with the detachable parts in a safe connection state. This embodiment can improve the speed and accuracy of model training through supervised training.
  • the machine learning model can be trained.
  • the machine learning model can be a neural network model, such as a convolutional neural network model, etc.
  • the structural design of the machine learning model can be flexibly implemented according to needs; among them, the structural design of the model is also an important aspect of the training process. , which will affect the prediction accuracy of the model.
  • the connection status of the detachable component and the movable platform has multiple categories
  • the machine learning model has multiple network branches, and each network branch is used to identify the detachable component respectively. Whether the connection status of different categories is the secure connection status, so one network branch is used to identify one connection status, so the model recognition accuracy can be improved.
  • each machine learning model corresponds to one type of detachable parts, and each machine learning model is used to use the image to identify the machine learning model. Whether the detachable parts corresponding to the model are in a safe connection state. Therefore, if there are multiple types of detachable parts of the movable platform, by designing a machine learning model for each type of part through this embodiment, the model recognition accuracy can be significantly improved.
  • a machine learning model can be pre-trained, and the machine learning model can be installed in the mobile platform or in a server connected to the mobile platform.
  • the machine learning model can be pre-trained by the business party, and the trained machine learning model can be stored in the removable platform, so that the removable platform can complete the identification of whether the detachable parts are in a safe connection state.
  • the movable platform can also send the collected images to the server, and the machine learning model configured on the server can identify from the image whether the detachable parts are in a safe connection state, and return the identification results to the movable platform.
  • the machine learning model can also be placed in other devices, and the other devices can recognize the image and determine whether the detachable parts of the movable platform are in a safe state.
  • prompt information can be output as needed. For example, if the detachable component is not in a safe connection state, a prompt message for instructing to reconnect the detachable component to the safe connection state is output, so that the user can notice that the detachable component is not in a safe connection state through the prompt information.
  • Connection Status For example, the prompt information may be used only to report an error to the user, that is, to prompt the user that the detachable component is not in a secure connection state; in other examples, the prompt information may also be used to indicate that the detachable component is not in a safe connection state. The component is reconnected from the current state to the safe connection state.
  • the prompt information may include the process of reconnecting the detachable component from the current state to the safe connection state, thereby instructing the user how to reconnect the detachable component from the current state. Connect to the safe connection state, thereby achieving better prompting effects and allowing users to reconnect detachable parts more efficiently.
  • the prompt information can be output in a variety of ways. For example, it can be output by a device that is communicatively connected to the movable platform, and/or it can be output by the movable platform.
  • the client connected to the mobile platform may include any device, such as a remote control, a smart phone, a wearable device, etc.
  • the movable platform in this embodiment takes the drone 110 as an example, and other devices that are connected to the drone 110 are explained in this embodiment by taking the remote control 210 and the smartphone 220 as examples.
  • the remote control 210 is provided with a display.
  • the remote control and the display can be detachably connected, and the display can also be fixedly mounted on the remote control.
  • Examples of other types of devices communicating with drones may include, but are not limited to, communication via: Internet, Local Area Network (LAN), Wide Area Network (WAN), Bluetooth, Near Field Communication (NFC) technology, based on technologies such as General Packet Radio Service (GPRS), GSM, Enhanced Data GSM Environment (EDGE), 3G, 4G or Long Term Evolution (LTE) protocols, infrared (IR) communications technology, and/or WiFi, and may be wireless, Wired type, or combination thereof.
  • GPRS General Packet Radio Service
  • GSM Global System for Mobile communications
  • EDGE Enhanced Data GSM Environment
  • 3G Third Generation
  • 4G Long Term Evolution
  • LTE Long Term Evolution
  • WiFi wireless, Wired type, or combination thereof.
  • the prompt information can be implemented in a variety of ways, for example, it can include image information, text information, video information, voice information or lighting information, etc.
  • the prompt information can be output in a variety of ways.
  • the output of prompt information includes any of the following: outputting prompt information on the user interface; and/or controlling one or more target components on the movable platform to output prompt information. For example, you can display images, text, and videos in the user interface of other devices, or control the playback component to play voice information, or control the lighting component to display lighting information, and so on.
  • the user interface displays a schematic image of the detachable component in a safe connection state.
  • a schematic image of reconnecting the detachable component from the current state to the safe connection state is displayed on the user interface.
  • Figure 2C it is a schematic diagram of outputting prompt information in this embodiment.
  • Figure 2C takes a smartphone as an example.
  • the screen of the smartphone displays the propeller blades in the drone from the current uninstalled state to the state connected to An indication of a safe connection on the drone's arm.
  • the prompt message adopts the text message "Dear, the propeller is not installed on this arm. Please see the picture below for the safe connection status. Please install it before flying.” This way, the user can be clearly guided on how to install the propeller through the above method.
  • the target component outputs prompt information including any of the following: the light-emitting component outputs light prompt information, the sound-generating component outputs sound information, or the motor vibration of the propeller outputs vibration information.
  • each arm of the drone is provided with a target component, and the prompt information can be output by the target component provided on the arm.
  • the prompter can include a sound-emitting component and/or a light-emitting component, and the sound can be emitted through the sound-emitting component.
  • the light-emitting component can emit light to prompt the user through the drone.
  • vibration information can also be output through the motor vibration of the propeller.
  • the flight control in the drone controls the rotation of the motor through electronic regulation, and the motor drives the propeller to rotate.
  • the rotation angle of the motor used to drive the propeller is large, the force of the blades pushing the air to move backward is large, and at the same time, they are pushed by the reaction force of the air, causing the rotorcraft to move.
  • the control of the propeller connected to the power motor to rotate at a preset angle may be performed after the rotorcraft stops moving, and the propeller may be controlled to rotate at a preset angle, and the preset angle may be at an angle at which the rotorcraft does not move.
  • the propeller rotation angle is smaller, and its force to push the air is smaller, so that the rotorcraft does not move.
  • the user can see one or more motors in the rotorcraft turning on the same arm as the detachable part that is not securely attached.
  • the rotation angle of the propeller is small, that is, the rotation angle of the power motor is small, so that the motor repeatedly rotates at a smaller rotation angle, causing the motor's rotor to vibrate, so the motor can vibrate and make sounds, so that the user can hear
  • the sound emitted by the motor reminds the user that the arm or propeller where the motor making the sound is located is not in a safe connection state.
  • the detachable component includes a machine arm
  • the target component for outputting prompt information includes: a component on the machine arm that is not in a safe connection state.
  • the detachable components include propeller blades, which are installed on the machine arm
  • the target components for outputting prompt information include: propeller blades that are not in a safe connection state are installed on the machine arm. parts. Based on this, the user can know which arm or propeller on the drone is not in a safe connection state, thereby achieving a better prompt effect.
  • the method may further include: if it is determined that the detachable component is in a secure connection state based on the image, prompting the user that the movable platform can perform a moving operation.
  • the user can be prompted that the movable platform can perform a mobile operation to improve the user's experience.
  • the movable platform is a four-rotor UAV as an example, and the detachable parts in the UAV are an arm and a propeller as an example; the UAV in this embodiment has one or more visual sensors.
  • the visual sensor is used to collect environmental observation data around the drone.
  • the drone can use the environmental observation data for flight control.
  • the collected environmental observation data can also be used for more flight tasks, such as storing the environment.
  • the observation data can be used for later data processing and can also be transmitted in real time to other devices that can communicate with the drone, such as other drones, ground stations, remote control devices or user's portable devices, etc.
  • the drone's visual sensor can be used to collect images of the drone itself.
  • the detachable parts are used as the collection target, and images containing the detachable parts are collected; the detachable parts are identified based on the images.
  • the connection status of the component is implemented using a machine learning model.
  • the process of image collection in this embodiment can be either the visual sensor of the drone collecting images, or the drone manually capturing images.
  • the left and right binocular images of the drone can be collected, and four images can be obtained, corresponding to the left eye of the drone, the left eye of the left eye, the right eye of the right eye, and the right eye of the right eye.
  • other embodiments are also optional, for example, a top view of the entire drone can be collected from the drone, and so on.
  • image data of the arms can be collected.
  • images of the four arms in normal deployment and all other abnormal deployment states can be collected.
  • image data can be collected with and without blades installed, as well as image data with multiple different blade models; as an example, images can be collected with multiple different types of blades installed on a drone, as well as with unmanned aerial vehicles. Images are collected after the machine is installed with matching propellers; optionally, four images can be collected simultaneously in timestamp units; of course, the number of images in actual applications can be flexibly configured as needed.
  • image data representing blades with different installation methods can also be collected.
  • images of blades with different installation methods and images of correctly installed blades can be collected.
  • Four images can be collected at one time based on timestamps.
  • the number of images in actual applications can be flexibly configured as needed.
  • the image data can cover as many scenarios as needed, that is, the drone is collected in different take-off scenarios.
  • Image data such as asphalt road, lawn, beach, gravel road and other take-off scenarios.
  • this embodiment uses a supervised training method.
  • other training methods such as unsupervised training can be used as needed.
  • the image data of the machine arm can be marked.
  • Each image is marked with whether the machine arm is in a normal state, which is a two-class classification task. It includes all the machine arm images of four pictures: left-view left eye, left-view right eye, right-view left eye and right-view right eye.
  • FIG 3A it is a schematic diagram of a machine learning model in an embodiment of the present application; as an example, in order to meet the requirement of being able to run in real time on an embedded device, the convolution layer uses a convolution kernel size of 3x3 and a step size of 2. For example, this can gradually reduce the size of the feature map and reduce the amount of calculation.
  • the convolution layer uses a convolution kernel size of 3x3 and a step size of 2. For example, this can gradually reduce the size of the feature map and reduce the amount of calculation.
  • the structure can be seen in Figure 3A.
  • the detachable components are the arm and the propeller as an example. Therefore, two network models are constructed as an example. One is called the arm deployment recognition model, and the other is called the blade recognition model.
  • the arm expansion recognition model can use a single image as input, or images from multiple perspectives, such as inputting images from four directions respectively, to determine whether each arm in the drone is deployed normally.
  • the blade recognition model can take 4 images with the same timestamp as input together; this embodiment is based on the three categories of blade connection status.
  • the blade recognition model is divided into three branches in the middle, one of which determines whether the blade is For installation, one branch determines whether the blade model is correct, and the other branch classifies which category the blade installation status belongs to.
  • the two models can be trained separately.
  • the input is a combination of four images.
  • the annotation results include whether the blade is installed, whether the blade model is correct, and which category the blade installation belongs to.
  • the network is divided into three branches, one of which It is used to determine whether the blade model is correct.
  • the other branch is used to determine whether the blade is installed. It is a two-category task.
  • the other branch is used to identify whether the blade is installed correctly or incorrectly. It is a multi-category task.
  • Two Both branches can use the cross-entropy loss function to jointly supervise the training of the entire blade recognition model.
  • Loss loss1+a*loss2+a*b*loss3
  • loss1 is the loss to determine whether the blade is installed
  • loss2 is the loss to determine whether the blade model is correct
  • loss3 is the loss to classify the blade installation.
  • a represents whether the propeller is installed (1 represents installed, 0 represents not installed)
  • b represents whether the propeller model is correct (1 represents the correct model, 0 represents incorrect).
  • the blade installation methods can be classified into multiple categories; the blades in this embodiment are divided into forward propellers and reverse propellers, and the forward propellers and reverse propellers need to be installed in the correct position.
  • FIG. 3B it is a schematic diagram of six blade installation methods of a quad-rotor UAV according to an exemplary embodiment of the present application. Wherein, Correction costs.
  • the arm deployment recognition model and the blade recognition model are obtained at the end of the training.
  • the models can be set on the movable platform or on the server.
  • the movable platform controls the visual sensor to collect images of the machine arm and propeller, and the two models identify whether the machine arm and propeller are in a safe state.
  • the recognition can be performed before the UAV is ready to take off.
  • the left-viewing left-viewing, left-viewing right-viewing, right-viewing left-viewing, and right-viewing right-viewing images are respectively input into the arm deployment recognition model.
  • the arm deployment model recognizes the four images respectively.
  • the recognition result of image recognition is whether the arm is unfolded. Based on the perspective of the collected image, each image corresponds to one arm of the drone.
  • the corresponding The position of the machine arm can be determined accurately, so that prompt information can be output, for example, to remind the user which arm needs to be unfolded, and to remind the user of the corresponding arm by flashing the signal light set on the corresponding arm.
  • the four images with the same timestamp are merged and input into the blade recognition model.
  • the recognition results of whether the blades are installed can be obtained. If the blades are not installed, the recognition results can be obtained through The application reminds the user to install it. If the blade has been installed, it can identify whether the paddle type is correct. If the type is incorrect, the application can remind the user that the paddle type is incorrect. If the paddle type is correct, the paddle can be identified. According to the classification results of the installation, if the installation is correct, it can be determined that the arms and blades are in a safe connection state, so it is determined that the drone is ready to take off.
  • Figure 3C is a schematic diagram of the user interface outputting prompt information according to an exemplary embodiment of the present application.
  • the blade recognition model It is recognized that the installation positions of the forward and reverse propellers are incorrectly installed in Category 1 as shown in Figure 3B. Therefore, the prompt method shown in Figure 3C can be given, in which the image on the left in Figure 3C guides the user to install the two propellers. Reverse the positions. The image on the right shows the correct installation position of the forward and reverse propellers in the drone.
  • the execution subject for implementing this embodiment is not limited to the movable platform, and may also be other electronic devices, such as a device with a display screen. Remote control device, smart phone, personal digital assistant, tablet computer, multimedia device, wearable device or personal computer, etc. It is not necessary whether the electronic device is connected to the mobile platform.
  • this application illustrates an information prompting method according to an exemplary embodiment.
  • the method includes:
  • step 402 images are collected of propeller blades of the movable platform; the movable platform includes propeller blades of different types, and the propeller blades are detachably connected to the movable platform; wherein different types of propeller blades are detachably connected to the movable platform.
  • the connection positions of the propeller blades and the movable platform are different;
  • step 402 based on the collected image, determine whether the type of propeller blade connected to the connection position of the movable platform matches the preset propeller blade type corresponding to the connection position;
  • step 402 if there is no match, prompt information is output, and the prompt information is used to instruct the connection position to be reconnected to the propeller blade.
  • the movable platform includes different types of propeller blades, and the propeller blades are detachably connected to the movable platform; wherein the connection positions of different types of propeller blades to the movable platform are different.
  • FIG. 5 it is a schematic diagram of another information prompting method according to an exemplary embodiment of the present application.
  • the method includes:
  • step 502 images are collected from the detachable parts of the movable platform
  • step 504 determine whether the detachable component is in a safe connection state based on the collected image
  • step 506 if the detachable component is not in the safe connection state, prompt information is output, and the prompt information is used to instruct the reconnection of the detachable component from the current state to the safe connection state.
  • the above method embodiments can be implemented by software, or can be implemented by hardware or a combination of software and hardware.
  • a device in a logical sense it is formed by reading the corresponding computer program instructions in the non-volatile memory into the memory and running it through the image processing processor where it is located.
  • Figure 6 is a structural diagram of a movable platform control device 600 according to an embodiment of the present application.
  • the movable platform control device 600 includes a processor 601, a memory 602, and A computer program on the memory that can be executed by the processor. When the processor executes the computer program, the following methods are implemented:
  • the movable platform includes at least one visual sensor, and at least one detachable component is used to be mounted on the movable platform;
  • the detachable component If the detachable component is not in the safe connection state, output prompt information, the prompt information is used to instruct the reconnection of the detachable component to the safe connection state;
  • the movable platform After determining that the detachable component is in a safe connection state based on the image, the movable platform is controlled to move in space based on the environmental image collected by the visual sensor.
  • the prompt information is used to indicate reconnecting the detachable component from the current state to the safe connection state.
  • determining whether the detachable component is in a safe connection state based on the collected images includes any of the following:
  • the detachable parts include different types of propeller blades, wherein different types of propeller blades are connected to the movable platform at different locations;
  • Determining whether the detachable component is in a safe connection state based on the collected image includes:
  • the type of the propeller blade includes: appearance information that characterizes the appearance of the propeller blade, and the appearance information is any of the following: shape information, length information, blade number information, color information, material information, or size. information.
  • the detachable component includes an arm
  • Determining whether the detachable component is in a safe connection state based on the collected image includes:
  • the step of collecting images of the detachable component of the movable platform based on the visual sensor the step of collecting images of the detachable component of the movable platform based on the visual sensor
  • the visual sensor is fixedly disposed on the movable platform, and the field of view of the visual sensor covers the connection position of the movable platform for connecting the detachable components.
  • the vision sensor is mounted on the movable platform through a pan/tilt
  • the image is collected by controlling the pan/tilt to make the visual sensor face the connection position of the movable platform for connecting the detachable component.
  • determining whether the detachable component is in a safe connection state based on the collected image includes:
  • the machine learning model determines whether the detachable component is in a safe connection state based on the collected images.
  • each machine learning model corresponds to one type of detachable parts, and each machine learning model is used to use the image to identify the machine learning model. Whether the detachable parts corresponding to the model are in a safe connection state.
  • connection status of the detachable component and the movable platform has multiple categories
  • the machine learning model has multiple network branches, and each network branch is used to identify different characteristics of the detachable component. Whether the connection status of the category is the secure connection status.
  • the machine learning model is trained through a sample image set.
  • the sample image set includes: sample images marked that the detachable component is not in a safe connection state and sample images marked that the detachable component is in a safe connection state. Sample image of status.
  • the output prompt information includes any of the following:
  • the user interface includes: user interfaces of other devices communicatively connected to the removable platform.
  • the user interface displays a schematic image of the detachable component in a securely connected state.
  • a schematic image of reconnecting the detachable component from the current state to the secure connection state is displayed on the user interface.
  • the target component outputs prompt information: including any of the following:
  • the lighting component outputs light prompt information, the speaker outputs sound information, or the propeller motor vibration outputs vibration information.
  • the detachable component includes a machine arm
  • the target component for outputting prompt information includes: a component on the machine arm that is not in a safe connection state.
  • the detachable component includes a propeller blade installed on a machine arm
  • the target component for outputting prompt information includes: a machine arm on which a propeller blade that is not in a safe connection state is installed. parts on.
  • the method further includes:
  • the user is prompted that the movable platform can perform a moving operation.
  • the device embodiment since it basically corresponds to the method embodiment, please refer to the partial description of the method embodiment for relevant details.
  • the device embodiments described above are only illustrative.
  • the units described as separate components may or may not be physically separated.
  • the components shown as units may or may not be physical units, that is, they may be located in One location, or it can be distributed across multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application. Persons of ordinary skill in the art can understand and implement the method without any creative effort.
  • control device of the movable platform of the present application may also include memory, network interface, non-volatile memory, etc.
  • the movable platform where the device is located in the embodiment is usually based on the actual function of the movable platform, and may also be Including other hardware, I won’t go into details about this.
  • this embodiment also provides an information prompting device 700.
  • the device 700 includes a processor 701, a memory 702, and a computer program stored on the memory 702 that can be executed by the processor 701.
  • the processor executes the computer program, the aforementioned embodiment of the information prompting method is implemented.
  • a prompt message indicating adjustment of the blade installation status is output.
  • the rotorcraft includes one or more image sensors, the observation range of the one or more image sensors covers multiple rotors of the rotorcraft, and the image data is based on the one or more image sensors. collected by the image sensor.
  • the observation range of the image sensor simultaneously covers the surrounding environment of the rotorcraft, and the image sensor is used to observe the surrounding environment when the rotor is in a flight state, so that multiple The rotorcraft adjusts one or more rotor operating parameters based on the observed information.
  • the prompt information includes image prompt information displayed on a ground controller of the rotorcraft;
  • the image prompt information includes: rotor identification indicating a plurality of rotors, and a first blade identification indicating a preset blade model adapted to the rotor.
  • the image prompt information also includes: a second blade identification of the currently installed blade model of any of the rotors.
  • the image prompt information includes: an adjustment guide identifier indicating an association between the first blade identifier of a preset blade model adapted to the rotor and the corresponding rotor.
  • the image prompt information includes: a fuselage identification indicating the fuselage of the rotorcraft;
  • the display position relationship between the rotor logo and the fuselage logo is consistent with the position distribution of the rotor on the fuselage of the rotorcraft.
  • processor when the processor executes the computer program, it may also implement:
  • prompt information is output, and the prompt information is used to instruct the connection position to be reconnected to the propeller blade.
  • processor when the processor executes the computer program, it may also implement:
  • prompt information is output, and the prompt information is used to instruct the reconnection of the detachable component from the current state to the safe connection state.
  • this embodiment also provides a movable platform 800.
  • the movable platform 800 includes at least one visual sensor 803, and at least one detachable component 804 for connecting with the movable platform;
  • the mobile platform also includes a processor 801, a memory, and a computer program stored on the memory 802 that can be executed by the processor 801;
  • the processor executes the computer program, it can implement:
  • a prompt message indicating adjustment of the blade installation status is output.
  • the detachable component If the detachable component is not in the safe connection state, output prompt information, the prompt information is used to instruct the reconnection of the detachable component to the safe connection state;
  • the movable platform After determining that the detachable component is in a safe connection state based on the image, the movable platform is controlled to move in space based on the environmental image collected by the visual sensor.
  • the collected image determine whether the type of propeller blade connected to the connection position of the movable platform matches the preset propeller blade type corresponding to the connection position;
  • prompt information is output, and the prompt information is used to instruct the connection position to be reconnected to the propeller blade.
  • prompt information is output, and the prompt information is used to instruct the reconnection of the detachable component from the current state to the safe connection state.
  • the electronic device 900 includes a processor 901, a memory 902, and a computer program stored on the memory 902 that can be executed by the processor;
  • the processor executes the computer program, it can implement:
  • a prompt message indicating adjustment of the blade installation status is output.
  • the detachable component If the detachable component is not in the safe connection state, output prompt information, the prompt information is used to instruct the reconnection of the detachable component to the safe connection state;
  • the movable platform After determining that the detachable component is in a safe connection state based on the image, the movable platform is controlled to move in space based on the environmental image collected by the visual sensor.
  • prompt information is output, and the prompt information is used to instruct the connection position to be reconnected to the propeller blade.
  • prompt information is output, and the prompt information is used to instruct the reconnection of the detachable component from the current state to the safe connection state.
  • This embodiment also provides a computer-readable storage medium.
  • Several computer instructions are stored on the computer-readable storage medium. When the computer instructions are executed, the embodiments of the control method of the movable platform and/or the Embodiment of information prompting method.
  • Embodiments of the present description may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having program code embodied therein.
  • Storage media available for computers include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information.
  • Information may be computer-readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, Magnetic tape cassettes, tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium can be used to store information that can be accessed by a computing device.
  • PRAM phase change memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • RAM random access memory
  • ROM read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • flash memory or other memory technology
  • CD-ROM compact disc read-only memory
  • DVD digital versatile disc
  • Magnetic tape cassettes tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium can be used to store information that can be accessed by
  • the device embodiment since it basically corresponds to the method embodiment, please refer to the partial description of the method embodiment for relevant details.
  • the device embodiments described above are only illustrative.
  • the units described as separate components may or may not be physically separated.
  • the components shown as units may or may not be physical units, that is, they may be located in One location, or it can be distributed across multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Persons of ordinary skill in the art can understand and implement the method without any creative effort.

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Transportation (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

本申请提供一种可移动平台及其控制方法、信息提示方法、装置、可移动平台、电子设备及计算机可读存储介质，其中，旋翼飞行器的多个旋翼包括可拆卸的桨叶，多个所述旋翼包括适配不同桨叶型号的不同旋翼，该信息提示方法包括：获取多个所述旋翼的预设桨叶型号(1110)；基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号(1112)；若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息(1114)。

Description

可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质 技术领域
本申请涉及可移动平台技术领域，具体而言，涉及一种可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质。
背景技术
目前，可移动平台广泛用于航拍、农业、植保、运输、测绘、救灾等领域。可移动平台可自动地或在用户控制下在空间中移动，如何保证可移动平台的安全移动，是本领域一直关注的技术问题。
发明内容
有鉴于此，本申请提供一种可移动平台的控制方法、信息提示方法、装置、可移动平台、电子设备及计算机可读存储介质，以解决相关技术中可移动平台的安全移动问题。
第一方面，提供一种旋翼飞行器的信息提示方法，所述旋翼飞行器的多个旋翼包括可拆卸的桨叶，多个所述旋翼包括适配不同桨叶型号的不同旋翼；
所述方法包括：
获取多个所述旋翼的预设桨叶型号；
基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号；
若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息。
第二方面，提供一种可移动平台的控制方法，所述可移动平台包括至少一个视觉传感器，至少一个可拆卸部件用于与所述可移动平台连接；所述方法包括：
基于所述视觉传感器所述可移动平台的所述可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件重新连接至所述安全连接状态；
在基于所述图像确定所述可拆卸部件处于安全连接状态下，基于所述视觉传感器采集的环境图像控制所述可移动平台在空间中运动。
第三方面，提供一种可移动平台的信息提示方法，所述可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接；其中，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；所述方法包括：
对所述可移动平台的所述螺旋桨桨叶采集图像；
根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配；
若未匹配，输出提示信息，所述提示信息用于指示将所述连接位置与螺旋桨桨叶 重新连接。
第四方面，提供一种可移动平台的信息提示方法，所述可移动平台包括至少一个可拆卸部件；所述方法包括：
对所述可移动平台的所述可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
第五方面，提供一种可移动平台的控制装置，所述装置包括处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现第一方面所述的方法。
第六方面，提供一种信息提示装置，所述装置包括处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现第一方面、第三方面或第四方面所述的方法。
第七方面，提供一种可移动平台，所述可移动平台包括至少一个视觉传感器，至少一个可拆卸部件用于与所述可移动平台连接；所述可移动平台还包括处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序；
所述处理器执行所述计算机程序时实现第一方面至第四方面任一所述的方法。
第八方面，提供一种电子设备，所述处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序；
所述处理器执行所述计算机程序时实现第一方面至第四方面任一所述的方法。
第九方面，提供一种计算机可读存储介质，所述计算机可读存储介质上存储有若干计算机指令，所述计算机指令被执行时实现第一方面至第四方面任一所述方法的步骤。
应用本申请提供的可移动平台的控制方案，可移动平台可以自动检测可拆卸部件是否处于安全连接状态，无需用户额外操作，该方案是基于可移动平台的视觉传感器对可拆卸部件采集的图像，来识别可拆卸部件是否处于安全连接状态，因此不需要额外硬件成本；由于可以输出用于指示将所述可拆卸部件重新连接至所述安全连接状态的提示信息，因此用户及时发现可拆卸部件未处于安全连接状态，进一步保证可移动平台的安全移动。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1A是本申请一个实施例的无人飞行系统的示意性架构图。
图1B是本申请一个实施例的一种四旋翼无人机处于收纳状态的示意图。
图1C是本申请一个实施例的一种四旋翼无人机处于可工作状态的示意图。
图1D是本申请一个实施例的旋翼飞行器的信息提示方法的流程示意图。
图2A是本申请一个实施例的可移动平台的控制方法的流程示意图。
图2B是本申请一个实施例的无人机与其他设备的通信示意图。
图2C是本申请一个实施例的一种输出提示信息的示意图。
图3A是本申请一个实施例的一种机器学习模型的示意图。
图3B是本申请一个实施例的四旋翼无人机的六种桨叶安装方式的示意图。
图3C是本申请一个实施例的另一种输出提示信息的示意图。
图4是本申请一个实施例的信息提示方法的流程示意图。
图5是本申请另一个实施例的信息提示方法的流程示意图。
图6是本申请一个实施例的可移动平台的控制装置的结构图。
图7是本申请一个实施例的信息提示装置的结构图。
图8是本申请一个实施例的可移动平台的结构图。
图9是本申请一个实施例的电子设备的结构图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。
本申请实施例的可移动平台可以包括多种具有移动能力的设备，例如无人机、车辆、机器人、清洁设备或无人船等等。其中，可移动平台可以连接多种部件。
以无人机场景为例进行说明，图1A是本申请实施例的无人飞行系统的示意性架构图，无人飞行系统100可以包括无人机110、显示设备130和遥控设备140。其中，无人机110可以包括动力系统150、飞行控制系统160(简称飞控)、机架和承载在机架上的云台120；无人机110可以包括旋翼无人机。无人机110可以与遥控设备140和显示设备130进行无线通信。
机架可以包括机身和脚架(也称为起落架)。机身可以包括中心架以及与中心架连接的一个或多个机臂，一个或多个机臂呈辐射状从中心架延伸出。脚架与机身连接，用于在无人机110着陆时起支撑作用。
动力系统150可以包括一个或多个电子调速器(简称为电调)151、一个或多个螺旋桨153以及与一个或多个螺旋桨153相对应的一个或多个动力电机152，其中动力电机152连接在电子调速器151与螺旋桨153之间，动力电机152和螺旋桨153可设置在无人机110的机臂上，例如机臂远离机身的一端；电子调速器151用于接收飞行控制系统160产生的驱动信号，并根据驱动信号提供驱动电流给动力电机152，以控制动力电机152的转速。动力电机152用于驱动螺旋桨旋转，从而为无人机110的飞行提供动力，该动力使得无人机110能够实现一个或多个自由度的运动。在某些实施例中，无人机110可以围绕一个或多个旋转轴旋转。例如，上述旋转轴可以包括横滚轴(Roll)、偏航轴(Yaw)和俯仰轴(pitch)。应理解，电机152可以是直流电机，也可以交流电机。另外，电机152可以是无刷电机，也可以是有刷电机。
飞行控制系统160可以包括飞行控制器161和传感系统162。传感系统162用于测量无人机的姿态信息，即无人机110在空间的位置信息和状态信息，例如，三维位置、三维角度、三维速度、三维加速度和三维角速度等；传感系统162还可采集其他信息，例如定位信息，或者无人机所在空间中景物的信息，例如深度信息或热度信息等等。传感系统162例如可以包括陀螺仪、超声传感器、电子罗盘、惯性测量单元 (Inertial Measurement Unit，IMU)、视觉传感器、热度仪、全球导航卫星系统和气压计等等传感器中的至少一种。例如，全球导航卫星系统可以是全球定位系统(Global Positioning System，GPS)。飞行控制器161用于控制无人机110的飞行，例如，可以根据传感系统162测量的姿态信息控制无人机110的飞行。应理解，飞行控制器161可以按照预先编好的程序指令对无人机110进行控制，也可以通过响应来自遥控设备140的一个或多个遥控信号对无人机110进行控制。
云台120可以包括电机122。云台用于携带如拍摄装置123等多种装置。飞行控制器161可以通过电机122控制云台120的运动。可选的，作为另一实施例，云台120还可以包括控制器，用于通过控制电机122来控制云台120的运动。应理解，云台120可以独立于无人机110，也可以为无人机110的一部分。应理解，电机122可以是直流电机，也可以是交流电机。另外，电机122可以是无刷电机，也可以是有刷电机。还应理解，云台的位置有多种实现方式，例如可以位于无人机的顶部，也可以位于无人机的底部等等。
拍摄装置123例如可以是照相机或摄像机等用于捕获图像的设备，拍摄装置123可以与飞行控制器通信，并在飞行控制器的控制下进行拍摄。本实施例的拍摄装置123至少包括感光元件，该感光元件例如为互补金属氧化物半导体(Complementary Metal Oxide Semiconductor，CMOS)传感器或电荷耦合元件(Charge-coupled Device，CCD)传感器。可以理解，拍摄装置123也可直接固定于无人机110上，从而云台120可以省略。
显示设备130位于无人飞行系统100的地面端，可以通过无线方式与无人机110进行通信，并且可以用于显示无人机110的姿态信息。另外，还可以在显示设备130上显示拍摄装置123拍摄的图像。应理解，显示设备130可以是独立的设备，也可以集成在遥控设备140中。
遥控设备140位于无人飞行系统100的地面端，可以通过无线方式与无人机110进行通信，用于对无人机110进行远程操纵。
应理解，上述对于无人飞行系统各组成部分的命名仅是出于标识的目的，并不应理解为对本申请的实施例的限制。
如前所述，如无人机等可移动平台可以连接多种部件。一些部件的连接状态可能会影响到可移动平台的安全移动。例如，一些可拆卸部件，由于这些部件的可拆卸特点，出于多种原因，例如用户拆卸、用户错误安装或部件松动等，导致这些部件与可移动平台具有多种不同连接状态，而一些连接状态可能影响到可移动平台的安全移动。
仍以无人机为例进行说明，如图1B所示，是本申请根据一示例性实施例示出的一种四旋翼无人机的示意图，该图1B中无人机110处于非工作的收纳状态，图1B无人机包括机身170和四个机臂，以机臂171为例，该机臂171与机身170可折叠连接，该机臂171并未展开；图1B中的无人机110还包括四个螺旋桨，以螺旋桨153为例，该螺旋桨153的桨叶也是可折叠的，该螺旋桨153的桨叶也未展开。
如图1C所示，是本申请根据一示例性实施例示出的一种四旋翼无人机的另一示意图，图1C中无人机110处于可工作状态，四个机臂及四个螺旋桨均处于展开状态。
可拆卸部件可以包括与可移动平台移动有关的部件、与可移动平台正常执行任务有关的部件。由于可拆卸部件的可拆卸特点，可拆卸部件与可移动平台可能具有多种 连接状态，只有可拆卸部件在安全连接状态下，才可让可移动平台处于正常工作状态，才可保证可移动平台的安全移动。相关技术中，通常是由可移动平台的厂商提供产品的使用说明书，由用户阅读使用说明书了解可拆卸部件的安全连接状态，然后由用户操作可拆卸部件，将可拆卸部件操作至安全连接状态。此种方式需要用户自行查阅使用说明书，给用户带来较大麻烦；而且，可能出现用户未能将可拆卸部件操作至安全连接状态的情况，例如用户忘记可拆卸部件的安全连接状态、用户操作不到位、或者用户将可移动平台交给其他不熟悉可移动平台的用户使用等等；另外，可移动平台可能连接多个可拆卸部件，这多个可拆卸部件外观上相似度较大，用户难以区分，但又要求连接在可移动平台的不同位置；这些情况都可能造成可拆卸部件未能处于安全连接状态，因此会影响到可移动平台后续的工作，给可移动平台的移动带来较大的安全隐患。
基于此，如图1D所示，是本说明书根据一示例性实施例示出的一种旋翼飞行器的信息提示方法，所述旋翼飞行器的多个旋翼包括可拆卸的桨叶，多个所述旋翼包括适配不同桨叶型号的不同旋翼；所述方法包括：
在步骤1110中，获取多个所述旋翼的预设桨叶型号；
在步骤1112中，基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号；
在步骤1114中，若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息。
本实施例方案可由旋翼飞行器或与所述旋翼飞行器连接的地面控制器执行，其中，图像数据的采集可以是旋翼飞行器上的图像传感器，或者是与旋翼飞行器连接的地面控制器的图像传感器，执行该实施例的旋翼飞行器或电子设备可以获取到上述任一图像传感器采集的图像数据。
旋翼无人机的旋翼数量包括二旋翼、四旋翼、六旋翼、八旋翼等等。通常情况下，一个动力电机驱动一个螺旋桨旋转，通过多个螺旋桨的配合，可以实现飞机的俯仰、横滚和偏航的动作。在一些例子中，以四旋翼为例，飞机中的飞行控制系统可以调节四个电机转速来改变螺旋桨转速，实现升力的变化，从而控制飞机的姿态和位置。其中，螺旋桨转动过程中由于空气阻力作用会形成与转动方向相反的反扭矩，为了克服反扭矩影响，可使四个螺旋桨中的两个正转，两个反转，且对角线上的各个旋翼转动方向相同。因此，旋翼飞机中螺旋桨桨叶通常是成对设计，包括正桨桨叶和反桨桨叶，正桨桨叶和反桨桨叶需要安装在无人机中正确的位置，例如正桨桨叶和反桨桨叶设置在同一对角线上。可选的，螺旋桨的桨叶数量通常是2个或更多，螺旋桨的长度也不尽相同，不同无人机设计有不同类型的螺旋桨。
因此，本实施例中每个旋翼可以预先设置有对应的预设桨叶型号，基于此，通过图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号，若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息，从而对用户进行提示，使用户可以调整桨叶安装状态，以保证旋翼飞行器的安全移动。
桨叶型号可以指示不同的桨叶构型。例如，多旋翼飞机的正桨或者反桨(正常安装的情况下正桨与反桨互为反向旋转以部分抵消水平方向上的扭矩)。
还可以指示适配不同飞行条件下的桨叶构型，例如，平原桨叶、静音桨叶、高原 桨叶等等。高原桨叶相较于平原桨叶，桨叶长度可能更长，以在相同转速下提供更高的升力，同时单位时间的功耗更大。静音桨叶可能与平原桨叶在尺寸上差别不大，但是桨叶末端做了特殊的材质、形状处理，降低了桨叶旋转切割空气引起的风噪。
在一般的多旋翼配件中，为了保证飞行安全，满足用户不同的飞行任务需求，在多旋翼飞行器的包装中往往配置有多种不同型号的桨叶。
桨叶与对应的旋翼可拆装连接，用户可以手动拆卸或者安装桨叶。在面对多个旋翼可能适配不同的桨叶型号的情况下，人工操作难免会引入装配差错。例如，在机身上同时安装了高原桨叶和平原桨叶，由于桨叶产生的升力不一致，会严重的影响飞行器的飞行安全。再比如，在机身上同时安装了平原桨叶和静音桨叶，这种安装可能对飞行器的稳定影响可控，但是对降噪来说可能无法满足用户作业需求。
因此，给飞行器适配满足要求的桨叶型号是至关重要的。
在一些例子中，所述桨叶型号可以包括表征所述桨叶外观的外观信息等，所述外观信息如下任一：形状信息、长度信息、叶片数量信息、颜色信息、材质信息、或大小信息等等，基于此，通过外观信息的识别，可以准确地判断出每个旋翼所连接的桨叶是否正确。
在一些例子中，所述旋翼飞行器包括一个或者多个图像传感器，一个或者多个图像传感器的观测范围覆盖所述旋翼飞行器的多个所述旋翼，所述图像数据是基于所述一个或者多个图像传感器采集得到的。本实施例中，可以利用旋翼飞行器上搭载的图像传感器采集数据。
在一些例子中，所述图像传感器的所述观测范围同时覆盖所述旋翼飞行器的周围环境，所述图像传感器用于在所述旋翼处于飞行状态时对所述周围环境进行观测，以使多个所述旋翼飞行器基于所述观测得到的信息调整一个或者多个所述旋翼运行参数。本实施例中，可以利用旋翼飞行器已有的用于避障的视觉传感器进行图像采集，通过复用视觉传感器可以减少成本。
在一些例子中，所述提示信息包括在所述旋翼飞行器的地面控制器上显示的图像提示信息；所述图像提示信息包括：指示多个所述旋翼的旋翼标识，以及指示所述旋翼适配的预设桨叶型号的第一桨叶标识。本实施例中，所述旋翼飞行器的地面控制器可以包括移动终端或遥控器等，地面控制器包括显示器，地面控制器可以显示图像提示信息，由于图像提示信息包括指示多个所述旋翼的旋翼标识以及指示所述旋翼适配的预设桨叶型号的第一桨叶标识，从而使用户直观地查阅到每个旋翼及其适配的预设桨叶型号，使用户知道该如何正确地安装旋翼。
在一些例子中，所述图像提示信息还可包括：任一所述旋翼的当前安装的桨叶型号的第二桨叶标识，基于此，可以使用户直观地查阅到旋翼飞行器上当前哪个旋翼安装了错误的桨叶，使用户可以清楚对哪个旋翼重新安装适配的桨叶。
在一些例子中，所述图像提示信息包括：指示所述旋翼适配的预设桨叶型号的所述第一桨叶标识与对应的所述旋翼关联关系的调整指引标识，基于此，针对安装错误的旋翼，该调整指引标识能够指示用户该旋翼适配的预设桨叶型号，使用户快速正确地对该旋翼的桨叶进行调整。
在一些例子中，所述图像提示信息包括：指示所述旋翼飞行器的机身的机身标识；所述旋翼标识与所述机身标识显示的显示位置关系，与所述旋翼在所述旋翼飞行器的 机身的位置分布一致，本实施例中可以按照旋翼飞行器与旋翼的实际位置关系进行图像显示，使用户可以通过图像提示信息，对照旋翼飞行器中的各个旋翼，便于用户进行桨叶的调整。
相对应的，本说明书还提供了一种可移动平台的控制方法的实施例，如图2A所示，是该可移动平台的控制方法的流程图，包括如下步骤：
在步骤202中、基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像。
在步骤204中、根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态。
在步骤206中、若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件重新连接至所述安全连接状态。
在步骤208中、在基于所述图像确定所述可拆卸部件处于安全连接状态下，基于所述视觉传感器采集的环境图像控制所述可移动平台在空间中运动。
本实施例中，可以由可移动平台自动检测可拆卸部件是否处于安全连接状态，无需用户额外操作，该方案是基于可移动平台的视觉传感器对可拆卸部件采集的图像，来识别可拆卸部件是否处于安全连接状态，因此不需要额外硬件成本；由于可以输出用于指示将所述可拆卸部件重新连接至所述安全连接状态的提示信息，因此可以提醒用户，进一步保证可移动平台的安全移动。
本实施例方案的执行时机，实际应用中可根据需要设定。示例性的，所述基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像的步骤，是基于与所述可移动平台通信的其他设备发送的消息执行的；和/或，是基于可移动平台的移动状态信息执行的；和/或，是基于所述可移动平台开机后产生的触发信号执行的。
例如，可以是可移动平台执行作业任务开始之前执行，例如可以在检测到作业任务开始后生成触发信号，以自动触发执行本实施例方案。或者，也可以是基于与所述可移动平台通信的其他设备发送的消息执行的，该消息可以包括指示可移动平台开始作业的消息，可移动平台通过该消息确定用户指示可移动平台开始作业后，执行本实施例方案。或者，还可以基于可移动平台的移动状态信息执行，例如，本实施例方案的执行可以是在可移动平台未处于移动状态下执行的，所述移动状态信息可以通过可移动平台的惯性测量传感器采集的数据而确定到。或者，是基于所述可移动平台开机后产生的触发信号执行的，该触发信号可以是可移动平台检测到某些物理按键被触发后产生，例如开机键，产生触发信号的方式可以是每次开机键被触发后产生，也可以是按照设定周期后产生。或者，还可以结合其他检测方式确定本实施例方案的执行时机，例如可移动平台可以检测到处于某些场景下才执行等等，例如，可以通过视觉传感器、激光雷达或测距传感器等采集可移动平台的环境观测数据，通过环境观测数据检测到可移动平台处于安全环境下等。
本实施例中，可移动平台包括至少一个视觉传感器。可移动平台在正常工作状态下，可移动平台可以在其他与可移动平台通信的设备的控制下进行移动，也可以是可移动平台自动控制移动。
在一些例子中，视觉传感器可以包括用于避障的图像传感器，用于避障的视觉传感器采集的图像可用于获取深度信息，例如，可移动平台在移动的过程中，视觉传感器可以对视野范围内的环境采集图像，基于采集的图像，可以通过视觉定位算法确定 出视野范围内的景物的深度信息，利用确定出的深度信息，可移动平台可以控制自身在空间中安全移动。其中，视觉传感器采集的图像供可移动平台计算深度信息，该图像未向用户展示；可选的，在一些场景中，根据需要，视觉传感器采集的图像也可以向用户展示。在其他场景中，视觉传感器也可以包括其他功能的图像传感器，例如用于采集向用户展示的图像的传感器等，本实施例对此不进行限定。
在一些例子中，可移动平台包括一个或多个视觉传感器，视觉传感器在可移动平台可以固定或非固定设置于可移动平台上。其中，根据需要可以选取一个或多个视觉传感器采集图像。例如，可以根据视觉传感器的视野范围和/或所需要识别的可拆卸部件在可移动平台上的连接位置来选取。例如，针对需要识别的可拆卸部件，选取视野范围覆盖所述可拆卸部件与可移动平台的连接位置的视觉传感器。
在一些例子中，所述视觉传感器固定设置于所述可移动平台上，所述视觉传感器的视野范围覆盖所述可移动平台的用于连接所述可拆卸部件的连接位置。在另一些例子中，所述视觉传感器通过云台搭载于所述可移动平台上；所述图像，是通过控制云台使所述视觉传感器朝向所述可移动平台的用于连接所述可拆卸部件的连接位置后采集的。
示例性的，所述采集的图像可以是一个视觉传感器采集的图像，也可以是多个视觉传感器的图像。例如，可以是同一视觉传感器采集的一张或多张图像，也可以是多个传感器分别采集的一张图像，还可以是多个传感器中不同传感器采集的多张图像等。其中，所述采集的图像可以包括可移动平台的全部或局部，即图像内容包含整个可移动平台，也可以只是可移动平台的部分。实际应用中，根据可拆卸部件的可拆卸特点，所述采集的图像不限定是否包括有可拆卸部件。例如，可拆卸部件被拆卸，并未与可移动平台连接，因此采集到的图像中并未包括有可拆卸部件；或者，可拆卸部件并未与可移动平台安全连接，因此采集到的图像中包括至少部分可拆卸部件，例如可拆卸部件的全部或部分。或者，可拆卸部件与可移动平台安全连接下，采集的图像中包括有可拆卸部件的全部，也可以是与可移动平台连接处等等。实际应用中，可以根据需要进行配置，只要采集的图像能用于检测可拆卸部件是否与可移动平台安全连接即可。
本实施例中，所述可拆卸部件与可移动平台可拆卸连接，可以包括与可移动平台移动有关的部件、与可移动平台正常执行任务有关的部件。在一些例子中，所述可移动平台包括无人机。所述可拆卸部件包括如下任一：机臂、螺旋桨桨叶、起落架、云台、云台保护部件、天线或电池组件等。在另一些例子中，所述可移动平台包括清洁机器人。所述可拆卸部件包括如下任一：清洁部件、清洁部件的保护部件、驱动部件或清洁机器人的保护部件等等。在另一些例子中，所述可移动平台包括车辆。所述可拆卸部件包括如下任一：车门、车窗、车轮、车后镜、车灯、机盖、天线、雨刷或传感器等等。
在步骤204中、根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；实际应用中可以有多种实现方式。例如，可以预先设置可拆卸部件的安全连接状态，通过对采集的图像进行图像识别，与预先设置的可拆卸部件的安全连接状态进行比对，根据比对结果确定可拆卸部件是否处于安全连接状态。例如，可以通过对图像中的一个或多个表征可拆卸部件与可移动平台的连接位置的像素区域进行图像识别，以分析该像素区域中可拆卸部件是否处于安全连接状态。或者，将采集的图像与表征可拆卸 部件处于安全连接状态的图像进行匹配，根据图像匹配结果可以确定可拆卸部件是否安全连接；或者，还可以通过机器学习模型的方式实现等等。
示例性的，所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括如下任一：识别所述图像中所述可移动平台的用于连接所述可拆卸部件的连接位置是否连接有部件；识别所述图像中所述连接位置连接的部件的类型，判断识别出的部件的类型是否与预设类型匹配；或，识别所述图像中所述连接位置连接的部件与所述连接位置的相对位置关系，判断识别出的相对位置关系是否与预设相对位置关系匹配。
实际应用中，上述实现方式根据需要可以采用一种或多种的组合。例如，若可拆卸部件未安装，通过识别所述图像中所述可移动平台的用于连接所述可拆卸部件的连接位置是否连接有部件，若未安装，可以确定可拆卸部件未处于安全连接状态。在一些例子中，用户可能安装错可拆卸部件，若确定有安装于可移动平台，可进一步识别所述图像中所述连接位置连接的部件的类型，判断识别出的部件的类型是否与预设类型匹配。可选的，用户也可能安装正确的可拆卸部件，但可拆卸部件的并未能安装在正确的位置，因此，还可以通过识别所述图像中所述连接位置连接的部件与所述连接位置的相对位置关系，判断识别出的相对位置关系是否与预设相对位置关系匹配。通过上述方式，可以准确地根据采集的图像判断所述可拆卸部件是否处于安全连接状态。
接下来以旋翼飞机为例说明根据采集的图像判断可拆卸部件是否处于安全连接状态的实施例。以旋翼飞机为例，旋翼飞机包括机身，其中，机臂与机身可拆卸连接，即机臂可以被拆卸，并且，机臂在连接上无人机时是可活动连接，可活动连接是指旋翼飞机未工作时可以机臂可折叠，但旋翼飞机在飞行过程中，机臂展开。因此，机臂作为可拆卸部件，其是否处于安全状态将影响旋翼飞机的安全飞行。
基于此，在一些例子中，所述可拆卸部件包括机臂；所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括：根据采集的图像判断所述机臂是否处于安全展开状态。因此，本实施例通过判断所述机臂是否处于安全展开状态，可以保证无人机的安全飞行。
旋翼飞机中螺旋桨桨叶也影响无人机的安全飞行。旋翼无人机的旋翼数量包括二旋翼、四旋翼、六旋翼、八旋翼等等。通常情况下，一个动力电机驱动一个螺旋桨旋转，通过多个螺旋桨的配合，可以实现飞机的俯仰、横滚和偏航的动作。在一些例子中，以四旋翼为例，飞机中的飞行控制系统可以调节四个电机转速来改变螺旋桨转速，实现升力的变化，从而控制飞机的姿态和位置。其中，螺旋桨转动过程中由于空气阻力作用会形成与转动方向相反的反扭矩，为了克服反扭矩影响，可使四个螺旋桨中的两个正转，两个反转，且对角线上的各个旋翼转动方向相同。因此，旋翼飞机中螺旋桨桨叶通常是成对设计，包括正桨桨叶和反桨桨叶，正桨桨叶和反桨桨叶需要安装在无人机中正确的位置，例如正桨桨叶和反桨桨叶设置在同一对角线上。可选的，螺旋桨的桨叶数量通常是2个或更多，螺旋桨的长度也不尽相同，不同无人机设计有不同类型的螺旋桨。综上可知，无人机中机臂及螺旋桨作为可拆卸部件，其连接状态是否安全，会影响无人机的飞行安全。
示例性的，所述可拆卸部件包括不同类型的螺旋桨桨叶，其中，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括：根据采集的图像，判断所述可移动平台的所述连接位 置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配。本实施例中，可移动平台可以连接多个螺旋桨桨叶，由于螺旋桨桨叶具有不同的类型，本实施例中可移动平台每个与螺旋桨桨叶的连接位置，可以设置有对应的预设螺旋桨桨叶类型，基于此，通过判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配，可以准确地识别可移动平台上各个螺旋桨桨叶是否安装在正确的连接位置上。
在一些例子中，可以通过图像识别可移动平台的连接位置上的螺旋桨桨叶的外观信息，来判断连接位置上的螺旋桨桨叶是否正确，可选的，所述螺旋桨桨叶的类型包括：表征所述螺旋桨桨叶外观的外观信息，所述外观信息如下任一：形状信息、长度信息、叶片数量信息、颜色信息、材质信息或大小信息等等，基于此，通过外观信息的识别，可以准确地判断出可移动平台的各个连接位置上所连接的螺旋桨桨叶是否正确。
在一些例子中，还可以通过机器学习模型的方式来识别可拆卸部件是否处于安全连接状态。示例性的，所述根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态，包括：由机器学习模型根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态。
模型的训练过程可以是：先通过建模表示出一个模型，再通过构建评价函数对模型进行评价，最后根据样本数据及最优化方法对评价函数进行优化，把模型调整到最优。
其中，建模是将实际问题转化成为计算机可以理解的问题，即将实际的问题转换成计算机可以表示的方式。建模一般是指基于大量样本数据估计出来模型的目标函数的过程。
评价的目标是判断已建好的模型的优劣。对于第一步中建好的模型，评价是一个指标，用于表示模型的优劣。这里就会涉及一些评价的指标以及一些评价函数的设计。在机器学习中会有针对性的评价指标。例如，在建模完成后，需要为模型设计一个损失函数，来评价模型的输出误差。
优化的目标是评价函数。即利用最优化方法，对评价函数进行最优化求解，找到评价最高的模型。例如，可以通过诸如梯度下降法等最优化方法，找到损失函数的输出误差的最小值(最优解)，将模型的参数调到最优。
可以这么理解，要训练一个机器学习模型之前，首先确定出一个合适的参数估计方法，再利用这种参数估计方法，把这个模型的目标函数中的各个参数估计出来，进而确定出目标函数最终的数学表达式。
在机器学习领域，如上所述，从建模至训练阶段涉及非常多的环节，例如样本数据的选择与处理、数据特征的设计、模型的设计、损失函数的设计或优化方法的设计等等，任一环节的细微差别都是导致预测准确度细微缺陷的因素。
本实施例中模型训练可以是有监督训练，半监督训练或无监督训练等等。在一些例子中，可以采用有监督训练方式以提高训练速度，样本数据中可以标注真实值，例如，所述机器学习模型通过样本图像集训练得到，所述样本图像集中包括：标注有所述可拆卸部件未处于安全连接状态的样本图像和标注有所述可拆卸部件处于安全连接状态的样本图像。本实施例通过有监督的训练方式，可以提高模型训练的速度和精确 度。
利用上述样本数据后，可以对机器学习模型训练得到。机器学习模型可以是神经网络模型等，例如基于卷积神经网络模型等等，实际应用中，可以根据需要灵活实现机器学习模型的结构设计；其中，模型的结构设计也是训练过程的其中一个重要方面，会影响到模型的预测精确度。基于此，示例性的，所述可拆卸部件与所述可移动平台的连接状态有多种类别，所述机器学习模型具有多个网络分支，每个网络分支分别用于识别所述可拆卸部件的不同类别的连接状态是否为所述安全连接状态，因此一个网络分支用于识别一种连接状态，因此可以提升模型识别准确度。
示例性的，所述机器学习模型可以有多个，所述可拆卸部件有多类，每个机器学习模型对应一类可拆卸部件，每个机器学习模型用于利用所述图像识别该机器学习模型对应的该类可拆卸部件是否处于安全连接状态。因此，若可移动平台的可拆卸部件有多类，通过本实施例设计每类部件对应一个机器学习模型，可以显著地提升模型识别准确度。
本申请实施例可以预先训练有机器学习模型，该机器学习模型可以设置于可移动平台中，也可以设置于与可移动平台通信连接的服务端中。在一些例子中，机器学习模型可以由业务方预先进行训练，训练后的机器学习模型可以存储在可移动平台中，以由可移动平台完成可拆卸部件是否处于安全连接状态的识别。在另一些例子中，还可以是可移动平台将采集的图像发送给服务端，由配置于服务端的机器学习模型从图像中识别可拆卸部件是否处于安全连接状态，并向可移动平台返回识别结果。在其他例子中，还可以是将机器学习模型置于其他设备中，由其他设备识别图像并判断可移动平台的可拆卸部件是否处于安全状态。
通过上述判断方式，可以根据需要输出提示信息。例如，若所述可拆卸部件未处于安全连接状态，输出用于指示将所述可拆卸部件重新连接至所述安全连接状态的提示信息，通过该提示信息使用户关注到可拆卸部件未处于安全连接状态。示例性的，所述提示信息可以仅用于向用户报错，即仅用于提示用户可拆卸部件未处于安全连接状态；在其他例子中，所述提示信息还可以用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态，例如，该提示信息可以包括可拆卸部件从当前状态重新连接至所述安全连接状态的过程，从而可以指导用户如何将可拆卸部件从当前状态重新连接至所述安全连接状态，从而可以实现更好的提示效果，使用户更高效地重新连接可拆卸部件。
在一些例子中，所述提示信息的输出方式可以有多种方式实现，作为例子，例如可以是与所述可移动平台通讯连接的设备输出的，和/或，是所述可移动平台输出的。其中，与所述可移动平台通讯连接的用户端可以包括任意设备，例如遥控器、智能手机或可穿戴设备等等。
示例性的，如图2B所示，本实施例可移动平台以无人机110为例，与无人机110通讯连接的其他设备，本实施例以遥控器210及智能手机220为例进行说明。其中，该遥控器210带有显示器，该遥控器和显示器可以可拆卸连接，该显示器也可以固定设置于遥控器上。其他设备与无人机通信的类型的示例可以包括但不限于经由以下方式的通信：因特网，局域网(LAN)，广域网(WAN)，蓝牙，近场通信(NFC)技术，基于诸如通用分组无线电服务(GPRS)、GSM、增强型数据GSM环境(EDGE)、 3G、4G或长期演进(LTE)协议的移动数据协议的网络，红外线(IR)通信技术，和/或WiFi，并且可以是无线式、有线式、或其组合。
其中，提示信息可以有多种方式实现，例如可以包括图像信息、文本信息、视频信息、语音信息或灯光信息等，相对应的，提示信息的输出的方式可以有多种实现方式。示例性的，所述输出提示信息，包括如下任一：在用户界面上输出提示信息；和/或，控制所述可移动平台上的一个或多个目标部件输出提示信息。例如可以其他设备的用户界面中显示图像、显示文本、显示视频，或者通过控制播放组件播放语音信息，或者控制照明组件显示灯光信息等等。
示例性的，所述用户界面中展示有所述可拆卸部件处于安全连接状态的示意图像。或者，示例性的，所述用户界面上展示有将所述可拆卸部件从当前状态重新连接至所述安全连接状态的示意图像。
如图2C所示，是本实施例中一种输出提示信息的示意图，图2C以智能手机为例，该智能手机的屏幕显示有无人机中螺旋桨桨叶从当前的未安装状态至连接至无人机机臂上处于安全连接状态的示意。可选的，提示信息采用文本信息“亲，这个机臂上未安装螺旋桨，安全连接状态请见下图，请安装后再飞哦”，从而可以通过上述方式可以清楚地向用户指导如何将可拆卸部件从当前状态重新连接至安全连接状态的过程。
示例性的，所述目标部件输出提示信息：包括如下任一：发光部件输出灯光提示信息、发声组件输出声音信息或螺旋桨的电机振动输出振动信息。
作为例子，无人机的各机臂上设置有目标部件，提示信息可以由设置在机臂上的目标部件输出，该提示器可以包括发声组件和/或发光组件，通过发声组件可以发出声音，通过发光组件可以发光，从而通过无人机对用户进行提示。
示例性的，还可以通过螺旋桨的电机振动输出振动信息。无人机中飞控通过电调控制电机转动，由电机驱动螺旋桨转动。所述电机用于驱动螺旋桨的转动角度较大时，桨叶推动空气向后运动的力较大，同时受到空气的反作用力推动，使得旋翼飞机运动。本实施例中，控制与所述动力电机连接的螺旋桨转动预设角度可以是旋翼飞机停止运动后执行，并且，可以控制螺旋桨转动预设角度，该预设角度可以处于使旋翼飞机未运动的角度范围内，因此螺旋桨转动角度较小，其推动空气的力较小，使得旋翼飞机未运动。因此，用户可以看到旋翼飞机中的一个或多个电机转动，转动的电机所在的机臂即未安全连接的可拆卸部件所在的机臂。其中，螺旋桨转动角度较小，也即是动力电机的转动角度较小，使得电机反复地在以较小的转动角度转动，使得电机的转子产生振动，因此电机能够振动发声，使得用户可以听到电机发出的声音，从而提示用户发出声音的电机所在的机臂或螺旋桨未处于安全连接状态，通过上述方式可以实现较好的提示效果，用户可以直观地查阅到未安全连接的可拆卸部件所在的位置。
示例性的，所述可拆卸部件包括机臂，所述输出提示信息的目标部件包括：未处于安全连接状态的机臂上的部件。示例性的，所述可拆卸部件包括螺旋桨桨叶，所述螺旋桨桨叶安装于机臂上，所述输出提示信息的目标部件包括：未处于安全连接状态的螺旋桨桨叶所安装的机臂上的部件。基于此，可以让用户知道未处于安全连接状态的是无人机上的哪个机臂或哪个螺旋桨，从而达到更好的提示效果。
在一些例子中，所述方法还可包括：若基于所述图像确定所述可拆卸部件处于安全连接状态下，提示用户所述可移动平台可执行移动操作。本实施例中，在基于所述 图像确定所述可拆卸部件处于安全连接状态下，可以向用户提示所述可移动平台可执行移动操作，提高用户的使用体验。
接下来再通过一实施例进行说明。本实施例以可移动平台为四旋翼无人机为例，该无人机中的可拆卸部件以机臂和螺旋桨为例；本实施例的无人机具有一个或多个视觉传感器，无人机飞行时，视觉传感器用于采集无人机周围的环境观测数据，无人机可利用环境观测数据进行飞行控制，采集的环境观测数据还可以用于更多的飞行任务等，例如可以存储环境观测数据以供后期数据处理使用，还可以实时传输给可与无人机通信的其他设备，例如其他无人机、地面站、遥控设备或用户的便携式设备等等。
本实施例方案中希望通过无人机的视觉传感器可以对无人机自身进行图像采集，具体的，以可拆卸部件作为采集目标，采集到包含有可拆卸部件的图像；基于图像来识别可拆卸部件的连接状态。可选的，为了准确快速地通过图像识别可拆卸部件的连接状态，作为例子，本实施例可以采用机器学习模型的方式来实现。
首先举例说明机器学习模型的训练阶段的实施例。
(1)准备数据
本实施例中图像采集的过程，既可以是无人机的视觉传感器采集图像，还可以是人工对无人机拍摄图像。其中，图像从哪些角度采集根据需要也有多种实现方式。作为例子，可以采集无人机的左右双目的图像，可以得到4张图像，分别对应无人机的左视左目，左视右目，右视左目，右视右目。当然，其他实施例也是可选的，例如可以对无人机采集无人机整体的俯视图等等。
一方面，可以采集机臂的图像数据，以四旋翼无人机为例，可以分别采集四个机臂正常展开以及其他所有非正常展开的状态的图像。
另一方面，可以采集桨叶安装和未安装的图像数据，以及多种不同桨叶型号的图像数据；作为例子，可以在无人机上安装多种不同类型的桨叶后采集图像，以及无人机安装相匹配的桨叶后采集图像；可选的，可以以时间戳为单位，每个时间戳同时采集四张图；当然，实际应用中图像的数量可以根据需要灵活配置。
其他的，还可以采集表示不同安装方式的桨叶的图像数据，例如分别采集桨叶不同的安装方式的图像，以及桨叶正确安装的图像，以时间戳为单元一次性采集四张图像；当然，实际应用中图像的数量可以根据需要灵活配置。
可选的，实际业务中无人机起飞的场景多种多样，为了进一步提高图像识别的准确性，图像数据根据需要可以覆盖尽可能多的场景，即将无人机置于不同的起飞场景中采集图像数据，比如沥青路、草坪、沙滩、石子路等等多种起飞场景。
(2)为图像配置标签
可选的，本实施例采用有监督的训练方式，实际应用中可以根据需要采用无监督等其他训练方式。
作为例子，可以标注机臂的图像数据，每张图像标注机臂是否是正常状态，属于二分类任务。其中包含左视左目、左视右目、右视左目及右视右目四张图的所有机臂图像。
接着，可标注桨叶是否安装，此标注只关注桨叶是否安装；若桨叶已经安装则标注桨叶型号是否正确，作为二分类标注任务，其中同一个时间戳的四张图像作为一组数据，只要四个机臂的电机中桨叶有错误的类型或者没有安装桨叶就标注为不正确的 状态，此标注只关注桨叶类型。进一步的，还可以标注桨叶安装错误的类别，分为六个类别，每个类别代表不同的安装状态，根据具体的错误类别进行多分类标注。
(3)网络构建与训练
如图3A所示，是本申请一实施例中机器学习模型的示意图；作为例子，为了满足能够在嵌入式设备实时运行的要求，卷积层以卷积核大小为3x3，步长为2为例，这样能够逐渐减小特征图的尺寸，降低计算量，结构可参考图3A所示。
本实施例以可拆卸部件为机臂和螺旋桨为例，因此以构建两个网络模型为例，一个称之为机臂展开识别模型，另一个称之为桨叶识别模型。
其中，机臂展开识别模型可以以单张图作为输入，也可以输入多个视角的图像，例如分别输入四个方向的图像等，可判断无人机中每个机臂是否正常展开。
其中，桨叶识别模型可将同一时间戳的4张图像一起作为输入；本实施例基于桨叶连接状态的三种类别，桨叶识别模型中间分出三个分支，其中一个分支判断桨叶是否安装，一个分支判断桨叶型号是否正确，另一个分支分类桨叶安装状态属于哪个类别。
两个模型可分别训练。训练机臂展开识别模型时，可以以单张图像作为输入，标注结果作为真值，使用交叉熵作为分类损失函数，反向传播训练卷积神经网络。训练桨叶识别模型时，输入为四张图像的组合，标注结果包含桨叶是否安装、桨叶的型号是否正确的标注位，以及桨叶安装属于哪个类别；网络分成三个分支，其中一个分支用于判断桨叶型号是否正确，另一个分支用于是否安装桨叶，属于二分类任务，另一个分支用于识别桨叶属于正确安装还是哪种类型的错误安装，属于多分类任务，两个分支都可采用交叉熵损失函数，共同监督整个桨叶识别模型训练。
训练桨叶识别模型时，由于是三个分支一起训练，但三个分支是属于前后关系的分支，所以在联合训练的时候，使用如下的规则进行联合训练，当未安装桨叶时，其他两个分支的损失为零，当安装桨叶但型号不正确时，桨叶安装分类网络的损失为零，可以得到如下的损失函数：
Loss＝loss1+a*loss2+a*b*loss3
其中loss1为判断是否安装桨叶的损失，loss2为桨叶型号是否正确的损失，loss3为桨叶安装分类的损失。a代表桨叶是否安装(1代表安装，0代表未安装)，b代表桨叶型号是否正确(1代表型号正确，0代表不正确)。根据上式的联合损失函数，能够有效的避免干扰，正确地训练桨叶识别模型。
本实施例中，可将桨叶安装方式分类为多个类别；本实施例的桨叶有正桨和反桨之分，正桨与反桨需要安装在正确的位置。如图3B所示，是本申请根据一示例性实施例示出的四旋翼无人机的六种桨叶安装方式的示意图。其中，×和O分别表示正桨和反桨，图中第0类属于正确安装的类别，其他类别属于安装错误的类别，根据识别错误的类别，能够准确的提示用户修正的方案，减小用户纠错成本。
通过上述训练过程，训练结束获得机臂展开识别模型和桨叶识别模型，模型可以设置于可移动平台中或服务端中。在需要时，可移动平台控制视觉传感器对机臂和螺旋桨采集图像，由这两个模型识别机臂和螺旋桨是否处于安全状态。
接下来对可移动平台识别机臂和螺旋桨是否处于安全状态的过程举例说明。
作为例子，可以在无人机准备起飞前执行识别，分别将左视左目、左视右目、右 视左目、右视右目的图像分别输入到机臂展开识别模型，机臂展开模型分别对四张图像识别机臂是否展开的识别结果，其中，基于采集的图像的视角，每张图像对应无人机的一个机臂，因此如果识别出哪张图像的识别结果是机臂没有正常展开，可相应地确定出机臂位置，从而可以输出提示信息，例如，提醒用户哪个机臂需要将机臂展开，还可配合对应机臂上设置的信号灯闪烁提醒用户相应的机臂。
在机臂检测完成后，当机臂都正常展开后，将同一时间戳的4张图像合并输入到桨叶识别模型，首先可以获取是否安装桨叶的识别结果，若未安装桨叶，可以通过应用程序提醒用户安装，若已经安装好桨叶，则可识别桨叶类型是否正确，若类型不正确，可以通过应用程序提醒用户桨叶类型不正确，若桨叶类型正确，则可识别桨叶安装的分类结果，若安装正确，可确定机臂和桨叶均处于安全连接状态，因此确定无人机处于可以起飞状态。如果桨叶安装不正确，则根据分类结果给用户相应的指导进行修正。例如，根据每个不同的分类结果，给与用户相应的指导，如图3C所示，是本申请根据一示例性实施例示出的用户界面输出提示信息的示意图，本实施例中桨叶识别模型识别出螺旋桨正桨和反桨的安装位置出现如图3B中类别1的错误安装，因此可以给出图3C中示出的提示方式，其中，图3C中左侧图像指导用户将其中两个螺旋桨调换位置，右侧图像给出了该无人机中螺旋桨正桨和反桨的正确安装位置。
上述实施例均以可移动平台为执行主体为例进行说明，在其他例子中，实施本实施例的执行主体并不限制于可移动平台，还可以是其他电子设备，例如带有显示屏的的遥控设备、智能手机、个人数字助理、平板电脑、多媒体设备、可穿戴设备或个人计算机等等，其中，该电子设备是否与可移动平台通信连接也是非必须的。
如图4所示，是本申请根据一示例性实施例示出的一种信息提示方法，所述方法包括：
在步骤402中，对所述可移动平台的螺旋桨桨叶采集图像；所述可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接；其中，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；
在步骤402中，根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配；
在步骤402中，若未匹配，输出提示信息，所述提示信息用于指示将所述连接位置与螺旋桨桨叶重新连接。
本实施例方案可应用于可移动平台或电子设备，其中，图像的采集可以是电子设备配置的视觉传感器采集到，也可以是电子设备采集的。具体的实现方式可参考前述实施例的描述，本实施例在此不进行赘述。
由上述实施例可见，可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接；其中，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同，通过对可移动平台采集图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配，因此能够准确地识别出可移动平台的连接位置上是否连接有正确的螺旋桨桨叶，若未正确连接，能够输出提示信息，以指示用户将所述连接位置与螺旋桨桨叶重新连接，从而可以保证可移动平台的安全移动。
如图5所示，是本申请根据一示例性实施例示出的另一种信息提示方法的示意图， 所述方法包括：
在步骤502中，对可移动平台的可拆卸部件采集图像；
在步骤504中，根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
在步骤506中，若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
本实施例方案可应用于可移动平台或电子设备，其中，图像的采集可以是电子设备配置的视觉传感器采集到，也可以是电子设备采集的。具体的实现方式可参考前述实施例的描述，本实施例在此不进行赘述。
由上述实施例可见，通过对可移动平台采集图像，根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态，若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息还可以用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态，从而可以指导用户如何将可拆卸部件从当前状态重新连接至所述安全连接状态，实现更好的提示效果，使用户更高效快速地重新连接可拆卸部件至安全连接状态。
上述方法实施例可以通过软件实现，也可以通过硬件或者软硬件结合的方式实现。以软件实现为例，作为一个逻辑意义上的装置，是通过其所在图像处理的处理器将非易失性存储器中对应的计算机程序指令读取到内存中运行形成的。从硬件层面而言，请参考图6，是本申请一个实施例的可移动平台的控制装置600的结构图，所述可移动平台的控制装置600包括处理器601、存储器602、存储在所述存储器上可被所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现如下方法：
基于所述视觉传感器对可移动平台的可拆卸部件采集图像；所述可移动平台包括至少一个视觉传感器，至少一个可拆卸部件用于搭载在所述可移动平台；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件重新连接至所述安全连接状态；
在基于所述图像确定所述可拆卸部件处于安全连接状态下，基于所述视觉传感器采集的环境图像控制所述可移动平台在空间中运动。
在一些例子中，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
在一些例子中，所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括如下任一：
识别所述图像中所述可移动平台的用于连接所述可拆卸部件的连接位置是否连接有部件；
识别所述图像中所述连接位置连接的部件的类型，判断识别出的部件的类型是否与预设类型匹配；或，
识别所述图像中所述连接位置连接的部件与所述连接位置的相对位置关系，判断识别出的相对位置关系是否与预设相对位置关系匹配。
所述可拆卸部件包括不同类型的螺旋桨桨叶，其中，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；
所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括：
根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配。
在一些例子中，所述螺旋桨桨叶的类型包括：表征所述螺旋桨桨叶外观的外观信息，所述外观信息如下任一：形状信息、长度信息、叶片数量信息、颜色信息、材质信息或大小信息。
在一些例子中，所述可拆卸部件包括机臂；
所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括：
根据采集的图像判断所述机臂是否处于安全展开状态。
在一些例子中，所述基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像的步骤，
是基于与所述可移动平台通信的其他设备发送的消息执行的；和/或，
是基于可移动平台的移动状态信息执行的；和/或，
是基于所述可移动平台开机后产生的触发信号执行的。
在一些例子中，所述视觉传感器固定设置于所述可移动平台上，所述视觉传感器的视野范围覆盖所述可移动平台的用于连接所述可拆卸部件的连接位置。
在一些例子中，所述视觉传感器通过云台搭载于所述可移动平台上；
所述图像，是通过控制云台使所述视觉传感器朝向所述可移动平台的用于连接所述可拆卸部件的连接位置后采集的。
在一些例子中，所述根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态，包括：
由机器学习模型根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态。
在一些例子中，所述机器学习模型有多个，所述可拆卸部件有多类，每个机器学习模型对应一类可拆卸部件，每个机器学习模型用于利用所述图像识别该机器学习模型对应的该类可拆卸部件是否处于安全连接状态。
在一些例子中，所述可拆卸部件与所述可移动平台的连接状态有多种类别，所述机器学习模型具有多个网络分支，每个网络分支分别用于识别所述可拆卸部件的不同类别的连接状态是否为所述安全连接状态。
在一些例子中，所述机器学习模型通过样本图像集训练得到，所述样本图像集中包括：标注有所述可拆卸部件未处于安全连接状态的样本图像和标注有所述可拆卸部件处于安全连接状态的样本图像。
在一些例子中，所述输出提示信息，包括如下任一：
在用户界面上输出提示信息；和/或，
控制所述可移动平台上的一个或多个目标部件输出提示信息。
在一些例子中，所述用户界面包括：与所述可移动平台通信连接的其他设备的用户界面。
在一些例子中，所述用户界面中展示有所述可拆卸部件处于安全连接状态的示意图像。
在一些例子中，所述用户界面上展示有将所述可拆卸部件从当前状态重新连接至所述安全连接状态的示意图像。
在一些例子中，所述目标部件输出提示信息：包括如下任一：
灯光部件输出灯光提示信息、扬声器输出声音信息或螺旋桨的电机振动输出振动信息。
在一些例子中，所述可拆卸部件包括机臂，所述输出提示信息的目标部件包括：未处于安全连接状态的机臂上的部件。
在一些例子中，所述可拆卸部件包括螺旋桨桨叶，所述螺旋桨桨叶安装于机臂上，所述输出提示信息的目标部件包括：未处于安全连接状态的螺旋桨桨叶所安装的机臂上的部件。
在一些例子中，所述方法还包括：
若基于所述图像确定所述可拆卸部件处于安全连接状态下，提示用户所述可移动平台可执行移动操作。
上述可移动平台的控制装置中各个单元的功能和作用的实现过程具体详见上述可移动平台的控制方法中对应步骤的实现过程，在此不再赘述。
对于装置实施例而言，由于其基本对应于方法实施例，所以相关之处参见方法实施例的部分说明即可。以上所描述的装置实施例仅仅是示意性的，其中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模块来实现本申请方案的目的。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。
实际应用中，本申请可移动平台的控制装置还可以包括内存、网络接口、以及非易失性存储器等等，实施例中装置所在的可移动平台通常根据该可移动平台的实际功能，还可以包括其他硬件，对此不再赘述。
如图7所示，本实施例还提供一种信息提示装置700，所述装置700包括处理器701、存储器702、存储在所述存储器702上可被所述处理器701执行的计算机程序，所述处理器执行所述计算机程序时实现前述所述信息提示方法的实施例。
作为例子，所述处理器执行所述计算机程序时实现：
获取多个所述旋翼的预设桨叶型号；
基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号；
若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息。
在一些例子中，所述旋翼飞行器包括一个或者多个图像传感器，一个或者多个图像传感器的观测范围覆盖所述旋翼飞行器的多个所述旋翼，所述图像数据是基于所述一个或者多个图像传感器采集得到的。
在一些例子中，所述图像传感器的所述观测范围同时覆盖所述旋翼飞行器的周围环境，所述图像传感器用于在所述旋翼处于飞行状态时对所述周围环境进行观测，以使多个所述旋翼飞行器基于所述观测得到的信息调整一个或者多个所述旋翼运行参数。
在一些例子中，所述提示信息包括在所述旋翼飞行器的地面控制器上显示的图像提示信息；
所述图像提示信息包括:指示多个所述旋翼的旋翼标识，以及指示所述旋翼适配的预设桨叶型号的第一桨叶标识。
在一些例子中，所述图像提示信息还包括：任一所述旋翼的当前安装的桨叶型号的第二桨叶标识。
在一些例子中，所述图像提示信息包括：指示所述旋翼适配的预设桨叶型号的所述第一桨叶标识与对应的所述旋翼关联关系的调整指引标识。
在一些例子中，所述图像提示信息包括：指示所述旋翼飞行器的机身的机身标识；
所述旋翼标识与所述机身标识显示的显示位置关系，与所述旋翼在所述旋翼飞行器的机身的位置分布一致。
或者，所述处理器执行所述计算机程序时还可实现：
对可移动平台的螺旋桨桨叶采集图像；其中，所述可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；
根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配；
若未匹配，输出提示信息，所述提示信息用于指示将所述连接位置与螺旋桨桨叶重新连接。
或者，所述处理器执行所述计算机程序时还可实现：
对可移动平台的至少一个可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
如图8所示，本实施例还提供一种可移动平台800，所述可移动平台800包括至少一个视觉传感器803，至少一个可拆卸部件804用于与所述可移动平台连接；所述可移动平台还包括处理器801、存储器、存储在所述存储器802上可被所述处理器801执行的计算机程序；
所述处理器执行所述计算机程序时实现前述任一方法的实施例。
例如，所述处理器执行所述计算机程序时可实现：
获取多个所述旋翼的预设桨叶型号；
基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号；
若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息。
或者，所述处理器执行所述计算机程序时可实现：
基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件重新连接至所述安全连接状态；
在基于所述图像确定所述可拆卸部件处于安全连接状态下，基于所述视觉传感器采集的环境图像控制所述可移动平台在空间中运动。
或者，所述处理器执行所述计算机程序时可实现：
对可移动平台的螺旋桨桨叶采集图像；其中，所述可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接，不同类型的螺旋桨桨叶与 所述可移动平台的连接位置不同；
根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配；
若未匹配，输出提示信息，所述提示信息用于指示将所述连接位置与螺旋桨桨叶重新连接。
或者，所述处理器执行所述计算机程序时可实现：
对可移动平台的至少一个可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
如图9所示，本实施例还提供一种电子设备900，所述电子设备900包括处理器901、存储器902、存储在所述存储器902上可被所述处理器执行的计算机程序；
所述处理器执行所述计算机程序时实现前述任一方法的实施例。
例如，所述处理器执行所述计算机程序时可实现：
获取多个所述旋翼的预设桨叶型号；
基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号；
若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息。
或者，所述处理器执行所述计算机程序时可实现：
基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件重新连接至所述安全连接状态；
在基于所述图像确定所述可拆卸部件处于安全连接状态下，基于所述视觉传感器采集的环境图像控制所述可移动平台在空间中运动。
或者，所述处理器执行所述计算机程序时可实现：
对可移动平台的螺旋桨桨叶采集图像；其中，所述可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；
根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配；
若未匹配，输出提示信息，所述提示信息用于指示将所述连接位置与螺旋桨桨叶重新连接。
或者，所述处理器执行所述计算机程序时可实现：
对可移动平台的至少一个可拆卸部件采集图像；
根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
本实施例还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有若干计算机指令，所述计算机指令被执行时实现前述可移动平台的控制方法的实施例和/ 或所述信息提示方法的实施例。
本说明书实施例可采用在一个或多个其中包含有程序代码的存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。计算机可用存储介质包括永久性和非永久性、可移动和非可移动媒体，可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括但不限于：相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。
对于装置实施例而言，由于其基本对应于方法实施例，所以相关之处参见方法实施例的部分说明即可。以上所描述的装置实施例仅仅是示意性的，其中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模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上对本发明实施例所提供的方法和装置进行了详细介绍，本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本发明的限制。

Claims (35)

  1. 一种旋翼飞行器的信息提示方法，其特征在于，所述旋翼飞行器的多个旋翼包括可拆卸的桨叶，多个所述旋翼包括适配不同桨叶型号的不同旋翼；
    所述方法包括：
    获取多个所述旋翼的预设桨叶型号；
    基于图像数据识别安装在所述旋翼飞行器每个所述旋翼的桨叶型号；
    若识别的所述桨叶的型号与所述预设桨叶型号不匹配，输出用于指示调整所述桨叶安装状态的提示信息。
  2. 根据权利要求1所述的方法，其特征在于，所述旋翼飞行器包括一个或者多个图像传感器，一个或者多个图像传感器的观测范围覆盖所述旋翼飞行器的多个所述旋翼，所述图像数据是基于所述一个或者多个图像传感器采集得到的。
  3. 根据权利要求1所述的方法，其特征在于，所述图像传感器的观测范围同时覆盖所述旋翼飞行器的周围环境，所述图像传感器用于在所述旋翼处于飞行状态时对所述周围环境进行观测，以使多个所述旋翼飞行器基于所述观测得到的信息调整一个或者多个所述旋翼运行参数。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，所述提示信息包括在所述旋翼飞行器的地面控制器上显示的图像提示信息；
    所述图像提示信息包括：指示多个所述旋翼的旋翼标识，以及指示所述旋翼适配的预设桨叶型号的第一桨叶标识。
  5. 根据权利要求4所述的方法，其特征在于，所述图像提示信息还包括：任一所述旋翼的当前安装的桨叶型号的第二桨叶标识。
  6. 根据权利要求4所述的方法，其特征在于，所述图像提示信息包括：指示所述旋翼适配的预设桨叶型号的所述第一桨叶标识与对应的所述旋翼关联关系的调整指引标识。
  7. 根据权利要求4所述的方法，其特征在于，所述图像提示信息包括：指示所述旋翼飞行器的机身的机身标识；
    所述旋翼标识与所述机身标识显示的显示位置关系，与所述旋翼在所述旋翼飞行器的机身的位置分布一致。
  8. 一种可移动平台的控制方法，其特征在于，所述可移动平台包括至少一个视觉传感器，至少一个可拆卸部件用于搭载在所述可移动平台；所述方法还包括：
    基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像；
    根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
    若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件重新连接至所述安全连接状态；
    在基于所述图像确定所述可拆卸部件处于安全连接状态下，基于所述视觉传感器采集的环境图像控制所述可移动平台在空间中运动。
  9. 根据权利要求8所述的方法，其特征在于，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
  10. 根据权利要求8所述的方法，其特征在于，所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括如下任一：
    识别所述图像中所述可移动平台的用于连接所述可拆卸部件的连接位置是否连接有部件；
    识别所述图像中所述连接位置连接的部件的类型，判断识别出的部件的类型是否与预设类型匹配；或，
    识别所述图像中所述连接位置连接的部件与所述连接位置的相对位置关系，判断识别出的相对位置关系是否与预设相对位置关系匹配。
  11. 根据权利要求8所述的方法，其特征在于，所述可拆卸部件包括不同类型的螺旋桨桨叶，其中，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；
    所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括：
    根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配。
  12. 根据权利要求11所述的方法，其特征在于，所述螺旋桨桨叶的类型包括：表征所述螺旋桨桨叶外观的外观信息，所述外观信息如下任一：形状信息、长度信息、叶片数量信息、颜色信息、材质信息或大小信息。
  13. 根据权利要求8所述的方法，其特征在于，所述可拆卸部件包括机臂；
    所述根据采集的图像判断所述可拆卸部件是否处于安全连接状态，包括：
    根据采集的图像判断所述机臂是否处于安全展开状态。
  14. 根据权利要求8所述的方法，其特征在于，所述基于所述视觉传感器对所述可移动平台的所述可拆卸部件采集图像的步骤，
    是基于与所述可移动平台通信的其他设备发送的消息执行的；和/或，
    是基于可移动平台的移动状态信息执行的；和/或，
    是基于所述可移动平台开机后产生的触发信号执行的。
  15. 根据权利要求8所述的方法，其特征在于，所述视觉传感器固定设置于所述可移动平台上，所述视觉传感器的视野范围覆盖所述可移动平台的用于连接所述可拆卸部件的连接位置。
  16. 根据权利要求8所述的方法，其特征在于，所述视觉传感器通过云台搭载于所述可移动平台上；
    所述图像，是通过控制云台使所述视觉传感器朝向所述可移动平台的用于连接所述可拆卸部件的连接位置后采集的。
  17. 根据权利要求8所述的方法，其特征在于，所述根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态，包括：
    由机器学习模型根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态。
  18. 根据权利要求17所述的方法，其特征在于，所述机器学习模型有多个，所述可拆卸部件有多类，每个机器学习模型对应一类可拆卸部件，每个机器学习模型用于利用所述图像识别该机器学习模型对应的该类可拆卸部件是否处于安全连接状态。
  19. 根据权利要求17所述的方法，其特征在于，所述可拆卸部件与所述可移动平台的连接状态有多种类别，所述机器学习模型具有多个网络分支，每个网络分支分别用于识别所述可拆卸部件的不同类别的连接状态是否为所述安全连接状态。
  20. 根据权利要求17所述的方法，其特征在于，所述机器学习模型通过样本图像集训练得到，所述样本图像集中包括：标注有所述可拆卸部件未处于安全连接状态的 样本图像和标注有所述可拆卸部件处于安全连接状态的样本图像。
  21. 根据权利要求8至20任一所述的方法，其特征在于，所述输出提示信息，包括如下任一：
    在用户界面上输出提示信息；和/或，
    控制所述可移动平台上的一个或多个目标部件输出提示信息。
  22. 根据权利要求21所述的方法，其特征在于，所述用户界面包括：与所述可移动平台通信连接的其他设备的用户界面。
  23. 根据权利要求21所述的方法，其特征在于，所述用户界面中展示有所述可拆卸部件处于安全连接状态的示意图像。
  24. 根据权利要求21所述的方法，其特征在于，所述用户界面上展示有将所述可拆卸部件从当前状态重新连接至所述安全连接状态的示意图像。
  25. 根据权利要求21所述的方法，其特征在于，所述目标部件输出提示信息：包括如下任一：
    灯光部件输出灯光提示信息、扬声器输出声音信息或螺旋桨的电机振动输出振动信息。
  26. 根据权利要求21所述的方法，其特征在于，所述可拆卸部件包括机臂，所述输出提示信息的目标部件包括：未处于安全连接状态的机臂上的部件。
  27. 根据权利要求21所述的方法，其特征在于，所述可拆卸部件包括螺旋桨桨叶，所述螺旋桨桨叶安装于机臂上，所述输出提示信息的目标部件包括：未处于安全连接状态的螺旋桨桨叶所安装的机臂上的部件。
  28. 根据权利要求8所述的方法，其特征在于，所述方法还包括：
    若基于所述图像确定所述可拆卸部件处于安全连接状态下，提示用户所述可移动平台可执行移动操作。
  29. 一种信息提示方法，其特征在于，所述方法包括：
    对可移动平台的螺旋桨桨叶采集图像；其中，所述可移动平台包括不同类型的螺旋桨桨叶，所述螺旋桨桨叶与所述可移动平台可拆卸连接，不同类型的螺旋桨桨叶与所述可移动平台的连接位置不同；
    根据采集的图像，判断所述可移动平台的所述连接位置上连接的螺旋桨桨叶的类型与该连接位置对应的预设螺旋桨桨叶类型是否匹配；
    若未匹配，输出提示信息，所述提示信息用于指示将所述连接位置与螺旋桨桨叶重新连接。
  30. 一种信息提示方法，其特征在于，所述方法包括：
    对可移动平台的至少一个可拆卸部件采集图像；
    根据采集的所述图像判断所述可拆卸部件是否处于安全连接状态；
    若所述可拆卸部件未处于安全连接状态，输出提示信息，所述提示信息用于指示将所述可拆卸部件从当前状态重新连接至所述安全连接状态。
  31. 一种可移动平台的控制装置，其特征在于，所述装置包括处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现权利要求8至28任一所述的方法。
  32. 一种信息提示装置，其特征在于，所述装置包括处理器、存储器、存储在所 述存储器上可被所述处理器执行的计算机程序，所述处理器执行所述计算机程序时实现权利要求1至7、29或30任一所述的方法。
  33. 一种可移动平台，其特征在于，所述可移动平台包括至少一个视觉传感器，至少一个可拆卸部件用于与所述可移动平台连接；所述可移动平台还包括处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序；
    所述处理器执行所述计算机程序时实现权利要求8至28任一所述的方法。
  34. 一种电子设备，其特征在于，所述电子设备包括处理器、存储器、存储在所述存储器上可被所述处理器执行的计算机程序；
    所述处理器执行所述计算机程序时实现权利要求1至7、29或30任一所述的方法。
  35. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有若干计算机指令，所述计算机指令被执行时实现权利要求1至30任一项所述方法的步骤。
PCT/CN2022/081105 2022-03-16 2022-03-16 可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质 WO2023173307A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/081105 WO2023173307A1 (zh) 2022-03-16 2022-03-16 可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/081105 WO2023173307A1 (zh) 2022-03-16 2022-03-16 可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2023173307A1 true WO2023173307A1 (zh) 2023-09-21

Family

ID=88022094

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/081105 WO2023173307A1 (zh) 2022-03-16 2022-03-16 可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质

Country Status (1)

Country Link
WO (1) WO2023173307A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117234221A (zh) * 2023-11-14 2023-12-15 科沃斯家用机器人有限公司 自移动设备控制方法、自移动设备和存储介质
CN117975313A (zh) * 2024-03-28 2024-05-03 浙江华视智检科技有限公司 无人机状态检测方法、无人机控制装置和无人机

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200361634A1 (en) * 2019-05-16 2020-11-19 United Parcel Service Of America, Inc. Autonomous drone diagnosis
CN113302128A (zh) * 2020-08-24 2021-08-24 深圳市大疆创新科技有限公司 螺旋桨异常检测方法、无人机、控制终端、系统及介质
CN113945164A (zh) * 2021-10-14 2022-01-18 中国航空工业集团公司北京长城计量测试技术研究所 一种直升机桨叶自动测量装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200361634A1 (en) * 2019-05-16 2020-11-19 United Parcel Service Of America, Inc. Autonomous drone diagnosis
CN113302128A (zh) * 2020-08-24 2021-08-24 深圳市大疆创新科技有限公司 螺旋桨异常检测方法、无人机、控制终端、系统及介质
CN113945164A (zh) * 2021-10-14 2022-01-18 中国航空工业集团公司北京长城计量测试技术研究所 一种直升机桨叶自动测量装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117234221A (zh) * 2023-11-14 2023-12-15 科沃斯家用机器人有限公司 自移动设备控制方法、自移动设备和存储介质
CN117975313A (zh) * 2024-03-28 2024-05-03 浙江华视智检科技有限公司 无人机状态检测方法、无人机控制装置和无人机

Similar Documents

Publication Publication Date Title
US11530047B2 (en) Unmanned aerial vehicle with rotating and overlapping rotor arms
US11042074B2 (en) Flying camera with string assembly for localization and interaction
CN107531322B (zh) 空中捕获平台
US11649052B2 (en) System and method for providing autonomous photography and videography
US10648809B2 (en) Adaptive compass calibration based on local field conditions
JP6559677B2 (ja) データ記録および分析するシステム、方法、およびデータレコーダ
WO2023173307A1 (zh) 可移动平台及其控制方法、信息提示方法、装置、电子设备、计算机可读存储介质
CN108780316B (zh) 用于飞行装置的移动控制的方法和系统
CN110347171B (zh) 一种飞行器控制方法及飞行器
JP6899846B2 (ja) 飛行経路表示方法、モバイルプラットフォーム、飛行システム、記録媒体及びプログラム
CN105045279A (zh) 一种利用无人飞行器航拍自动生成全景照片的系统及方法
WO2020143677A1 (zh) 一种飞行控制方法及飞行控制系统
US11530038B2 (en) Detachable protection structure for unmanned aerial systems
WO2020062178A1 (zh) 基于地图识别目标对象的方法与控制终端
WO2018214155A1 (zh) 用于设备姿态调整的方法、设备、系统和计算机可读存储介质
JP6583840B1 (ja) 検査システム
WO2019227287A1 (zh) 无人机的数据处理方法和设备
EP3926976B1 (en) Information processing device, information processing method, and program
WO2021251441A1 (ja) 方法、システムおよびプログラム
CN116234751A (zh) 无人飞行器的系统和结构
WO2020062255A1 (zh) 拍摄控制方法和无人机
WO2020061941A1 (zh) 通信方法、系统和可移动平台
US12024284B2 (en) Information processing device, information processing method, and recording medium
US20230118521A1 (en) Aerial capture platform
CN116194369A (zh) 具有用户交互部件和可折叠结构的无人飞行器

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: 22931345

Country of ref document: EP

Kind code of ref document: A1