Scenario identification method, control apparatus, movable platform, and storage medium

Abstract

A scenario identification method, a control apparatus, a movable platform, and a computer-readable storage medium. The method comprises: controlling a detection apparatus to emit detection light, so as to detect the current scenario, wherein the detection apparatus comprises an area-array-type photoelectric sensor (S101); when the area-array-type photoelectric sensor receives reflection echoes of the detection light, acquiring a plurality of signals outputted by means of a plurality of photoelectric units of the area-array-type photoelectric sensor (S102); and on the basis of signal parameters of the plurality of signals, identifying whether the current scenario is a target scenario, wherein the target scenario at least comprises a water surface scenario (S103). Scenario identification steps are simple, thereby facilitating quick acquisition of a scenario identification result.

Description

Scene recognition method, control device, movable platform and storage medium Technical Field

[0001] The present application relates to the field of scene recognition technology, and in particular to a scene recognition method, a control device, a mobile platform, and a computer-readable storage medium. Background Art

[0002] With the continuous development of science and technology, mobile platforms are gradually becoming important auxiliary tools in people's lives and work. Mobile platforms refer to machine platforms with autonomous mobility capabilities, which can move freely and perform tasks in different environments.

[0003] Accurately identifying the current scene during the movement of a mobile platform is a crucial factor in ensuring its safe movement. Currently, scene identification is often achieved through visual observation by the driver or with the aid of visual sensors. However, in autonomous driving scenarios or in scenes that are not visually identifiable, effective methods for determining scene type are lacking.

[0004] Summary of the Invention

[0005] In view of this, one of the objectives of the present application is to provide a scene recognition method, a control device, a movable platform and a computer-readable storage medium.

[0006] In a first aspect, an embodiment of the present application provides a scene recognition method, comprising:

[0007] Controlling the detection device to emit detection light to detect the current scene; the detection device includes an array photoelectric sensor;

[0008] When the area array photoelectric sensor receives the reflected echo of the detection light, acquiring a plurality of signals output by a plurality of photoelectric units of the area array photoelectric sensor;

[0009] It is identified whether the current scene is a target scene according to the signal parameters of the multiple signals, and the target scene at least includes a water surface scene.

[0010] In a second aspect, an embodiment of the present application provides a scene recognition method, including:

[0011] Controlling a detection device to emit detection light to detect a current scene, wherein the detection device includes an array photoelectric sensor;

[0012] When the area array photoelectric sensor receives the reflected echo of the detection light, acquiring a plurality of signals output by a plurality of photoelectric units of the area array photoelectric sensor;

[0013] Whether the current scene is a target scene is identified based on the distribution and / or change of the signal parameters of the multiple signals.

[0014] In a third aspect, an embodiment of the present application provides a scene recognition method, including:

[0015] Acquire an echo signal of a reflective surface in the current scene, wherein the echo signal is an optical signal reflected back by the reflective surface when a detection device actively emits detection light, or an electrical signal converted from the optical signal;

[0016] Determining reflection information of the reflecting surface according to the echo signal, wherein the reflection information includes at least one of the following: distribution information and change information of signal parameters of multiple echo signals of the reflecting surface;

[0017] According to the reflection information of the reflection surface, it is determined whether the current scene is a target scene, and the target scene at least includes a water surface scene.

[0018] In a fourth aspect, an embodiment of the present application provides a control device, including:

[0019] a memory for storing executable instructions;

[0020] one or more processors;

[0021] When the one or more processors execute the executable instructions, they are individually or collectively configured to execute the method described in any one of the first aspect, the second aspect, or the third aspect.

[0022] In a fifth aspect, an embodiment of the present application provides a movable platform, including:

[0023] body;

[0024] a power system, mounted on the machine body, for providing power to the movable platform;

[0025] a detection device, mounted on the body, for emitting detection light to detect the current scene; and

[0026] A control device as described in the fourth aspect.

[0027] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores executable instructions, and when the executable instructions are executed by a processor, the method described in any one of the first aspect, the second aspect, or the third aspect is implemented.

[0028] The embodiments of the present application can identify target scenes based on the signal parameters of multiple signals output by multiple photoelectric units of a planar array photoelectric sensor. For example, it can identify whether the current scene is a water surface scene. For example, scene recognition can be performed based on the distribution and / or changes of the signal parameters of the multiple signals. This solves the problem that visual sensors cannot accurately identify water surface scenes, improves scene recognition accuracy, and improves the safety of the movable platform during movement. The embodiments and beneficial effects will be further elaborated below. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0030] FIG1 is a schematic structural diagram of a movable platform provided in an embodiment of the present application.

[0031] FIG2 is a schematic structural diagram of a power system of an unmanned aerial vehicle provided in an embodiment of the present application.

[0032] FIG3A is a schematic structural diagram of a detection device provided in an embodiment of the present application.

[0033] FIG3B is a schematic diagram of a detection scene of the detection device provided in an embodiment of the present application.

[0034] FIG4 is a schematic diagram of an application scenario of a movable platform provided in an embodiment of the present application.

[0035] FIG5 is a schematic diagram of an application scenario of another movable platform provided in an embodiment of the present application.

[0036] FIG6 is a flow chart of a scene recognition method provided in an embodiment of the present application.

[0037] FIG7 is a schematic diagram of multiple photoelectric units of the area array photoelectric sensor provided in an embodiment of the present application.

[0038] FIG8 is a flow chart of another scene recognition method provided in an embodiment of the present application.

[0039] FIG9 is a flow chart of another scene recognition method provided in an embodiment of the present application.

[0040] FIG10 is a schematic structural diagram of a control device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0042] Accurately identifying the current scene is a crucial factor in ensuring safe movement for mobile platforms. By accurately perceiving and understanding the surrounding environment, mobile platforms can make timely decisions and adjustments. Currently, scene recognition is often achieved through visual observation by the driver or with the help of visual sensors. However, in low-light or high-light scenes, or scenes with no or weak textures, visual sensors often cannot accurately identify the scene type, and there is a lack of effective methods for determining scene type.

[0043] In response to the problems in the related art, an embodiment of the present application provides a scene recognition method, which can perform scene recognition with the help of a detection device. First, the detection device can be controlled to emit detection light to detect the current scene. The detection device includes a planar array photoelectric sensor, so that when the planar array photoelectric sensor receives the reflected echo of the detection light, it can obtain multiple signals output by multiple photoelectric units of the planar array photoelectric sensor, and identify whether the current scene is the target scene based on the signal parameters of the multiple signals, thereby realizing scene recognition in a non-visual manner, improving the accuracy of scene recognition and the safety of the movable platform during movement. Moreover, it is only necessary to perform scene judgment based on signal parameters, and the recognition steps are simple, which helps to quickly obtain scene recognition results. Furthermore, the target scene includes at least a water surface scene, and it is possible to directly and quickly determine whether the current scene is a water surface scene based on the signal parameters of multiple signals, solving the problem that the water surface scene cannot be accurately identified by the visual sensor.

[0044] In response to the problems in the related art, the embodiments of the present application provide another scene recognition method, which can be used to perform scene recognition with the help of a detection device. First, the detection device can be controlled to emit detection light to detect the current scene. The detection device includes a planar array photoelectric sensor, so that when the planar array photoelectric sensor receives the reflected echo of the detection light, it can obtain multiple signals output by multiple photoelectric units of the planar array photoelectric sensor, and identify whether the current scene is the target scene based on the distribution and / or change of the signal parameters of the multiple signals; the scene recognition is realized by actively emitting a detection light source, which improves the accuracy of scene recognition and the safety of the movable platform during movement. Moreover, the scene judgment only needs to be made based on the distribution and / or change of the signal parameters. The recognition steps are simple, which helps to quickly obtain the scene recognition results.

[0045] In response to the problems in the related art, the embodiment of the present application provides another scene recognition method, which can be used to perform scene recognition with the help of a detection device. First, the detection device actively emits detection light, and then obtains the echo signal emitted by the reflective surface of the current scene; based on the echo signal, the reflection information of the reflective surface is determined, and the reflection information includes at least one of the following: distribution information and change information of the signal parameters of multiple echo signals of the reflective surface; based on the reflection information of the reflective surface, it is determined whether the current scene is a target scene, and the target scene includes at least a water surface scene. The scene recognition is achieved by non-visual means, which improves the accuracy of scene recognition and the safety of the movable platform during movement. Moreover, it is only necessary to directly and quickly determine whether the current scene is a water surface scene based on the distribution information and / or change information of the signal parameters of multiple echo signals of the reflective surface, which solves the problem that the water surface scene cannot be accurately identified by the visual sensor, and the recognition step is simple, and the scene recognition result can be quickly obtained for the subsequent rapid decision-making process of the movable platform.

[0046] The scene recognition method provided in the embodiment of the present application can be executed by a control device.

[0047] On the one hand, the control device can be a computer software product integrated in a mobile platform, and the computer software product includes an application program that can execute the scene recognition method provided in the embodiment of the present application.

[0048] On the other hand, the control device includes a processor. When executing the scene recognition method, the processor can perform the method by calling an executable program in a memory or by a logic operation circuit.

[0049] The control device may execute the scene recognition method in the background, or may display the method to the user via a graphical interface, or may execute part of the method in the background and display the other part to the user. Furthermore, the control device may execute the scene recognition method completely autonomously, or partially autonomously while requiring human intervention for the other part.

[0050] The control device may be disposed on a movable platform. Movable platforms include, but are not limited to, aircraft, unmanned aerial vehicles, vehicles, movable carts, or mobile robots. Unmanned aerial vehicles include, but are not limited to, unmanned helicopters, unmanned fixed-wing aircraft, unmanned multi-rotor aircraft, unmanned gliders, unmanned helicopter balloons, or the like.

[0051] Please refer to Figure 1. The movable platform 100 includes a body 10; a power system 20 installed on the body 10 for providing power to the movable platform; a detection device 30 installed on the body 10 for emitting detection light to detect the current scene; and the above-mentioned control device 40.

[0052] The body 10 refers to the structure of the mobile platform 100, including its outer shape, frame, and connecting components. The body 10 is a crucial component of the mobile platform 100, supporting and protecting the other components of the mobile platform 100 and determining its motion performance, stability, and adaptability.

[0053] The power system 20 can provide the required power and energy support for the movable platform 100 .

[0054] Taking an unmanned aerial vehicle as an example, please refer to Figure 2. The power system 20 may include one or more electronic speed regulators 21 (referred to as ESCs for short), one or more propellers 22, and one or more motors 23 corresponding to the one or more propellers, wherein the motor 23 is connected between the electronic speed regulator 21 and the propeller 22, and the motor 23 and the propeller 22 are arranged on the wings of the unmanned aerial vehicle.

[0055] Taking a fuel vehicle as an example, the power system 20 may include an internal combustion engine, which uses fuel for combustion and drives the vehicle by rotating the engine.

[0056] Taking new energy vehicles as an example, the power system 20 consists of a battery pack, a motor and an electronic control system. The battery pack supplies power to the motor, and the motor converts electrical energy into mechanical energy to drive the vehicle. Among them, the electronic control system is used to control the vehicle's power output and driving mode, as well as manage the battery's charging and discharging process.

[0057] The detection device 30 is an important tool for the movable platform 100 to perceive the surrounding environment. The detection light emitted by the detection device 30 can be used to detect the current scene in which the movable platform 100 is located, thereby assisting the subsequent movement decision process of the movable platform 100.

[0058] The detection device 30 may be a device that measures distance based on the time of flight (TOF) method of light. For example, the detection device 30 may be a 3D TOF device; or the detection device 30 may be a laser radar, an infrared radar, or the like.

[0059] Referring to Figure 3A , a detection device 30 may include a transmitter 31 and a receiver 32. The transmitter may emit detection light. The detection light is scattered, reflected, or refracted by objects in the environment. A portion of the detection light is reflected back after a period of time, returning to the receiver, which receives the reflected echo. The detection light may include, but is not limited to, infrared light and laser light.

[0060] Please refer to Figure 3B. The emitter 31 in the detection device 30 includes an emitting light source, which can be set as a surface light source. Specifically, the emitter can be enabled to emit detection light in the form of surface light through optical phased array, mechanical scanning, etc. Compared with a point light source, the coverage range of a surface light source is roughly cone-shaped or cylindrical (it can also be composed of rays emitted by multiple point light sources), which can provide a wider range of illumination coverage. By emitting the detection light emitted by the light source, at least a part of the current scene (including one or more reflective surfaces) can be illuminated, so that a more comprehensive measurement result of the current scene can be obtained based on the reflected light. Exemplarily, the surface light source can be implemented using optoelectronic devices such as solid-state laser tubes or light-emitting diodes (LEDs). Among them, different surface light sources correspond to different FOVs (Field of View), and the FOV of a surface light source indicates the range or angle that it can illuminate or radiate. The field of view of a surface light source depends on the nature of the light source and the application requirements. In actual applications, it may be necessary to set it according to the specific situation.

[0061] The receiver 32 in the detection device 30 comprises an area array photoelectric sensor. An area array photoelectric sensor consists of multiple photoelectric elements (e.g., photodiodes or photoresistors) arranged in a two-dimensional array. When light strikes a photoelectric element, each element generates and outputs a signal (an electrical signal). The signal strength is proportional to the light intensity.

[0062] For example, the detection device 30 can be installed at different positions in the movable platform 100, so as to detect in different directions of the movable platform 100. For example, the installation position of the detection device 30 in the movable platform 100 supports the detection device 30 to detect in the moving direction of the movable platform 100. In some embodiments, the detection device can be movably arranged relative to the main body of the movable platform 100, so that the detection device can change its own detection direction by changing its relative position with the main body of the movable platform 100.

[0063] In one example, please refer to Figures 3B and 4. Taking the movable platform 100 as an aircraft (manned aircraft or unmanned aircraft) as an example, the detection device 30 can be installed on the bottom of the aircraft, so that during the flight of the aircraft, the detection device 30 can detect the scene below the aircraft. It can be understood that the light source of the transmitter of the detection device 30 mentioned above can be set as a surface light source, and the surface light source can specifically form a field of view range of a pyramid or a cone, so there can be a corresponding field of view angle. And the aircraft can achieve one or more degrees of freedom of movement, such as rotating around one or more rotation axes (roll axis (Roll), yaw axis (Yaw) and pitch axis (Pitch)), then as the posture of the aircraft changes, the scene below the aircraft detected by the detection device 30 will also be different.

[0064] Referring to Figure 3B, the detection device 30 can be installed on the bottom of the aircraft. The detection device 30 is used to detect the scene below the aircraft, thereby determining the altitude of the aircraft. This requires that no matter what attitude the aircraft flies in, at least one of the detection lights emitted by the surface light source is perpendicular to the reflective surface of the scene below the aircraft and can be reflected back to the surface array photoelectric sensor, so that the altitude of the aircraft can be measured. Therefore, the field of view angle of the surface light source can be determined according to the attitude angle of the aircraft to meet the above requirements. For example, in order to achieve the ability to measure the altitude of the aircraft regardless of the attitude the aircraft flies in, the field of view angle of the surface light source in the detection device installed on the aircraft is greater than or equal to twice the maximum attitude angle of the aircraft.

[0065] In another example, please refer to Figure 5. Taking the movable platform 100 as a vehicle as an example, the detection device 30 can be installed at least one of the front and rear of the vehicle, so that when the vehicle moves, the detection device 30 can detect the scene in front of or behind the vehicle accordingly.

[0066] The control device 40 can control the detection device 30 to emit detection light to detect the current scene. When the area array photoelectric sensor receives the reflected echo of the detection light, the control device 40 can perform scene recognition based on the signal parameters of the multiple signals output by the multiple photoelectric units of the area array photoelectric sensor. In addition, the control device 40 can also make subsequent movement decisions based on the scene recognition and control the movement of the movable platform 100 based on the decision results.

[0067] Of course, the movable platform 100 may also include other types of sensors, such as at least one of a gyroscope, an ultrasonic sensor, an electronic compass, an inertial measurement unit (IMU), a visual sensor, a global navigation satellite system, and a barometer.

[0068] For example, the mobile platform 100 may also be connected to a remote control terminal for remotely controlling the mobile platform 100. The remote control terminal includes but is not limited to a mobile phone, a remote controller, a tablet, a personal digital assistant, a computer, or a wearable device.

[0069] The following is an exemplary description of the scene recognition method provided in the embodiment of the present application:

[0070] Please refer to Figure 6, which is a flow chart of a scene recognition method provided in an embodiment of the present application. The scene recognition method is described by applying it to a mobile platform as an example, and the mobile platform is equipped with a detection device. The method includes:

[0071] In S101 , a detection device is controlled to emit detection light to detect a current scene; the detection device includes an array photoelectric sensor.

[0072] As the movable platform moves, the detection device can be controlled to emit detection light, including but not limited to laser or infrared light. In some embodiments, the wavelength range of the detection light can be different from the wavelength range of visible light to reduce the impact of ambient light on recognition results. When the detection light encounters an object in the current environment, it is reflected back and, after a period of time, returns to the area array photoelectric sensor, which receives the reflected echo of the detection light.

[0073] In S102 , when the area array photoelectric sensor receives the reflected echo of the detection light, a plurality of signals output by a plurality of photoelectric units of the area array photoelectric sensor are acquired.

[0074] An area array photoelectric sensor consists of multiple photoelectric cells arranged in a two-dimensional array. When light strikes a cell, each generates and outputs a signal. This allows for the acquisition of multiple signals from the cells within the area array photoelectric sensor.

[0075] In S103 , it is identified whether the current scene is a target scene according to the signal parameters of the plurality of signals, where the target scene at least includes a water surface scene.

[0076] In this embodiment, scene determination can be performed directly based on the signal parameters of the multiple signals output by the multiple photoelectric units in the area array photoelectric sensor to identify whether the current scene is a water surface scene. This avoids the problem of passive sensors such as visual sensors being unable to accurately identify water surface scenes in low or strong light due to their requirements for ambient light brightness. Actively emitting detection light for water surface scene recognition improves scene recognition accuracy and the safety of the movable platform during movement. Furthermore, the recognition process is simple, enabling rapid acquisition of scene recognition results.

[0077] In some embodiments, the signal parameter includes at least one of signal strength and reflectivity corresponding to the signal output by the photoelectric unit. The detection device detects the current scene by actively emitting detection light and receiving reflected echoes via a planar array photoelectric sensor. The depth corresponding to the signal output by each photoelectric unit can be calculated based on the phase difference between the actively emitted detection light and the reflected echo; the signal strength corresponding to the signal output by each photoelectric unit can be calculated based on the reflected echo; and the reflectivity corresponding to the signal output by each photoelectric unit can be determined based on the signal strength and depth corresponding to the signal output by each photoelectric unit.

[0078] In this way, by actively emitting detection light and determining the current scene based on the signal strength and / or reflectivity corresponding to the reflected echo signal, the problem of being unable to accurately identify the water surface scene due to the lack of texture features on the water surface can be avoided, thereby improving the accuracy of identifying textureless / weak texture scenes.

[0079] In some embodiments, considering that different target scenes have their own scene characteristics, after the detection device detects different target scenes, the signal parameters of the multiple signals output by the multiple photoelectric units in the array photoelectric sensor also vary. Therefore, after obtaining the signal parameters corresponding to the multiple signals, the movable platform can identify whether the current scene is the target scene based on at least one of the distribution and change of the signal parameters of the multiple signals. This embodiment achieves accurate scene recognition based on the reflection of the scene characteristics of the target scene in the signal parameters, and the recognition process is simple, which helps to improve scene recognition efficiency.

[0080] Exemplarily, the distribution of the signal parameter includes at least one of the following: the spatial distribution of the signal parameter, and the distribution of the numerical value of the signal parameter. The change of the signal parameter includes at least one of the following: the spatial change of the signal parameter, and the temporal change of the signal parameter.

[0081] Taking a target scene including a water surface as an example, the reflective surface in a water surface scene is smooth, primarily involving specular reflection. Depending on the incident angle of light and the normal direction of the water surface, light will be reflected at an angle equal to the incident angle and within the same plane. This characteristic of the scene means that when a planar array photoelectric sensor receives reflected echoes of detection light, the signal output from the photoelectric units in one area will have high signal strength and reflectivity, while the signal output from the photoelectric units in other areas will have low signal strength and reflectivity. Therefore, based on at least one of the distribution and variation of the signal parameters of the multiple signals, it can be determined whether the current scene is a water surface scene.

[0082] Taking a target scene including fog as an example, in most foggy scenes, the fog accounts for more than 50%; while in most non-foggy scenes, the fog accounts for less than 50%. Fog has a relatively low reflectivity to light because water droplets and tiny particles in the fog scatter, absorb, and refract light, thereby reducing light reflection. This scene characteristic results in the signal strength and reflectivity of the signals output by most of the multiple photoelectric units of the area array photoelectric sensor when receiving the reflected echo of the detection light. Therefore, based on at least one of the distribution and variation of the signal parameters, it is possible to identify whether the current scene is foggy or cloudy.

[0083] In some embodiments, the multiple signals can generate at least a portion of a point cloud image, and each pixel in the point cloud image corresponds one-to-one to a photoelectric unit of the area array photoelectric sensor.

[0084] Exemplarily, point cloud images include but are not limited to depth maps, reflectivity maps, signal strength maps, or point cloud maps. (1) Each pixel in the depth map represents the distance to the detection device, and grayscale values ​​or color coding can be used to represent distance information; for example, in the depth map, brighter pixels represent closer objects, while darker pixels represent farther objects. (2) Each pixel in the reflectivity map represents the reflectivity of the detection position, and grayscale values ​​or color coding can be used to represent the magnitude of the reflectivity. (3) Each pixel in the signal strength map represents signal strength information, and grayscale values ​​or color coding can be used to represent the magnitude of the signal strength. (4) Each pixel in the point cloud map represents a point in three-dimensional space and contains the position of the point and other possible attribute information, such as signal strength, reflectivity, etc.

[0085] The detection device can emit detection light at a preset frequency and generate a single point cloud image based on multiple signals output by multiple photoelectric units of the area array photoelectric sensor within a time period. Multiple point cloud images can also be generated based on multiple signals output by multiple photoelectric units of the area array photoelectric sensor within multiple different time periods, where the multiple different time periods are continuous. The mobile platform can identify whether the current scene is the target scene based on the single point cloud image. Alternatively, to improve scene recognition accuracy, the mobile platform can identify whether the current scene is the target scene based on multiple point cloud images.

[0086] Exemplarily, the point cloud image includes signal strength information and / or reflectivity information; the signal strength information and reflectivity information of each pixel in the point cloud image correspond one-to-one to the signal parameters of the signal output by the photoelectric unit in the array.

[0087] In some possible implementations, the mobile platform may identify whether the current scene is the target scene based on the distribution and / or changes in signal strength information within a frame of point cloud images. Alternatively, the mobile platform may identify whether the current scene is the target scene based on the distribution and / or changes in reflectivity information within a frame of point cloud images.

[0088] In other possible implementations, to improve the accuracy of scene recognition, the mobile platform may identify whether the current scene is a target scene based on the distribution and / or changes in signal strength information in multiple frames of point cloud images. Alternatively, the mobile platform may identify whether the current scene is a target scene based on the distribution and / or changes in reflectivity information in multiple frames of point cloud images. For example, when it is determined that the reflectivity information of multiple consecutive frames of point cloud images all meets the scene characteristics of a target scene, the current scene is determined to be the target scene.

[0089] In other embodiments, in addition to performing scene recognition based on point cloud images, the movable platform can also perform scene recognition directly based on the signal parameters of the signals output by the photoelectric units. As mentioned above, the detection device can emit detection light at a preset frequency. In one possible implementation, the movable platform can identify whether the current scene is a target scene based on the distribution and / or changes in the signal parameters of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor within a time period. Alternatively, in another possible implementation, in order to improve the accuracy of scene recognition, it is also possible to identify whether the current scene is a target scene based on the distribution and / or changes in the multiple signals output by the multiple photoelectric units of the array photoelectric sensor within multiple different time periods; wherein the multiple different time periods are continuous time periods.

[0090] In some embodiments, the target scene may include at least one of a water surface scene, a cloud and fog scene, a land scene with low reflectivity, and a non-flat land scene.

[0091] (1) Here is an example of the recognition process of the water surface scene:

[0092] The reflective surface in the water scene is a smooth surface, which mainly involves specular reflection. Depending on the incident angle of the light and the normal direction of the water surface, the light will be reflected at an angle equal to the incident angle and within the same plane. In other words, specular reflection will cause the light to be reflected back at the same angle, resulting in a relatively high signal strength or reflectivity in specific areas, while the light in other areas may be reflected or absorbed at different angles, resulting in lower signal strength or reflectivity.

[0093] In some embodiments, water surface scene recognition can be performed based on the spatial variation of signal strength. The mobile platform can perform scene recognition based on the spatial variation and / or spatial distribution of signal strength information for multiple pixels in the point cloud image to determine whether the current scene is a water surface scene. Alternatively, the mobile platform can perform scene recognition based on the spatial variation and / or spatial distribution of the signal strength of multiple signals output by multiple photoelectric units of a planar array photoelectric sensor to determine whether the current scene is a water surface scene.

[0094] For example, a first preset condition related to signal strength can be set based on the aforementioned characteristics of the water surface scene. When the signal strengths corresponding to the multiple photoelectric units meet the first preset condition, the mobile platform determines that the current scene is a water surface scene. More specifically, the mobile platform can determine that the current scene is a target scene when the spatial variation and / or spatial distribution of the signal strengths corresponding to the multiple photoelectric units meet the first preset condition.

[0095] Referring to the multiple photoelectric units forming a two-dimensional array included in the area array photoelectric sensor shown in FIG7 , the first preset condition indicates that, within the same time period, the signal strengths of multiple first photoelectric units located in a first area are concentrated, the signal strengths corresponding to the multiple first photoelectric units are greater than the signal strengths corresponding to multiple second photoelectric units located in a second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or the first preset condition indicates that a region of photoelectric units has a signal strength gradient greater than a threshold. This embodiment performs water surface scene recognition based on the scenario characteristics of high signal strength in specific areas and low signal strength in other areas, thereby ensuring scene recognition accuracy.

[0096] The first region may be any region in a two-dimensional array composed of a plurality of photoelectric units, and this embodiment does not impose any limitation thereto. The first region shown in FIG. 7 is only for illustration.

[0097] Specifically, the first preset condition may include at least one of the following: (1) the proportion of first photoelectric units with signal strength higher than the first preset threshold is less than the first preset proportion; (2) the first photoelectric units with signal strength higher than the first preset threshold are concentrated in an area of ​​a preset size; (3) the proportion of second photoelectric units with signal strength lower than the second preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the second preset threshold is less than or equal to the first preset threshold; (4) there are multiple adjacent photoelectric units whose corresponding signal strength change gradient values ​​are greater than the third preset threshold.

[0098] It can be understood that the specific values ​​of the first preset threshold, the second preset threshold, the third preset threshold, the first preset proportion, the second preset proportion and the preset size of the area can be set according to the actual application scenario, and this embodiment does not impose any restrictions on this.

[0099] In some embodiments, water surface scene recognition can also be performed based on the spatial variation of reflectivity. The mobile platform can perform scene recognition based on the spatial variation and / or spatial distribution of reflectivity information of multiple pixels in the point cloud image to determine whether the current scene is a water surface scene. In other words, the mobile platform can perform scene recognition based on the spatial variation and / or spatial distribution of the reflectivity of multiple signals output by multiple photoelectric units of the area array photoelectric sensor to determine whether the current scene is a water surface scene.

[0100] For example, a second preset condition related to reflectivity can be set based on the aforementioned characteristics of the water surface scene. If the reflectivity corresponding to the multiple photoelectric units meets the second preset condition, the movable platform determines that the current scene is a water surface scene. More specifically, the movable platform can determine that the current scene is a target scene if the spatial variation and / or spatial distribution of the reflectivity corresponding to the multiple photoelectric units meets the second preset condition.

[0101] Referring to the multiple photoelectric units forming a two-dimensional array included in the area array photoelectric sensor shown in FIG7 , the second preset condition is used to indicate that, within the multiple photoelectric units in the same time period, the reflectivity of multiple first photoelectric units located in a first region is concentrated, the reflectivity corresponding to the multiple first photoelectric units is greater than the reflectivity corresponding to multiple second photoelectric units located in a second region outside the first region, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or the second preset condition is used to indicate that a region of photoelectric units has a reflectivity gradient greater than a threshold. This embodiment performs water surface scene recognition based on the scene characteristics of high reflectivity in specific regions and low reflectivity in other regions, thereby ensuring scene recognition accuracy.

[0102] Specifically, the second preset condition may include at least one of the following: (1) the proportion of first photoelectric units with a reflectivity higher than a fourth preset threshold is less than the first preset proportion; (2) the first photoelectric units with a reflectivity higher than the fourth preset threshold are concentrated in an area of ​​a preset size; (3) the proportion of second photoelectric units with a reflectivity lower than a fifth preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the fifth preset threshold is less than or equal to the fourth preset threshold; (4) there are multiple adjacent photoelectric units whose corresponding reflectivity change gradient values ​​are greater than the sixth preset threshold.

[0103] It is understandable that the specific values ​​of the fourth preset threshold, the fifth preset threshold, and the sixth preset threshold can be set according to the actual application scenario. Among them, considering that the reflectivity of the signal output by the photoelectric unit is also determined based on the signal strength depth corresponding to the signal, that is, there is a positive correlation between the reflectivity and signal strength of the signal output by the same photoelectric unit. The photoelectric unit with high signal strength has a correspondingly higher reflectivity. Therefore, the first preset proportion, the second preset proportion, and the area of ​​preset size in the first preset condition and the first preset proportion, the second preset proportion, and the area of ​​preset size in the second preset condition can be set to the same values. Of course, in some scenarios, the two can also be set to different values, and this embodiment does not impose any restrictions on this.

[0104] In some embodiments, water surface scene recognition can also be performed based on the numerical value of the signal strength (the numerical value of the signal strength in most areas is low). The first interval and the first preset range related to the signal strength can be set based on the above-mentioned scene characteristics of the water surface scene; among the signal strengths of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor, if the proportion of the signal strengths with values ​​in the first interval reaches the first preset range, it can be determined that the current scene is a water surface scene. It can be understood that the upper limit value of the first interval is less than or equal to the first preset threshold value in the above-mentioned first preset condition. The first preset range can be [90%, 100%]. This embodiment performs water surface scene recognition based on the scene characteristics that the numerical value of the signal strength in most areas is low, thereby ensuring the accuracy of scene recognition.

[0105] In some embodiments, water surface scene recognition can also be performed based on the numerical value of the reflectivity (the reflectivity values ​​in most areas are relatively low). A second interval and a second preset range related to the reflectivity can be set based on the above-mentioned scene characteristics of the water surface scene; among the reflectivities of multiple signals output by multiple photoelectric units of the array photoelectric sensor, if the proportion of reflectivity values ​​in the second interval reaches the second preset range, it can be determined that the current scene is a water surface scene. It can be understood that the upper limit value of the second interval is less than or equal to the fourth preset threshold in the above-mentioned second preset condition. The second preset range can be [90%, 100%], and the first preset range and the second preset range can be the same. This embodiment performs water surface scene recognition based on the scene characteristic that the reflectivity values ​​in most areas are relatively low, thereby ensuring the accuracy of scene recognition.

[0106] It can be understood that in actual application scenarios, the above four methods of identifying water surface scenes can be used in only one way or in any combination, and this embodiment does not impose any restrictions on this. A frame of point cloud image (or the signal parameters of multiple signals output by multiple photoelectric units in a time period) can be used to identify whether the current scene is a water surface scene in accordance with any one or more of the above methods; or, in order to improve the accuracy of scene recognition, multiple frames of point cloud images (or the signal parameters of multiple signals output by multiple photoelectric units in different time periods) can be used to identify whether the current scene is a water surface scene in accordance with any one or more of the above methods. For example, M frames of point cloud images are used for scene recognition in accordance with any one of the above methods, where at least N frames of point cloud images determine that the current scene is a water surface scene, and then it can be finally confirmed that the water surface scene has been identified, M>N, and N is an integer greater than 1; wherein, multiple frames of point cloud images can reflect the changes in signal parameters over time.

[0107] In some embodiments, in addition to the mirror reflection phenomenon in water surface scenes, there are other scenes such as ice and glass that also have mirror reflection phenomena. Therefore, in order to improve the recognition accuracy of water surface scenes, in addition to using at least one of the above methods to identify water surface scenes, it is also possible to further combine visual information for judgment. Exemplarily, the movable platform is also equipped with a visual sensor. The visual information collected by the visual sensor for the current scene can be obtained; then, based on the visual information and the signal parameters of multiple signals, it is identified whether the current scene is a water surface scene; wherein the visual information is at least used to exclude other mirror scenes other than the water surface scene. This embodiment realizes the exclusion of other mirror scenes based on visual information, which is conducive to improving the recognition accuracy of water surface scenes.

[0108] For example, a visual recognition model can be pre-trained. For example, this visual recognition model is obtained through supervised training based on positive and negative samples. The positive samples are visual information of water scenes, and the positive samples are labeled "yes." The negative samples are visual information of mirror scenes other than water scenes, and the positive samples are labeled "no." The trained visual recognition model is used to recognize water scenes based on the input visual information, and outputs two recognition results: "yes" or "no."

[0109] (2) Here is an example of the recognition process of cloud and fog scenes:

[0110] In most foggy scenes, the fog component is greater than 50%, while in most non-foggy scenes, it is less than 50%. Fog has a relatively low reflectivity to light. This is because the water droplets and tiny particles in the fog scatter, absorb, and refract light, reducing its reflection. The water droplets and particles in fog vary in size and shape, deflecting light from its original path and scattering it in different directions. This scattering causes light to propagate more randomly and dispersedly through fog, reducing direct reflections between the light and the observer. Furthermore, the water droplets and particles in the fog absorb some of the light's energy, gradually weakening it. This absorption effect is particularly pronounced in thicker clouds. These characteristics result in relatively low signal strength and reflectivity in the reflected echo.

[0111] In some embodiments, the aforementioned multiple signals can generate at least a portion of a point cloud image, with each pixel in the point cloud image corresponding one-to-one to a photoelectric unit of the area array photoelectric sensor. Scene recognition can then be performed based on the numerical distribution of at least one of the reflectivity information and signal strength information of the multiple pixels in the point cloud image to determine whether the current scene is foggy or cloudy. Alternatively, the mobile platform can perform scene recognition based on the numerical distribution of at least one of the reflectivity information and signal strength information of the multiple signals output by the multiple photoelectric units of the area array photoelectric sensor to determine whether the current scene is foggy or cloudy.

[0112] In one possible implementation, the cloud and fog scene can be judged by the number of cloud and fog points. The movable platform counts the number of cloud and fog points in at least one of the following ways: (1) Counting the number of multiple signals output by multiple photoelectric units of the array photoelectric sensor whose signal strength is less than the seventh preset threshold value to determine the number of cloud and fog points; (2) Counting the number of multiple signals whose reflectivity is less than the eighth preset threshold value to determine the number of cloud and fog points. After determining the number of cloud and fog points, if the number of cloud and fog points reaches the third preset range, it is determined that the current scene is a cloud and fog scene. Exemplarily, the lower limit of the third preset range is greater than or equal to half of the total number of photoelectric units contained in the array photoelectric sensor, that is, the proportion of cloud and fog points exceeds 50%.

[0113] It is understood that the seventh and eighth preset thresholds can be specifically set based on actual application scenarios, and this embodiment does not impose any restrictions on this. For example, taking reflectivity as an example, the reflectivity of objects in a natural environment to light signals is greater than 5%, while the reflectivity of fog to light signals is generally less than 2%. There may be a small number of areas with reflectivity between 2% and 5%. In this case, the eighth preset threshold can be set to 5%.

[0114] In another possible implementation, foggy scenes can be identified based on the low reflectivity of light signals from foggy scenes. A third preset condition can be set based on this characteristic of foggy scenes. The mobile platform can identify the current scene as foggy if the proportion of reflectivity values ​​of multiple signals in different intervals meets the third preset condition. The third preset condition is used to indicate that the reflectivity values ​​corresponding to the multiple signals are concentrated within a preset interval.

[0115] Exemplarily, the different intervals may include a third interval and a fourth interval, the upper limit of the third interval is less than or equal to the lower limit of the fourth interval; the third preset condition includes: the proportion of the reflectivity of multiple signals in the third interval is greater than the proportion in the fourth interval, and the sum of the proportions of the reflectivity of multiple signals in the third interval and the fourth interval is greater than the third preset proportion.

[0116] For example, as mentioned above, the reflectivity of objects in a natural environment to light signals is greater than 5%, while the reflectivity of fog to light signals is typically less than 2%. There may be a small number of areas with reflectivity between 2% and 5%. When fog is thick, the reflectivity of some areas may be greater than 5%. For example, the upper limit of the third interval is 2%, the lower limit of the fourth interval is 2%, and the upper limit can be set to 5% or greater. The third preset condition is used to indicate that the reflectivity values ​​of multiple signals are concentrated within the third and fourth intervals, and are primarily distributed in the third interval.

[0117] It can be understood that a frame of point cloud image (or the signal parameters of multiple signals output by multiple photoelectric units in a time period) can be used to identify whether the current scene is a foggy scene in accordance with any one or more of the above methods; or, in order to improve the accuracy of scene recognition, multiple frames of point cloud images (or the signal parameters of multiple signals output by multiple photoelectric units in different time periods) can be used to identify whether the current scene is a foggy scene in accordance with any one or more of the above methods. For example, M frames of point cloud images are used for scene recognition in accordance with any one of the above methods, and if at least N frames of point cloud images determine that the current scene is a foggy scene, it can be finally confirmed that the foggy scene has been identified, M>N, and N is an integer greater than 1; wherein, multiple frames of point cloud images can reflect the temporal changes of signal parameters.

[0118] In some embodiments, water surface scenes may also have relatively low reflectivity and signal strength, but water surface scenes primarily involve specular reflection, while foggy scenes primarily involve diffuse reflection. Water surface scenes may experience sudden changes in signal strength or reflectivity, while foggy scenes generally have lower signal strength or reflectivity. Therefore, whether the current scene is a water surface scene can be identified based on any of the aforementioned water surface scene identification methods. If the scene is determined to be a non-water surface scene, whether the current scene is a foggy scene can be identified based on any of the aforementioned foggy scene identification methods.

[0119] Exemplarily, the movable platform can determine whether the current scene is a water surface scene based on the first signal parameters of multiple signals. When the current scene is a non-water surface scene, it can identify whether the current scene is a foggy scene based on the second signal parameters of multiple signals. The first signal parameter is different from the second signal parameter; for example, the first signal parameter is the signal strength corresponding to the signal, and the second signal parameter is the reflectivity corresponding to the signal. Alternatively, the first signal parameter and the second signal parameter can also be the same, such as both are signal strength or both are reflectivity. The specific identification method of the water surface scene and the foggy scene can be found in the above description and will not be repeated here. When judging the foggy scene, this embodiment first excludes the water surface scene situation, which is conducive to improving the recognition accuracy of the foggy scene.

[0120] In some embodiments, in addition to identifying whether the current scene is a water surface scene according to any of the above-mentioned water surface scene recognition methods, other sensors can also be used, such as a barometer, to make a judgment to detect whether the air pressure value output by the barometer meets the water surface pressure conditions. If not, the water surface scene can be excluded.

[0121] In some embodiments, the purpose of scene judgment is to determine whether the depths of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor are usable. Considering that fog does not actually hinder the movement of the movable platform, the depths of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor are invalid. In order to improve processing efficiency, a detection threshold can be set. Before performing fog scene recognition, it is detected whether the reflectivity of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor is less than the detection threshold. If so, it is determined that the multiple signals output by the multiple photoelectric units of the array photoelectric sensor are invalid (that is, no useful depth information can be obtained). Then, there is no need to perform fog scene recognition again, and it can be directly determined that the depth information corresponding to the signal is unavailable.

[0122] The detection threshold may be a reflectivity lower than 2%. For example, if multiple consecutive point cloud image frames identify the current scene as a foggy scene, the detection threshold may be further increased, eliminating the need for foggy scene recognition, thereby further improving processing efficiency.

[0123] (3) Here we provide an example of the recognition of low-reflectivity land scenes:

[0124] The inventors discovered that, unlike foggy scenes, foggy scenes typically have a reflectivity of less than 2% for light signals. However, some land scenes with low reflectivity, such as asphalt roads, have a reflectivity of light signals mainly concentrated between 2% and 5%, so there is no sudden change in reflectivity. Therefore, a fourth preset condition can be set based on the characteristics of low-reflectivity land scenes. After acquiring multiple signals output by multiple photoelectric units of a planar array photoelectric sensor, if the reflectivity of the multiple signals meets the fourth preset condition, the current scene is determined to be a low-reflectivity land scene. The fourth preset condition includes: among the reflectivities of the multiple signals, the proportion of reflectivity values ​​in the fifth interval is greater than the fourth preset proportion; and / or the reflectivity gradient changes corresponding to the multiple signals are all less than a threshold.

[0125] Exemplarily, the lower limit of the fifth interval may be less than or equal to 2%, and the upper limit may be greater than or equal to 5%, and the fourth preset proportion may be greater than 50%.

[0126] It should be understood that due to the rich feature points / textures in the visual sensor, clouds and fog often cause the visual sensor to mistakenly identify clouds and fog as physical obstacles, causing the aircraft to trigger obstacle avoidance actions.

[0127] The embodiment of the present application actively emits detection light. Since the signal intensity and reflectivity parameters of the echo emitted by clouds and fog have distribution and / or variation characteristics, the cloud and fog scene can be identified based on the distribution and / or variation information of the signal parameters of multiple signals, so as to reduce the aircraft's unexpected actions such as deceleration and hovering caused by erroneous triggering of obstacle avoidance in cloud and fog scenes, thereby improving the aircraft's flight safety and flight operation efficiency.

[0128] (4) Here is an example of the recognition process of non-flat scenes:

[0129] The reflectivity of light signals in non-flat terrain scenes is generally high. Therefore, a fifth preset condition can be set based on the characteristics of non-flat terrain scenes. After acquiring multiple signals output by multiple photoelectric units of the area array photoelectric sensor, if the reflectivity of the multiple signals meets the fifth preset condition, the current scene is determined to be a non-flat terrain scene. The fifth preset condition includes: among the reflectivity values ​​of the multiple signals, the proportion of reflectivity values ​​in a sixth range is greater than a fifth preset proportion. For example, the lower limit of the sixth range can be greater than or equal to 5%, and the fifth preset proportion can be greater than 50%.

[0130] In non-flat terrain scenarios, the sensing range of the area array sensor can be used to identify terrain undulations, thereby further improving the safety of the mobile platform during operation.

[0131] In some embodiments, when the current scene is identified as the target scene, prompt information corresponding to the target scene is output. The prompt information may be at least one of visual information and auditory information. In some embodiments, the prompt information may also be an electrical signal indicating the target scene or an identifier in communication signaling. The prompt information may be output directly by the mobile platform; alternatively, the prompt information may be output via a remote control terminal communicatively connected to the mobile platform. The remote control terminal may be equipped with relevant devices for outputting prompt information, such as a display or a speaker.

[0132] For example, if the current scene is a foggy scene, the depth corresponding to the signal output by the photoelectric unit of the array photoelectric sensor may be the distance between the movable platform and the fog, but the fog does not actually hinder the movement of the movable platform, so the depth is invalid; then, instead of outputting the depth information, a prompt message indicating that "the current scene is a foggy scene and you can continue to pass" may be output.

[0133] For example, if the current scene is other than a foggy scene (such as a water surface, low-reflectivity land, or uneven terrain), and the water surface, low-reflectivity land, or uneven terrain affect the movement or landing of the movable platform, then the depth corresponding to the signal output by the photoelectric unit of the array photoelectric sensor is valid, and movement decisions can be made based on this depth, and prompt information corresponding to the target scene can be output. For example, the output prompt information for a water surface scene may be "Caution landing on water surface scene."

[0134] In some embodiments, referring to FIG3B and FIG4 , the detection device can be mounted on the bottom of the movable platform, or positioned toward the bottom of the movable platform. The current scene detected by the detection device can include the scene below the movable platform carrying the detection device, and the height of the movable platform can be determined by the detection device 30. Since the detection device has a field of view, no matter what attitude the movable platform flies in, and no matter whether the reflection from the reflective surface is diffuse reflection or specular reflection, as long as the field of view is large enough, there will always be a light beam perpendicular to the reflective surface of the current scene that can be reflected back to the area array photoelectric sensor. In this way, the height of the movable platform can be determined based on the light signal reflected back from the light beam perpendicular to the reflective surface of the current scene.

[0135] Furthermore, if the current scene is a specular reflection scene, if the detection device is located in front of or behind the movable platform, when detecting the scene diagonally below the movable platform, the detection light will be unable to receive the reflected echo due to specular reflection, resulting in an inability to effectively identify the target scene. In particular, since aircraft have the ability to move in the vertical direction, it is also very important to collect scene information above and below the aircraft.

[0136] In such scenarios, by installing the detection device at the bottom of the movable platform, or setting it toward the bottom of the movable platform, the movable platform can still receive reflected light in a mirror reflection scenario, so as to effectively identify the scene below the movable platform.

[0137] Therefore, when identifying the current scene as a scene other than a foggy scene (such as a water surface scene, a land scene with low reflectivity, and a non-flat ground scene, etc.), the movable platform can determine the height information of the movable platform based on the minimum value of the depths corresponding to the multiple signals output by the photoelectric unit of the array photoelectric sensor, and then output the height information of the movable platform.

[0138] For example, if the reflective surface of the current scene is a plane, the minimum value of the depths corresponding to the multiple signals output by the photoelectric units of the area array photoelectric sensor is the height of the movable platform.

[0139] For example, if the reflective surface of the current scene is an undulating surface, the movable platform needs to further combine the current posture of the movable platform to determine the height information of the movable platform according to the minimum value of the depths corresponding to the multiple signals and the current posture of the movable platform.

[0140] 3B and 4 , in some embodiments, the movable platform may be an aircraft. In order to measure the height of the aircraft from the reflective surface of the current scene in any attitude of the aircraft, that is, to achieve full coverage of the attitude angle, it is necessary to determine the field of view angle of the detection device provided on the movable platform according to the maximum attitude angle of the aircraft. Exemplarily, the field of view angle of the detection device provided on the aircraft is greater than or equal to twice the maximum attitude angle of the aircraft. The maximum attitude angle can be understood as the maximum angle between the propeller plane and the horizontal plane during the flight of the aircraft. In the case where the aircraft is a single-rotor aircraft, the propeller plane may be a plane passing through the center of the blade and perpendicular to the blade's axis of rotation. In the case where the aircraft is a multi-rotor aircraft, the propeller plane may be a plane formed by the line connecting the centers of multiple blades.

[0141] The following is an example of the control process of the mobile platform in different target scenarios:

[0142] (1) If the target scene is a foggy scene.

[0143] As mentioned above, fog and clouds do not actually hinder the movement of the movable platform, and therefore the depth corresponding to the signal output by the photoelectric unit of the area array photoelectric sensor is invalid. If the current scene is identified as fog and clouds, the movable platform can be controlled to move at its original speed. Alternatively, for safety reasons, the movable platform can be controlled to move at a speed lower than the original speed for a preset period of time before returning to its original speed. This embodiment allows the movable platform to move normally even when fog and clouds are identified, without the need for obstacle avoidance, thereby improving control accuracy.

[0144] (2) If the target scene is a water surface scene.

[0145] In the water surface scenario, the signal parameters of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor also include depth information, and the depth information represents the distance between the movable platform and the water surface. The water surface has an impact on the movement or landing of the movable platform, and the depth information is valid.

[0146] In some embodiments, the movable platform includes an aircraft; when the current scene is identified as a water surface, the aircraft can be controlled to maintain hovering at a preset height above the water surface. For example, the distance between the aircraft and the water surface can be determined based on the depths corresponding to the multiple signals, and the aircraft can then be controlled to maintain hovering at a preset height above the water surface based on the distance between the aircraft and the water surface. This embodiment enables hovering over the water surface, preventing the aircraft from accidentally landing on the water surface and ensuring the safety of the aircraft.

[0147] In some embodiments, the movable platform includes an aircraft; when the current scene is identified as a water surface scene, the aircraft can identify whether there are waves in the water surface scene based on the distribution and / or changes in the depths corresponding to the multiple signals. Exemplarily, the distribution of the depths corresponding to the multiple signals includes at least one of the following: the spatial distribution of the depths corresponding to the multiple signals, and the distribution of the numerical values ​​of the depths corresponding to the multiple signals. The changes in the depths corresponding to the multiple signals include at least one of the following: the spatial changes in the depths corresponding to the multiple signals, and the temporal changes in the depths corresponding to the multiple signals.

[0148] For example, considering the phenomenon of wave fluctuations, different areas may experience different heights at the same time. To account for this scenario, if, during time period T, the depths of signals output by the multiple photoelectric units of the area array photoelectric sensor differ from one another, while the depths of signals output by the photoelectric units in the same area are similar, then waves are determined to be present on the water surface.

[0149] For example, considering the phenomenon of wave fluctuations, the same area may experience waves rising and then receding at different times. Based on this scenario characteristic, if the depth of the signal output by the photoelectric units in a certain area of ​​the area array photoelectric sensor during time period T is different from the depth of the signal output by the photoelectric units in the same area of ​​the area array photoelectric sensor during time period T+1, the presence of waves on the water surface is determined.

[0150] If waves are detected in the water scene, the aircraft can be controlled to maintain a preset distance from the waves during the hovering process to prevent the waves from affecting the aircraft (such as waves hitting the aircraft and causing certain parts to malfunction or crash). Alternatively, the aircraft can be controlled to maintain a preset distance from the water surface during the hovering process to ensure that the hovering position of the aircraft does not change due to waves during the hovering process on the water surface, thereby ensuring the stability of the aircraft's hovering. Alternatively, the aircraft can also detect through depth information that the distance to the waves is less than a threshold, and then control the aircraft to move away from the waves, etc. The specific settings can be made according to actual conditions.

[0151] In some embodiments, the movable platform includes an aircraft. When the current scene is identified as a water surface scene, it can be determined that the depth information corresponding to the multiple signals is valid. The aircraft can identify whether there are water objects and / or underwater objects in the water surface scene based on the distribution and / or changes in the depths corresponding to the multiple signals. Exemplarily, the distribution of the depths corresponding to the multiple signals includes at least one of the following: the spatial distribution of the depths corresponding to the multiple signals, and the distribution of the numerical values ​​of the depths corresponding to the multiple signals. The changes in the depths corresponding to the multiple signals include at least one of the following: the spatial changes in the depths corresponding to the multiple signals, and the temporal changes in the depths corresponding to the multiple signals.

[0152] For example, considering the regional distribution of objects, if the depth of the signals output by the photoelectric units in a certain area among the multiple photoelectric units of the area array photoelectric sensor in the T time period is higher than the depth of the water surface, it can be determined that there is an object above the water; or, if the depth of the signals output by the photoelectric units in a certain area is lower than the depth of the water surface, it can be determined that there is an underwater object.

[0153] In one possible implementation, if an object on water is identified, the aircraft's landing location can also be determined based on the location of the object on water. The location of the object on water can be determined based on the spatial distribution of the depths of multiple signals output by multiple photoelectric units of the area array photoelectric sensor. This embodiment enables landing after an object on water is identified, preventing the aircraft from accidentally landing on the water surface and ensuring aircraft safety.

[0154] For example, to improve landing accuracy and safety, visual information can be combined to identify whether an object on the water is a landing point. Visual information collected by a visual sensor regarding the current scene can be obtained; if the object on the water is identified as a landing point based on the visual information, the landing position of the aircraft is determined based on the position of the object on the water.

[0155] In the process of landing at the location of the water object, the aircraft can be controlled to land slowly at a speed lower than the original speed to ensure the safety of the aircraft.

[0156] (3) If the target scene is a non-flat scene.

[0157] In a non-flat ground scenario, the signal parameters of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor also include depth information, and the depth information represents the distance between the movable platform and the non-flat ground. The non-flat ground has an impact on the movement or landing of the movable platform, and the depth information is valid.

[0158] In some embodiments, if the current scene is identified as non-flat, the depth distribution of multiple signals and the current posture of the movable platform can be used to determine whether the current scene contains stairs or slopes, and a prompt message can be output based on the identification result. For example, the prompt message includes but is not limited to "There is a slope here, please control carefully to avoid collision."

[0159] In some embodiments, the movable platform includes an aircraft. When the current scene is identified as non-flat, the aircraft can control the aircraft to maintain a preset distance from a slope or step by using the depth distribution of multiple signals and the aircraft's current attitude during the execution of tasks such as terrain simulation flight, thereby improving operational accuracy and efficiency.

[0160] Exemplarily, the movable platform includes an aircraft. Upon detecting that the current scene contains stairs or a slope, the aircraft can be controlled to hover while maintaining a preset distance from the stairs or slope, or to continue flying while maintaining a preset distance from the stairs or slope. This embodiment utilizes the recognition of stairs or slopes to achieve safe control of the aircraft and avoid collisions.

[0161] In another possible implementation, the movable platform includes an aircraft. If the current scene is identified as non-flat, the system can determine whether the current scene contains a landing guide slot based on the depth distribution of multiple signals and the current posture of the movable platform. If the current scene contains a landing guide slot, the aircraft is controlled to land within the guide slot. This embodiment enables precise landing of the aircraft through the identification of the airport guide slot.

[0162] (4) If the target scene is a land scene with low reflectivity.

[0163] In a low-reflectivity land scene, the signal parameters of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor also include depth information, and the depth information represents the distance between the movable platform and the land. The land has an impact on the movement or landing of the movable platform, and the depth information is valid.

[0164] In some embodiments, the movable platform includes an aircraft, and the aircraft can be controlled to land safely on land with low reflectivity according to the depths of the multiple signals output by the multiple photoelectric units of the area array photoelectric sensor.

[0165] The various technical features in the above embodiments can be arbitrarily combined as long as there is no conflict or contradiction between the combinations of features. Therefore, the arbitrary combination of the various technical features in the above embodiments also falls within the scope of disclosure of this specification.

[0166] Referring to FIG8 , this embodiment of the present application further provides another scene recognition method, including:

[0167] In S201 , a detection device is controlled to emit detection light to detect a current scene, wherein the detection device includes an array photoelectric sensor.

[0168] In S202 , when the area array photoelectric sensor receives the reflected echo of the detection light, a plurality of signals output by a plurality of photoelectric units of the area array photoelectric sensor are acquired.

[0169] In S203 , it is identified whether the current scene is a target scene based on the distribution and / or change of the signal parameters of the multiple signals.

[0170] This embodiment takes into account the characteristics of different target scenes, so that the signal parameters of the multiple signals output by the multiple photoelectric units of the array photoelectric sensor are different. Therefore, by actively emitting detection light, it is possible to directly identify whether the current scene is the target scene based on the distribution and / or changes in the signal parameters of the multiple signals of the received reflected echo. This can avoid the problem that passive sensors such as visual sensors cannot recognize the current scene in low light or strong light scenes due to their requirements for ambient light brightness. The active emission of detection light for scene recognition improves the accuracy of scene recognition and the safety of the movable platform during movement. The recognition steps are simple, which helps to quickly obtain scene recognition results.

[0171] In some embodiments, the signal parameter includes a signal strength corresponding to the signal and / or a reflectivity, wherein the reflectivity is determined based on the signal strength corresponding to the signal and the depth corresponding to the signal.

[0172] In some embodiments, the distribution of the signal parameter includes the spatial distribution of the signal parameter and / or the distribution of the numerical value of the signal parameter. The change of the signal parameter includes the spatial change of the signal parameter and / or the temporal change of the signal parameter.

[0173] In some embodiments, the multiple signals can generate at least a portion of a point cloud image, with each pixel in the point cloud image corresponding one-to-one to a photoelectric unit of the area array photoelectric sensor. The point cloud image includes signal strength information and / or reflectivity information; the signal strength information and reflectivity information of each pixel in the point cloud image correspond one-to-one to signal parameters of the signals output by the photoelectric units in the area array. S203 then includes: identifying whether the current scene is a target scene based on the distribution and / or variation of the signal strength information and / or reflectivity information in one or more frames of the point cloud image.

[0174] In some embodiments, the target scene includes a water surface scene, and S203 includes: performing scene recognition based on the spatial variation and / or spatial distribution of signal strength information of multiple pixel points in the point cloud image to identify whether the current scene is a water surface scene; and / or performing scene recognition based on the spatial variation and / or spatial distribution of reflectivity information of multiple pixel points in the point cloud image to identify whether the current scene is a water surface scene.

[0175] In some embodiments, the target scene includes a water surface scene, and S203 includes: when the signal strength corresponding to the plurality of photoelectric units satisfies a first preset condition, determining that the current scene is a water surface scene; the first preset condition is used to characterize that, among the plurality of photoelectric units in the same time period, the signal strengths of the plurality of first photoelectric units located in the first area are concentratedly distributed, the signal strengths corresponding to the plurality of first photoelectric units are greater than the signal strengths corresponding to the plurality of second photoelectric units located in the second area outside the first area, and the number of the plurality of first photoelectric units is less than the number of the plurality of second photoelectric units; and / or, the first preset condition is used to characterize a photoelectric unit area where there is a signal strength gradient change greater than a threshold.

[0176] The first preset condition includes at least one of the following: (1) the proportion of first photoelectric units with signal strength higher than a first preset threshold is less than the first preset proportion; (2) the first photoelectric units with signal strength higher than the first preset threshold are concentrated in an area of ​​a preset size; (3) the proportion of second photoelectric units with signal strength lower than a second preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the second preset threshold is less than or equal to the first preset threshold; (4) the change gradient value of the signal strength corresponding to multiple adjacent photoelectric units is greater than the third preset threshold.

[0177] In some embodiments, the target scene includes a water surface scene, and S203 includes: if the reflectivity corresponding to the plurality of photoelectric units satisfies a second preset condition, determining that the current scene is a water surface scene;

[0178] The second preset condition is used to characterize that among multiple photoelectric units in the same time period, the reflectivities of multiple first photoelectric units located in the first area are concentratedly distributed, the reflectivities corresponding to the multiple first photoelectric units are greater than the reflectivities corresponding to multiple second photoelectric units located in the second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the second preset condition is used to characterize a photoelectric unit area where there is a reflectivity gradient change greater than a threshold.

[0179] The second preset condition includes at least one of the following: (1) the proportion of the first photoelectric units with a reflectivity higher than the fourth preset threshold is less than the first preset proportion; (2) the first photoelectric units with a reflectivity higher than the fourth preset threshold are concentrated in an area of ​​a preset size; (3) the proportion of the second photoelectric units with a reflectivity lower than the fifth preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the fifth preset threshold is less than or equal to the fourth preset threshold; (4) there are multiple adjacent photoelectric units whose corresponding reflectivity change gradient values ​​are greater than the sixth preset threshold.

[0180] In some embodiments, the target scene includes a water surface scene. S203 includes: determining that the current scene is a water surface scene if, among the signal strengths of the multiple signals, the proportion of signal strengths within a first interval reaches a first preset range; and / or determining that the current scene is a water surface scene if, among the reflectivity values ​​of the multiple signals, the proportion of reflectivity values ​​within a second interval reaches a second preset range.

[0181] In some embodiments, the target scene also includes a foggy scene. S203 includes: counting the number of multiple signals whose signal strengths are less than a seventh preset threshold to determine the number of foggy points; and / or counting the number of multiple signals whose reflectivity is less than an eighth preset threshold to determine the number of foggy points. If the number of foggy points reaches a third preset range, the current scene is determined to be a foggy scene.

[0182] In some embodiments, the target scene also includes a foggy scene. S203 includes: when the proportion of the numerical values ​​of the reflectivity of multiple signals in different intervals meets the third preset condition, identifying the current scene as a foggy scene, the third preset condition is used to characterize that the numerical values ​​of the reflectivity corresponding to the multiple signals are concentrated in the preset interval. Wherein, the different intervals include the third interval and the fourth interval, and the upper limit of the third interval is less than or equal to the lower limit of the fourth interval; the third preset condition includes: the proportion of the reflectivity of the multiple signals in the third interval is greater than the proportion in the fourth interval, and the sum of the proportions of the reflectivity of the multiple signals in the third interval and the fourth interval is greater than the third preset proportion.

[0183] In some embodiments, the target scene further includes a foggy scene. S203 includes: if the current scene is determined to be a non-water surface scene based on first signal parameters of the multiple signals, identifying whether the current scene is a foggy scene based on second signal parameters of the multiple signals, wherein the first signal parameters are different from the second signal parameters.

[0184] In some embodiments, the method is applied to a movable platform, and the method further includes: when the current scene is identified as a water surface scene, controlling the movable platform to remain hovering at a preset height from the water surface.

[0185] In some embodiments, the method further includes: obtaining visual information collected by a visual sensor of the current scene. S203 includes: identifying whether the current scene is a water surface scene based on the visual information and signal parameters of the multiple signals; wherein the visual information is at least used to exclude other mirror scenes other than the water surface scene.

[0186] In some embodiments, the method is applied to a mobile platform, and the method further includes: when the current scene is identified as the target scene, outputting prompt information corresponding to the target scene.

[0187] In some embodiments, the method is applied to a movable platform, and the method further includes: when the current scene is identified as a foggy scene, controlling the movable platform to move at the original speed, or controlling the movable platform to move at the original speed after moving at a speed lower than the original speed for a preset period of time.

[0188] In some embodiments, the signal parameters of the plurality of signals further include depth information.

[0189] In some embodiments, it also includes: when the current scene is identified as a water surface scene, identifying whether there are water objects and / or underwater objects in the water surface scene based at least on the distribution and / or changes of depths corresponding to multiple signals.

[0190] In some embodiments, the method is applied to a movable platform, and the method further includes: if an aquatic object is identified in the water scene, determining a landing position of the movable platform according to the position of the aquatic object.

[0191] In some embodiments, the method further includes: when the current scene is identified as a water surface scene, identifying whether there are waves in the water surface scene based on the distribution and / or change of the depths of the multiple signals.

[0192] In some embodiments, the method is applied to a movable platform, and the method also includes: if waves are identified in the water surface scene, controlling the movable platform to maintain a preset distance from the waves during the hovering process; or, if waves are identified in the water surface scene, controlling the movable platform to maintain a preset distance from the water surface during the hovering process.

[0193] In some embodiments, the current scene is a scene below a movable platform carrying a detection device.

[0194] In some embodiments, the method further includes: outputting height information of the movable platform; the height information of the movable platform is determined according to a minimum value of depths corresponding to the plurality of signals.

[0195] In some embodiments, the height information of the movable platform is determined according to the minimum value of the depths respectively corresponding to the multiple signals and the current posture of the movable platform.

[0196] In some embodiments, the field of view angle of the detection device disposed on the movable platform is determined according to the maximum attitude angle of the movable platform.

[0197] In some embodiments, the field of view angle is greater than or equal to twice the maximum attitude angle of the movable platform.

[0198] In some embodiments, the target scene also includes a non-flat ground scene.

[0199] In some embodiments, the method further includes: identifying whether the current scene contains steps or slopes based on the depth distribution of the multiple signals and the current posture of the movable platform, and outputting prompt information based on the identification result.

[0200] In some embodiments, the signal parameters include the depth corresponding to the signal; and further include: identifying whether the current scene contains an airport guide groove capable of landing based on the distribution of the depths of multiple signals and the current posture of the movable platform; if the current scene contains an airport guide groove capable of landing, controlling the movable platform to land in the guide groove.

[0201] In some embodiments, the target scene also includes a low-reflectivity land scene. S203 includes: if the reflectivity of the multiple signals satisfies a fourth preset condition, determining that the current scene is a low-reflectivity land scene; the fourth preset condition includes: among the reflectivities of the multiple signals, a proportion of reflectivity values ​​in a fifth interval is greater than a fourth preset proportion; and / or the reflectivity gradient changes corresponding to the multiple signals are all less than a threshold.

[0202] Referring to FIG. 9 , an embodiment of the present application further provides a scene recognition method, including:

[0203] In S301, an echo signal of a reflective surface in a current scene is obtained, wherein the echo signal is an optical signal reflected back by a reflective surface when a detection device actively emits detection light, or an electrical signal converted from an optical signal.

[0204] In S302 , reflection information of the reflecting surface is determined according to the echo signal, wherein the reflection information includes at least one of the following: distribution information and change information of signal parameters of multiple echo signals of the reflecting surface.

[0205] In S303, it is determined whether the current scene is a target scene according to the reflection information of the reflective surface. The target scene at least includes a water surface scene.

[0206] This embodiment takes into account the scene characteristics of different target scenes, so that the reflection information corresponding to the multiple echo signals reflected by the reflecting surface is different. Therefore, based on the distribution and / or change of the signal parameters of the multiple echo signals reflected by the reflecting surface, it is possible to directly identify whether the current scene is the target scene. This can avoid the problem that passive sensors such as visual sensors cannot recognize the current scene in low light or strong light scenes due to the requirements of ambient light brightness. The scene recognition is realized by non-visual means, which improves the accuracy of scene recognition and the safety of the movable platform during movement. The recognition steps are simple, which helps to quickly obtain scene recognition results. Water surface scene recognition can also be performed based on the scene characteristics of the water surface scene.

[0207] In some embodiments, the detection device may also include a planar array photoelectric sensor, as in the above-described embodiments. The detection device's light source may be configured as a planar light source. The echo signal reflected from the reflective surface may be acquired by the detection device's planar array photoelectric sensor. The details are similar to those in the above-described embodiments and will not be further described here. It should be understood that the methods in the above-described embodiments may be combined with those in the present embodiment.

[0208] In some embodiments, the signal parameter includes a signal strength corresponding to the signal and / or a reflectivity, wherein the reflectivity is determined based on the signal strength corresponding to the signal and the depth corresponding to the signal.

[0209] In some embodiments, the distribution of the signal parameter includes the spatial distribution of the signal parameter and / or the distribution of the numerical value of the signal parameter. The change of the signal parameter includes the spatial change of the signal parameter and / or the temporal change of the signal parameter.

[0210] In some embodiments, the multiple reflection signals of the reflective surface are signals output by multiple photoelectric units on a planar array photoelectric sensor.

[0211] In some embodiments, multiple signals can generate at least a portion of a point cloud image, and each pixel in the point cloud image corresponds one-to-one to a photoelectric unit of the array photoelectric sensor; S303 includes: identifying whether the current scene is a target scene based on one or more frames of point cloud images.

[0212] In some embodiments, the point cloud image includes signal strength information and / or reflectivity information; the signal strength information and reflectivity information of each pixel in the point cloud image correspond one-to-one to the signal parameters of the signal output by the photoelectric unit in the array; S303 includes: performing scene recognition based on the spatial variation and / or spatial distribution of the signal strength information of multiple pixel points in the point cloud image to identify whether the current scene is a water surface scene; and / or performing scene recognition based on the spatial variation and / or spatial distribution of the reflectivity information of multiple pixel points in the point cloud image to identify whether the current scene is a water surface scene.

[0213] In some embodiments, S303 includes: when the signal strength corresponding to multiple photoelectric units meets a first preset condition, determining that the current scene is a water surface scene; the first preset condition is used to characterize that among multiple photoelectric units in the same time period, the signal strength of multiple first photoelectric units located in a first area is concentrated, the signal strength corresponding to the multiple first photoelectric units is greater than the signal strength corresponding to the multiple second photoelectric units located in a second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the first preset condition is used to characterize a photoelectric unit area where there is a signal strength gradient change greater than a threshold.

[0214] The first preset condition includes at least one of the following: (1) the proportion of first photoelectric units with signal strength higher than a first preset threshold is less than the first preset proportion; (2) the first photoelectric units with signal strength higher than the first preset threshold are concentrated in an area of ​​a preset size; (3) the proportion of second photoelectric units with signal strength lower than a second preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the second preset threshold is less than or equal to the first preset threshold; (4) the change gradient value of the signal strength corresponding to multiple adjacent photoelectric units is greater than the third preset threshold.

[0215] In some embodiments, S303 includes: when the reflectivity corresponding to multiple photoelectric units meets a second preset condition, determining that the current scene is a water surface scene; the second preset condition is used to characterize that among the multiple photoelectric units in the same time period, the reflectivity of multiple first photoelectric units located in the first area is concentrated, the reflectivity corresponding to the multiple first photoelectric units is greater than the reflectivity corresponding to the multiple second photoelectric units located in the second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the second preset condition is used to characterize a photoelectric unit area where there is a reflectivity gradient change greater than a threshold.

[0216] The second preset condition includes at least one of the following: (1) the proportion of the first photoelectric units with a reflectivity higher than the fourth preset threshold is less than the first preset proportion; (2) the first photoelectric units with a reflectivity higher than the fourth preset threshold are concentrated in an area of ​​a preset size; (3) the proportion of the second photoelectric units with a reflectivity lower than the fifth preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the fifth preset threshold is less than or equal to the fourth preset threshold; (4) there are multiple adjacent photoelectric units whose corresponding reflectivity change gradient values ​​are greater than the sixth preset threshold.

[0217] In some embodiments, S303 includes: among the signal strengths of multiple signals, if the proportion of signal strengths with values ​​in the first interval reaches a first preset range, determining that the current scene is a water surface scene; and / or among the reflectivity of multiple signals, if the proportion of reflectivity values ​​with values ​​in the second interval reaches a second preset range, determining that the current scene is a water surface scene.

[0218] In some embodiments, the target scene also includes a foggy scene. S303 then includes: counting the number of multiple signals whose signal strengths are less than a seventh preset threshold to determine the number of foggy points; and / or counting the number of multiple signals whose reflectivity is less than an eighth preset threshold to determine the number of foggy points. If the number of foggy points reaches a third preset range, the current scene is determined to be a foggy scene.

[0219] In some embodiments, the target scene also includes a foggy scene. S303 includes: if the proportions of the reflectivity values ​​of the multiple signals in different intervals meet a third preset condition, identifying the current scene as a foggy scene, where the third preset condition is used to indicate that the reflectivity values ​​corresponding to the multiple signals are concentrated in a preset interval.

[0220] Among them, the different intervals include a third interval and a fourth interval, the upper limit of the third interval is less than or equal to the lower limit of the fourth interval; the third preset condition includes: the proportion of the reflectivity of multiple signals in the third interval is greater than the proportion in the fourth interval, and the sum of the proportions of the reflectivity of multiple signals in the third interval and the fourth interval is greater than the third preset proportion.

[0221] In some embodiments, S303 includes: if the current scene is determined to be a non-surface scene based on first signal parameters of the multiple signals, identifying whether the current scene is a foggy scene based on second signal parameters of the multiple signals, wherein the first signal parameters are different from the second signal parameters. In some embodiments, the method is applied to a mobile platform, and the method further includes: if the current scene is determined to be a surface scene, controlling the mobile platform to remain hovering at a preset height above the water surface. In some embodiments, the method further includes: acquiring visual information of the current scene collected by a visual sensor. S303 includes: if the current scene is determined to be a surface scene based on the visual information and the signal parameters of the multiple signals, wherein the visual information is used to at least exclude mirror scenes other than surface scenes. In some embodiments, the method is applied to a mobile platform, and the method further includes: if the current scene is determined to be a target scene, outputting prompt information corresponding to the target scene. In some embodiments, the method is applied to a mobile platform, and the method further includes: if the current scene is determined to be foggy, controlling the mobile platform to move at an original speed, or controlling the mobile platform to move at a speed lower than the original speed for a preset period of time before returning to the original speed.

[0222] In some embodiments, the signal parameters of the multiple signals also include depth information. In some embodiments, the method further includes: if the current scene is identified as a water scene, identifying whether there are above-water objects and / or underwater objects in the water scene based at least on the distribution and / or changes in the depths corresponding to the multiple signals.

[0223] In some embodiments, the method is applied to a mobile platform and further includes: if an aquatic object is identified in the water scene, determining the landing location of the mobile platform based on the location of the aquatic object. In some embodiments, the method further includes: if the current scene is identified as a water scene, identifying whether there are waves in the water scene based on the depth distribution and / or changes of multiple signals. In some embodiments, the method is applied to a mobile platform and further includes: if waves are identified in the water scene, controlling the mobile platform to maintain a preset distance from the waves during hovering; or, if waves are identified in the water scene, controlling the mobile platform to maintain a preset distance from the water surface during hovering. In some embodiments, the current scene is the scene below the mobile platform equipped with a detection device. In some embodiments, the method further includes: outputting altitude information of the mobile platform; the altitude information of the mobile platform is determined based on the minimum value of the depths corresponding to the multiple signals. In some embodiments, the altitude information of the mobile platform is determined based on the minimum value of the depths corresponding to the multiple signals and the current attitude of the mobile platform. In some embodiments, the field of view of the detection device mounted on the mobile platform is determined based on the maximum attitude angle of the mobile platform. In some embodiments, the field of view is greater than or equal to twice the maximum attitude angle of the mobile platform. In some embodiments, the target scene also includes a non-flat scene. In some embodiments, it also includes: identifying whether the current scene contains steps or slopes based on the distribution of the depths of multiple signals and the current posture of the movable platform, and outputting prompt information based on the identification result. In some embodiments, the signal parameters include the depth corresponding to the signal; the method also includes: identifying whether the current scene contains an airport guide groove capable of landing based on the distribution of the depths of multiple signals and the current posture of the movable platform; if the current scene contains an airport guide groove capable of landing, controlling the movable platform to land in the guide groove. In some embodiments, the target scene also includes a land scene with low reflectivity. S303 includes: if the reflectivity of multiple signals meets the fourth preset condition, determining that the current scene is a land scene with low reflectivity; the fourth preset condition includes: among the reflectivities of multiple signals, the proportion of reflectivity values ​​in the fifth interval is greater than the fourth preset proportion; and / or the reflectivity gradient changes corresponding to multiple signals are all less than a threshold.

[0224] In some embodiments, referring to FIG. 10 , the present application also provides a control device 40 including:

[0225] A memory 42 for storing executable instructions; one or more processors 41; wherein, when the one or more processors 41 execute the executable instructions, they are individually or collectively configured to perform any one of the above methods.

[0226] For example, the control device may be installed in the movable platform shown in FIG1 . Of course, in some embodiments, the control device may also be electrically connected or communicatively connected to the movable platform; for example, the control device may also be a controller or mobile device that remotely controls the movable platform, and the like, which is not limited here.

[0227] The processor 41 executes the executable instructions included in the memory 42. The processor 41 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0228] The memory 42 stores executable instructions for the scene recognition method. The memory 42 may include at least one type of storage medium, including flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, a magnetic disk, an optical disk, etc. Furthermore, the device may cooperate with a network storage device that performs the storage function of the memory via a network connection. The memory 42 may be an internal storage unit of the control device 40, such as a hard disk or memory of the control device 40. The memory 42 may also be an external storage device of the control device 40, such as a plug-in hard disk equipped on the control device 40, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Furthermore, the memory 42 may include both the internal storage unit of the control device 40 and an external storage device. The memory 42 is used to store the computer program 44 and other programs and data required by the device. The memory 42 may also be used to temporarily store data that has been output or is to be output.

[0229] Various embodiments described herein can be implemented using a computer-readable medium such as computer software, hardware, or any combination thereof. For hardware implementation, the embodiments described herein can be implemented by using at least one of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, an electronic unit designed to perform the functions described herein. For software implementation, embodiments such as processes or functions can be implemented with a separate software module that allows execution of at least one function or operation. Software code can be implemented by a software application (or program) written in any appropriate programming language, and the software code can be stored in a memory and executed by a controller.

[0230] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0231] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions, such as a memory including instructions, is also provided. The instructions are executable by a processor of a device to perform the above method. For example, the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, or the like.

[0232] A non-transitory computer-readable storage medium enables the terminal to perform the above method when instructions in the storage medium are executed by a processor of the terminal.

[0233] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. The terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0234] The above is a detailed introduction to the methods and devices provided in the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the methods and core ideas of the present application. At the same time, for those skilled in the art, based on the ideas of the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A scene recognition method, characterized in that: include: Controlling the detection device to emit detection light to detect the current scene; The detection device includes an array photoelectric sensor; When the area array photoelectric sensor receives the reflected echo of the detection light, acquiring a plurality of signals output by a plurality of photoelectric units of the area array photoelectric sensor; It is identified whether the current scene is a target scene according to the signal parameters of the multiple signals, and the target scene at least includes a water surface scene.

2. A scene recognition method, characterized in that: include: Controlling the detection device to emit detection light to detect the current scene, wherein the detection device includes a planar array photoelectric sensor; When the area array photoelectric sensor receives the reflected echo of the detection light, acquiring a plurality of signals output by a plurality of photoelectric units of the area array photoelectric sensor; Whether the current scene is a target scene is identified according to the distribution and / or change of the signal parameters of the multiple signals.

3. The method according to claim 1 or 2, characterized in that: The signal parameters include signal strength and / or reflectivity corresponding to the signal.

4. The method according to claim 3, characterized in that The reflectivity is determined based on a signal strength corresponding to the signal and a depth corresponding to the signal.

5. The method according to any one of claims 1 to 4, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: Whether the current scene is a target scene is identified according to the distribution and / or change of the signal parameters of the multiple signals.

6. The method according to claim 5, characterized in that The distribution of the signal parameters includes the spatial distribution of the signal parameters and / or the distribution of the numerical values ​​of the signal parameters.

7. The method according to claim 5, characterized in that The variation of the signal parameter includes the variation of the signal parameter in space and / or the variation of the signal parameter in time.

8. The method according to any one of claims 1 to 7, characterized in that The multiple signals are capable of generating at least a portion of a point cloud image, and each pixel in the point cloud image corresponds one-to-one to a photoelectric unit of the array photoelectric sensor; The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: According to one or more frames of the point cloud image, it is identified whether the current scene is a target scene.

9. The method according to claim 8, characterized in that The point cloud image includes signal strength information and / or reflectivity information; the signal strength information and reflectivity information of each pixel in the point cloud image correspond one-to-one to the signal parameters of the signal output by the photoelectric unit in the array; The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals, wherein the target scene at least includes a water surface scene, comprises: Performing scene recognition based on signal strength information of multiple pixel points in the point cloud image, spatial changes and / or spatial distribution to identify whether the current scene is a water surface scene; and / or Scene recognition is performed based on the reflectivity information of multiple pixel points in the point cloud image, the spatial variation and / or the spatial distribution, so as to identify whether the current scene is a water surface scene.

10. The method according to any one of claims 1 to 9, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals, wherein the target scene at least includes a water surface scene, comprises: When the signal strengths corresponding to the plurality of photoelectric units meet a first preset condition, determining that the current scene is a water surface scene; The first preset condition is used to characterize that, among the multiple photoelectric units in the same time period, the signal strengths of the multiple first photoelectric units located in the first area are concentrated, the signal strengths corresponding to the multiple first photoelectric units are greater than the signal strengths corresponding to the multiple second photoelectric units located in the second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the first preset condition is used to characterize a photoelectric unit area where there is a signal strength gradient change greater than a threshold.

11. The method according to claim 10, characterized in that The first preset condition includes at least one of the following: The proportion of the first photoelectric units having a signal strength higher than the first preset threshold is less than the first preset proportion; The first photoelectric units with signal strength higher than the first preset threshold are concentrated in an area of ​​a preset size; The proportion of the second photoelectric units whose signal strength is lower than the second preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the second preset threshold is less than or equal to the first preset threshold; There are a plurality of adjacent photoelectric units whose corresponding signal strength change gradient values ​​are greater than a third preset threshold.

12. The method according to any one of claims 1 to 9, characterized in that: The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals, wherein the target scene at least includes a water surface scene, comprises: When the reflectivity corresponding to the plurality of photoelectric units meets the second preset condition, determining that the current scene is a water surface scene; The second preset condition is used to characterize that among the multiple photoelectric units in the same time period, the reflectivities of the multiple first photoelectric units located in the first area are concentratedly distributed, the reflectivities corresponding to the multiple first photoelectric units are greater than the reflectivities corresponding to the multiple second photoelectric units located in the second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the second preset condition is used to characterize a photoelectric unit area where there is a reflectivity gradient change greater than a threshold.

13. The method according to claim 12, characterized in that The second preset condition includes at least one of the following: The proportion of the first photoelectric units having a reflectivity higher than the fourth preset threshold is less than the first preset proportion; The first photoelectric units with reflectivity higher than a fourth preset threshold value are concentrated in an area of ​​a preset size; The proportion of the second photoelectric units whose reflectivity is lower than the fifth preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the fifth preset threshold is less than or equal to the fourth preset threshold; There are a plurality of adjacent photoelectric units whose corresponding reflectivity change gradient values ​​are greater than a sixth preset threshold.

14. The method according to any one of claims 1 to 6, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: If, among the signal strengths of the multiple signals, the proportion of the signal strengths whose values ​​are in the first interval reaches a first preset range, it is determined that the current scene is a water surface scene; and / or If, among the reflectivities of the multiple signals, the proportion of reflectivity values ​​whose values ​​are in the second interval reaches a second preset range, it is determined that the current scene is a water surface scene.

15. The method according to any one of claims 1 to 14, characterized in that The target scene also includes a cloud and fog scene.

16. The method according to claim 15, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: Counting the number of the plurality of signals whose signal strengths are less than a seventh preset threshold value to determine the number of cloud points; and / or counting the number of the plurality of signals whose reflectivity is less than an eighth preset threshold value to determine the number of cloud points; If the number of the cloud points reaches a third preset range, it is determined that the current scene is a cloud scene.

17. The method according to claim 15, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: When the proportion of the numerical values ​​of the reflectivity of the multiple signals in different intervals meets the third preset condition, the current scene is identified as a foggy scene, and the third preset condition is used to characterize that the numerical values ​​of the reflectivity corresponding to the multiple signals are concentrated in a preset interval.

18. The method according to claim 17, characterized in that The different intervals include a third interval and a fourth interval, and the upper limit of the third interval is less than or equal to the lower limit of the fourth interval; The third preset condition includes: the proportion of the reflectivity of the multiple signals in the third interval is greater than the proportion in the fourth interval, and the sum of the proportions of the reflectivity of the multiple signals in the third interval and the fourth interval is greater than the third preset proportion.

19. The method according to claim 15, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: In the case where the current scene is determined to be a non-water surface scene according to the first signal parameters of the multiple signals, whether the current scene is a cloud and fog scene is identified according to the second signal parameters of the multiple signals, wherein the first signal parameters are different from the second signal parameters.

20. The method according to any one of claims 1 to 14, characterized in that The method is applied to a movable platform, and the method further comprises: When the current scene is identified as a water surface scene, the movable platform is controlled to keep hovering at a preset height from the water surface.

21. The method according to any one of claims 1 to 14, characterized in that Also includes: Acquire visual information collected by a visual sensor on the current scene; The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: Whether the current scene is a water surface scene is identified according to the visual information and the signal parameters of the multiple signals; wherein the visual information is at least used to exclude other mirror scenes except the water surface scene.

22. The method according to any one of claims 1 to 19, characterized in that The method is applied to a movable platform, and the method further comprises: When the current scene is identified as the target scene, prompt information corresponding to the target scene is output.

23. The method according to any one of claims 15 to 19, characterized in that The method is applied to a movable platform, and the method further comprises: When the current scene is identified as a foggy scene, the movable platform is controlled to move at the original speed, or the movable platform is controlled to move at a speed lower than the original speed for a preset time and then move at the original speed.

24. The method according to any one of claims 1 to 19, characterized in that The signal parameters of the plurality of signals further include depth information.

25. The method according to claim 24, characterized in that Also includes: When the current scene is identified as a water surface scene, it is identified whether there are above-water objects and / or underwater objects in the water surface scene at least based on the distribution and / or change of the depths corresponding to the multiple signals.

26. The method according to claim 24, characterized in that The method is applied to a movable platform, and the method further comprises: If it is identified that there is an object on water in the water surface scene, the landing position of the movable platform is determined according to the position of the object on water.

27. The method according to claim 24, characterized in that Also includes: When the current scene is identified as a water surface scene, whether there are waves in the water surface scene is identified based on the distribution and / or change of the depths of the multiple signals.

28. The method according to claim 24, characterized in that The method is applied to a movable platform, and the method further comprises: If it is identified that there are waves in the water scene, the movable platform is controlled to maintain a preset distance from the waves during the hovering process; or, If it is identified that there are waves in the water surface scene, the movable platform is controlled to maintain a preset distance from the water surface during the hovering process.

29. The method according to claim 24, characterized in that The current scene is the scene below the movable platform carrying the detection device.

30. The method according to claim 29, characterized in that The method further comprises: Outputting the height information of the movable platform; the height information of the movable platform is determined according to the minimum value of the depths respectively corresponding to the multiple signals.

31. The method according to claim 29, characterized in that The height information of the movable platform is determined according to the minimum value of the depths respectively corresponding to the multiple signals and the current posture of the movable platform.

32. The method according to claim 1, 2 or 31, characterized in that The field of view angle of the detection device arranged on the movable platform is determined according to the maximum attitude angle of the movable platform.

33. The method according to claim 32, characterized in that The field of view angle is greater than or equal to twice the maximum attitude angle of the movable platform.

34. The method according to any one of claims 29 to 33, characterized in that The target scene also includes a non-flat ground scene.

35. The method according to claim 34, characterized in that Also includes: According to the distribution of the depths of the multiple signals and the current posture of the movable platform, it is identified whether the current scene contains stairs or slopes, and prompt information is output according to the identification result.

36. The method according to claim 34, characterized in that The signal parameters include the depth corresponding to the signal; The method further comprises: According to the depth distribution of the plurality of signals and the current posture of the movable platform, identifying whether the current scene includes an airport guide slot capable of landing; In the case that the current scene includes an airport guide slot capable of landing, the movable platform is controlled to land in the guide slot.

37. The method according to any one of claims 1 to 19, characterized in that The target scene also includes a land scene with low reflectivity.

38. The method according to claim 37, characterized in that The step of identifying whether the current scene is a target scene according to the signal parameters of the multiple signals includes: If the reflectivity of the multiple signals meets the fourth preset condition, determining that the current scene is a land scene with low reflectivity; The fourth preset condition includes: among the reflectances of multiple signals, the proportion of reflectance values ​​in the fifth interval is greater than the fourth preset proportion; and / or the reflectance gradient changes corresponding to the multiple signals are all less than a threshold.

39. A scene recognition method, characterized in that: include: Acquire an echo signal of a reflective surface in the current scene, wherein the echo signal is an optical signal reflected by the reflective surface when a detection device actively emits detection light, or an electrical signal converted from the optical signal; Determine, according to the echo signal, reflection information of the reflection surface, wherein the reflection information includes at least one of the following: distribution information and change information of signal parameters of multiple echo signals of the reflection surface; According to the reflection information of the reflection surface, it is determined whether the current scene is a target scene, and the target scene at least includes a water surface scene.

40. The method according to claim 39, characterized in that The signal parameters include a signal strength corresponding to the signal, and / or a reflectivity; wherein the reflectivity is determined based on the signal strength corresponding to the signal and the depth corresponding to the signal.

41. The method according to claim 39, characterized in that The distribution of the signal parameters includes the spatial distribution of the signal parameters and / or the distribution of the numerical values ​​of the signal parameters; the change of the signal parameters includes the spatial change of the signal parameters and / or the temporal change of the signal parameters.

42. The method according to any one of claims 1 to 41, characterized in that The multiple reflection signals of the reflection surface are signals output by multiple photoelectric units on the area array photoelectric sensor.

43. The method according to claim 42, characterized in that The multiple signals are capable of generating at least a portion of a point cloud image, and each pixel in the point cloud image corresponds one-to-one to a photoelectric unit of the array photoelectric sensor; Determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: According to one or more frames of the point cloud image, it is identified whether the current scene is a target scene.

44. The method according to claim 43, characterized in that The point cloud image includes signal strength information and / or reflectivity information; the signal strength information and reflectivity information of each pixel in the point cloud image correspond one-to-one to the signal parameters of the signal output by the photoelectric unit in the array; The determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: Performing scene recognition based on signal strength information of multiple pixel points in the point cloud image, spatial changes and / or spatial distribution to identify whether the current scene is a water surface scene; and / or According to the reflectivity information of multiple pixels in the point cloud image, the spatial variation and / or the spatial The scene recognition is performed based on the distribution of the pixels on the surface to identify whether the current scene is a water surface scene.

45. The method according to claim 42, characterized in that The determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: When the signal strengths corresponding to the plurality of photoelectric units meet a first preset condition, determining that the current scene is a water surface scene; The first preset condition is used to characterize that, among the multiple photoelectric units in the same time period, the signal strengths of the multiple first photoelectric units located in the first area are concentrated, the signal strengths corresponding to the multiple first photoelectric units are greater than the signal strengths corresponding to the multiple second photoelectric units located in the second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the first preset condition is used to characterize a photoelectric unit area where there is a signal strength gradient change greater than a threshold.

46. ​​The method according to claim 45, characterized in that The first preset condition includes at least one of the following: The proportion of the first photoelectric units having a signal strength higher than the first preset threshold is less than the first preset proportion; The first photoelectric units with signal strength higher than the first preset threshold are concentrated in an area of ​​a preset size; The proportion of the second photoelectric units whose signal strength is lower than the second preset threshold is greater than the second preset proportion; wherein the second preset proportion is greater than the first preset proportion, and the second preset threshold is less than or equal to the first preset threshold; There are a plurality of adjacent photoelectric units whose corresponding signal strength change gradient values ​​are greater than a third preset threshold.

47. The method according to claim 42, characterized in that The determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: When the reflectivity corresponding to the plurality of photoelectric units meets the second preset condition, determining that the current scene is a water surface scene; The second preset condition is used to characterize that among the multiple photoelectric units in the same time period, the reflectivities of the multiple first photoelectric units located in the first area are concentratedly distributed, the reflectivities corresponding to the multiple first photoelectric units are greater than the reflectivities corresponding to the multiple second photoelectric units located in the second area outside the first area, and the number of the multiple first photoelectric units is less than the number of the multiple second photoelectric units; and / or, the second preset condition is used to characterize a photoelectric unit area where there is a reflectivity gradient change greater than a threshold.

48. The method according to any one of claims 39 to 42, characterized in that Determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: If, among the signal strengths of the multiple signals, the proportion of the signal strengths whose values ​​are in the first interval reaches a first preset range, it is determined that the current scene is a water surface scene; and / or If, among the reflectivities of the multiple signals, the proportion of reflectivity values ​​whose values ​​are in the second interval reaches a second preset range, it is determined that the current scene is a water surface scene.

49. The method according to any one of claims 39 to 44, characterized in that The target scene also includes a cloud and fog scene.

50. The method according to claim 49, characterized in that Determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: Counting the number of the plurality of signals whose signal strengths are less than a seventh preset threshold value to determine the number of cloud points; and / or counting the number of the plurality of signals whose reflectivity is less than an eighth preset threshold value to determine the number of cloud points; If the number of the cloud points reaches a third preset range, it is determined that the current scene is a cloud scene.

51. The method according to claim 50, characterized in that Determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: When the proportion of the numerical values ​​of the reflectivity of the multiple signals in different intervals meets the third preset condition, the current scene is identified as a foggy scene, and the third preset condition is used to characterize that the numerical values ​​of the reflectivity corresponding to the multiple signals are concentrated in a preset interval.

52. The method according to claim 51, characterized in that The different intervals include a third interval and a fourth interval, and the upper limit of the third interval is less than or equal to the lower limit of the fourth interval; The third preset condition includes: the proportion of the reflectivity of the multiple signals in the third interval is greater than the proportion in the fourth interval, and the sum of the proportions of the reflectivity of the multiple signals in the third interval and the fourth interval is greater than the third preset proportion.

53. The method according to claim 49, characterized in that The determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: In the case where the current scene is determined to be a non-water surface scene according to the first signal parameters of the multiple signals, whether the current scene is a cloud and fog scene is identified according to the second signal parameters of the multiple signals, wherein the first signal parameters are different from the second signal parameters.

54. The method according to any one of claims 39 to 48, characterized in that The method is applied to a movable platform, and the method further comprises: When the current scene is identified as a water surface scene, the movable platform is controlled to keep hovering at a preset height from the water surface.

55. The method according to any one of claims 39 to 48, characterized in that Also includes: Acquire visual information collected by a visual sensor on the current scene; The determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: Whether the current scene is a water surface scene is identified according to the visual information and the signal parameters of the multiple signals; wherein the visual information is at least used to exclude other mirror scenes except the water surface scene.

56. The method according to any one of claims 39 to 48, characterized in that The method is applied to a movable platform, and the method further comprises: When the current scene is identified as the target scene, prompt information corresponding to the target scene is output.

57. The method according to any one of claims 49 to 53, characterized in that The method is applied to a movable platform, and the method further comprises: When the current scene is identified as a foggy scene, the movable platform is controlled to move at the original speed, or the movable platform is controlled to move at a speed lower than the original speed for a preset time and then move at the original speed.

58. The method according to any one of claims 39 to 48, characterized in that The signal parameters of the plurality of signals further include depth information.

59. The method according to claim 58, characterized in that Also includes: When the current scene is identified as a water surface scene, it is identified whether there are above-water objects and / or underwater objects in the water surface scene at least based on the distribution and / or change of the depths corresponding to the multiple signals.

60. The method according to claim 59, characterized in that The method is applied to a movable platform, and the method further comprises: If it is recognized that there is an object on the water surface in the water surface scene, the movable plane is determined according to the position of the object on the water surface. The landing position of the platform.

61. The method according to claim 58, characterized in that Also includes: When the current scene is identified as a water surface scene, whether there are waves in the water surface scene is identified based on the distribution and / or change of the depths of the multiple signals.

62. The method according to claim 61, characterized in that The method is applied to a movable platform, and the method further comprises: If it is identified that there are waves in the water scene, the movable platform is controlled to maintain a preset distance from the waves during the hovering process; or, If it is identified that there are waves in the water surface scene, the movable platform is controlled to maintain a preset distance from the water surface during the hovering process.

63. The method according to claim 58, characterized in that The current scene is the scene below the movable platform carrying the detection device.

64. The method according to claim 63, characterized in that The method further comprises: Outputting the height information of the movable platform; the height information of the movable platform is determined according to the minimum value of the depths respectively corresponding to the multiple signals.

65. The method according to claim 63, characterized in that The height information of the movable platform is determined according to the minimum value of the depths respectively corresponding to the multiple signals and the current posture of the movable platform.

66. The method according to claim 39 or 65, characterized in that The field of view angle of the detection device arranged on the movable platform is determined according to the maximum attitude angle of the movable platform.

67. The method according to claim 66, characterized in that The field of view angle is greater than or equal to twice the maximum attitude angle of the movable platform.

68. The method according to claim 58 or 59, characterized in that The target scene also includes a non-flat ground scene.

69. The method according to claim 68, characterized in that Also includes: According to the distribution of the depths of the multiple signals and the current posture of the movable platform, it is identified whether the current scene contains stairs or slopes, and prompt information is output according to the identification result.

70. The method of claim 68, wherein: The signal parameters include the depth corresponding to the signal; The method further comprises: According to the depth distribution of the plurality of signals and the current posture of the movable platform, identifying whether the current scene includes an airport guide slot capable of landing; In the case that the current scene includes an airport guide slot capable of landing, the movable platform is controlled to land in the guide slot.

71. The method according to any one of claims 39 to 43, characterized in that The target scene also includes a land scene with low reflectivity.

72. The method according to claim 71, characterized in that The determining whether the current scene is a target scene according to the reflection information of the reflection surface includes: If the reflectivity of the multiple signals meets the fourth preset condition, determining that the current scene is a land scene with low reflectivity; The fourth preset condition includes: among the reflectances of multiple signals, the proportion of reflectance values ​​in the fifth interval is greater than the fourth preset proportion; and / or the reflectance gradient changes corresponding to the multiple signals are all less than a threshold.

73. The method according to claim 39, characterized in that The detection device comprises a planar array photoelectric sensor, and the echo signal of the reflecting surface is received by the planar array photoelectric sensor.

74. A control device, characterized in that: include: a memory for storing executable instructions; one or more processors; Wherein, when the one or more processors execute the executable instructions, they are individually or collectively configured to execute the method of any one of claims 1 to 73.

75. A movable platform, characterized in that: include: Body; A power system, installed on the machine body, for providing power to the movable platform; a detection device, mounted on the body, for emitting detection light to detect the current scene; and A control device as claimed in claim 74.

76. The movable platform according to claim 75, characterized in that The detection device is arranged at the bottom of the machine body.

77. A computer-readable storage medium, characterized in that The computer-readable storage medium stores executable instructions, which, when executed by a processor, implement the method described in any one of claims 1 to 73.

Images