Low-altitude manned flight avoidance method, system, and manned aircraft

Abstract

The application relates to a low-altitude manned flight avoidance method, system and manned aircraft. The method comprises the following steps: acquiring multi-source data, wherein the multi-source data at least comprises digital terrain database data and image data identified by a sensor of the manned aircraft; generating avoidance reference information for the manned aircraft based on the multi-source data; acquiring pose data of the manned aircraft; generating a flight planning track of the manned aircraft according to the pose data and the avoidance reference information; and controlling flight according to the flight planning track. The scheme provided by the application can realize effective flight avoidance, improve the low-altitude flight environment perception capability of the manned aircraft and the low-altitude flight safety.

Description

Technical Field

[0001] This application relates to the field of low-altitude flight technology, and in particular to a low-altitude manned flight avoidance method, system and manned aircraft. Background Technology

[0002] With the rapid development of manned flight technology, urban air transportation is gradually becoming an important part of the future transportation system. Compared with traditional ground transportation, urban air transportation can effectively alleviate urban congestion and improve travel efficiency.

[0003] Currently, commercial airliners and helicopters typically use TCAS (Traffic Collision Avoidance System) and GPWS (Ground Proximity Warning System) to ensure safety during high-altitude flight. TCAS mainly relies on onboard transponder signals to detect surrounding aircraft and provide avoidance suggestions, while GPWS is based on onboard radar altimeters and terrain databases to monitor and warn of terrain and obstacles.

[0004] However, since TCAS and GPWS systems are mainly designed for high-altitude flight environments, they are difficult to meet the low-altitude flight requirements of urban air traffic and are not suitable for low-altitude manned flight scenarios, thus affecting low-altitude flight safety. Summary of the Invention

[0005] To address or partially address the problems existing in related technologies, this application provides a low-altitude manned flight avoidance method, system, and manned aircraft, which can be applied to avoidance handling in low-altitude manned flight scenarios and improve low-altitude flight safety.

[0006] The first aspect of this application provides a method for low-altitude manned flight obstacle avoidance, including: Acquire multi-source data, wherein the multi-source data includes at least digital terrain database data and image data identified by the sensors of the manned aircraft; Based on the multi-source data, obstacle avoidance reference information is generated for the manned aircraft. Acquire the pose data of the manned spacecraft; The flight planning trajectory of the manned aircraft is generated based on the pose data and the obstacle avoidance reference information, so as to control the flight according to the flight planning trajectory.

[0007] In one instance, the multi-source data further includes communication data from co-flying aircraft, and the step of generating obstacle avoidance reference information for the manned aircraft based on the multi-source data includes: The surrounding environment information of the manned aircraft is generated based on the digital terrain database data and the image data, and the flight trajectory of the co-flying aircraft is predicted based on the communication data of the co-flying aircraft and the image data.

[0008] In one example, generating the surrounding environment information of the manned aircraft based on the digital terrain database data and the image data includes: Extract DSM data from the digital terrain database; An environmental point cloud model is obtained by combining the DSM data and the image data. Based on the environmental point cloud model, vector information fusion is performed on the DSM data and the image data to obtain the surrounding environment information of the manned aircraft.

[0009] In one example, predicting the flight trajectory of the co-flying aircraft based on the communication data and the image data includes: The image data and the environmental point cloud model are used to detect the 3D pose information of dynamic obstacles around the manned aircraft, wherein the 3D pose information of the dynamic obstacles includes at least the 3D pose information of the aircraft in the same field. Based on the 3D pose information and communication data of the co-flying aircraft, a predicted flight trajectory of the co-flying aircraft is generated.

[0010] In one instance, the dynamic obstacle 3D pose information further includes the 3D pose information of flying obstacles, and the method further includes: The predicted flight trajectory of the flying obstacle is generated based on the 3D pose information of the flying obstacle.

[0011] In one instance, generating the flight planning trajectory of the manned aircraft based on the pose data and the obstacle avoidance reference information includes: Obstacle avoidance risk detection is performed based on the pose data and the obstacle avoidance reference information; Based on the obstacle avoidance risk detection results, the flight planning trajectory of the manned aircraft is generated.

[0012] A second aspect of this application provides a low-altitude manned flight obstacle avoidance system, the system including a perception and avoidance processor; wherein, the perception and avoidance processor includes: A perception module is used to acquire multi-source data, wherein the multi-source data includes at least digital terrain database data and image data identified by the sensors of the manned aircraft; and to generate obstacle avoidance reference information for the manned aircraft based on the multi-source data. The obstacle avoidance planning module is used to acquire the position and attitude data of the manned aircraft; generate the flight planning trajectory of the manned aircraft based on the position and attitude data and the obstacle avoidance reference information, and control the flight according to the flight planning trajectory.

[0013] In one example, the system further includes a multi-source data acquisition module, which comprises: The image acquisition module is used to acquire image data identified by the sensors of the manned aircraft through sensors and transmit it to the perception module; The digital terrain database module is used to collect digital terrain database data and transmit it to the sensing module; The communication module is used to collect communication data from the co-flying aircraft and transmit it to the sensing module; The perception module generates the surrounding environment information of the manned aircraft based on the digital terrain database data and the image data, and generates the predicted flight trajectory of the co-flying aircraft based on the communication data of the co-flying aircraft and the image data.

[0014] In one instance, the sensing module includes: The data acquisition submodule is used to acquire multi-source data, which includes digital terrain database data, image data identified by the sensors of the manned aircraft, and communication data of the co-flying aircraft. The data fusion submodule is used to extract DSM data from the digital terrain database; obtain an environmental point cloud model by combining the DSM data and the image data; and perform vector information fusion on the DSM data and the image data based on the environmental point cloud model to obtain the surrounding environment information of the manned aircraft. The tracking and prediction submodule is used to detect the 3D pose information of dynamic obstacles around the manned aircraft using the image data and the environmental point cloud model, wherein the 3D pose information of the dynamic obstacles includes at least the 3D pose information of the co-flying aircraft; based on the 3D pose information of the co-flying aircraft and the communication data of the co-flying aircraft, it generates a predicted flight trajectory of the co-flying aircraft; wherein the 3D pose information of the dynamic obstacles also includes the 3D pose information of flying obstacles, and the tracking and prediction submodule is also used to generate a predicted flight trajectory of the flying obstacles based on the 3D pose information of the flying obstacles.

[0015] In one instance, the obstacle avoidance planning module comprises: The collision avoidance risk assessment and detection submodule is used to perform obstacle avoidance risk detection based on the pose data and the obstacle avoidance reference information. The obstacle avoidance path planning submodule is used to generate the flight planning trajectory of the manned aircraft based on the detection results of obstacle avoidance risk detection.

[0016] A third aspect of this application provides a low-altitude manned aircraft, comprising: Processor; and A memory that stores executable code, which, when executed by the processor, causes the processor to perform the method described above.

[0017] The technical solution provided in this application may include the following beneficial effects: In this embodiment, multi-source data is acquired, including at least digital terrain database data and image data identified by the sensors of the manned aircraft. Obstacle avoidance reference information for the manned aircraft is generated based on the multi-source data. The pose data of the manned aircraft is acquired. The flight planning trajectory of the manned aircraft is generated based on the pose data and obstacle avoidance reference information, so as to control the flight according to the flight planning trajectory.

[0018] This application acquires multi-source data and utilizes digital terrain database data and sensor-identified image data from the multi-source data to generate obstacle avoidance reference information for manned aircraft. This provides reliable decision support for low-altitude flight avoidance of manned aircraft, ensuring that the flight planning trajectory for manned aircraft is accurately generated by combining attitude data and obstacle avoidance reference information. This enables effective flight avoidance, improves the low-altitude flight environment perception capability and low-altitude flight safety of manned aircraft.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0020] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0021] Figure 1 This is a flowchart illustrating a low-altitude manned flight obstacle avoidance method according to an embodiment of this application; Figure 2 This is another schematic flowchart illustrating a low-altitude manned flight obstacle avoidance method according to an embodiment of this application; Figure 3 This is a data flow diagram illustrating the perception and avoidance process in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a low-altitude manned flight avoidance system shown in the embodiments of this application; Figure 5This is a block diagram of the perception and avoidance system shown in the embodiments of this application; Figure 6 This is a schematic diagram of the structure of a low-altitude manned aircraft shown in an embodiment of this application. Detailed Implementation

[0022] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0023] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0024] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0025] Currently, commercial airliners and helicopters typically use TCAS (Traffic Collision Avoidance System) and GPWS (Ground Proximity Warning System) to ensure safety during high-altitude flight. TCAS mainly relies on onboard transponder signals to detect surrounding aircraft and provide avoidance suggestions, while GPWS is based on onboard radar altimeters and terrain databases to monitor and warn of terrain and obstacles.

[0026] However, since TCAS and GPWS systems are mainly designed for high-altitude flight environments, they are difficult to meet the low-altitude flight requirements of urban air traffic and are not suitable for low-altitude manned flight scenarios, thus affecting low-altitude flight safety.

[0027] To address the aforementioned issues, this application provides a method, system, and manned aircraft for obstacle avoidance in low-altitude manned flight scenarios, which can be applied to obstacle avoidance handling in low-altitude manned flight scenarios and improve low-altitude flight safety.

[0028] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0029] Figure 1 This is a schematic flowchart illustrating a low-altitude manned flight obstacle avoidance method according to an embodiment of this application. See also... Figure 1 This method, applied to low-altitude manned flight avoidance systems, includes at least the following steps: Step 101: Acquire multi-source data, which includes at least digital terrain database data and image data identified by the sensors of the manned aircraft.

[0030] Current systems such as TCAS and GPWS are mainly designed for mid-to-high altitude flight. Compared with the high-altitude flight environment, the low-altitude flight environment is more complex. It not only has static obstacles such as tall buildings and bridges, but also dynamic obstacles such as other aircraft and flocks of birds. These obstacles can easily interfere with the flight of manned aircraft and affect the safety of low-altitude manned flight.

[0031] In this application embodiment, in order to fully perceive the complex low-altitude flight environment, the low-altitude manned flight avoidance system of this application obtains relevant data from multiple different data sources and uses it as multi-source data for low-altitude manned flight avoidance, especially obtaining digital terrain database data and image data identified by the sensor of the manned aircraft.

[0032] Step 102: Generate obstacle avoidance reference information for manned aircraft based on multi-source data.

[0033] In this embodiment of the application, after the low-altitude manned flight avoidance system acquires multi-source data, it can generate obstacle avoidance reference information for the manned aircraft based on the multi-source data. This obstacle avoidance reference information provides data support for the low-altitude manned flight avoidance system to make avoidance decisions, so as to assess the collision risk between the manned aircraft and various obstacles and generate a flight planning trajectory.

[0034] Step 103: Obtain the attitude data of the manned spacecraft.

[0035] In actual flight, the attitude data of manned aircraft is critical data, which directly affects the flight path, speed adjustment and obstacle avoidance operation of manned aircraft in low-altitude flight environment. Therefore, before the low-altitude manned flight avoidance system plans the obstacle avoidance path, it is also necessary to obtain the attitude data of manned aircraft to ensure that a flight planning trajectory that conforms to the current flight state of manned aircraft is generated.

[0036] Optionally, pose data refers to position and attitude data that directly reflect the current flight state of a manned aircraft.

[0037] The location data can be the geographic coordinates of the manned aircraft, such as its longitude, latitude, and altitude. The attitude data can be the manned aircraft's yaw, pitch, and roll angles.

[0038] Step 104: Generate the flight planning trajectory of the manned aircraft based on the pose data and obstacle avoidance reference information, so as to control the flight according to the flight planning trajectory.

[0039] In this embodiment of the application, the low-altitude manned flight avoidance system generates a flight planning trajectory that conforms to the low-altitude flight environment based on the obstacle avoidance reference information obtained in step 102 and the pose data obtained in step 103, so as to control the manned aircraft to accurately avoid obstacles when flying at low altitude.

[0040] In this embodiment, multi-source data is acquired, including at least digital terrain database data and image data identified by the sensors of the manned aircraft. Obstacle avoidance reference information for the manned aircraft is generated based on the multi-source data. The pose data of the manned aircraft is acquired. The flight planning trajectory of the manned aircraft is generated based on the pose data and obstacle avoidance reference information, so as to control the flight according to the flight planning trajectory.

[0041] This application acquires multi-source data and utilizes digital terrain database data and sensor-identified image data from the multi-source data to generate obstacle avoidance reference information for manned aircraft. This provides reliable decision support for low-altitude flight avoidance of manned aircraft, ensuring that the flight planning trajectory for manned aircraft is accurately generated by combining attitude data and obstacle avoidance reference information. This enables effective flight avoidance, improves the low-altitude flight environment perception capability and low-altitude flight safety of manned aircraft.

[0042] Figure 2 This is a flowchart illustrating a low-altitude manned flight obstacle avoidance method according to another embodiment of this application. Figure 2 relatively Figure 1 The technical solution of the embodiments of this application is described in more detail. It is applied to a low-altitude manned flight avoidance system. The method may include the following steps: Step 201: Obtain multi-source data.

[0043] In this embodiment of the application, after the low-altitude manned flight avoidance system is initialized, the low-altitude manned flight avoidance system first obtains relevant data from different data sources. This data is the data basis for the low-altitude manned flight avoidance system to perform environmental perception and detection.

[0044] The multi-source data acquired by the low-altitude manned flight avoidance system includes at least digital terrain database data, image data identified by the sensors of the manned aircraft, and communication data of other aircraft in the same field.

[0045] To enable those skilled in the art to better understand the multi-source data involved in low-altitude manned flight avoidance, each type of data will be explained in detail below.

[0046] Digital terrain database data: Geographic information stored in the digital terrain database mainly includes the size, height, location, and type of objects such as high-rise buildings and transportation facilities, as well as relevant terrain features.

[0047] Among them, the digital terrain database refers to a database constructed using a large amount of geographic information and prior maps measured and collected through various means. This database contains detailed geographic information of static elements such as high-rise buildings, transportation facilities, and natural terrain.

[0048] It is understood that static elements in digital terrain databases are objectively existing geographical environmental features that will not change significantly in a short period of time. Therefore, this application marks static elements in digital terrain databases that affect the flight safety of manned aircraft as static obstacles, and considers static obstacles as one of the influencing factors during flight planning.

[0049] Image data identified by the sensors of manned aircraft: Environmental images collected by the sensors of manned aircraft. The sensors of manned aircraft can not only collect images of static elements, which may overlap with static elements in the digital terrain database, or may be other static elements that the digital terrain database has not yet updated, but also capture images of dynamic elements, such as other aircraft, birds, etc.

[0050] It is understandable that the trajectories of these dynamic elements also affect the flight safety of manned aircraft. Therefore, this application marks the dynamic elements identified by the sensors of manned aircraft as dynamic obstacles, and considers dynamic obstacles as one of the influencing factors during flight planning.

[0051] Among them, the sensor can be an image acquisition device or a vision sensor, which can realize environmental perception and target detection.

[0052] Communication data from other aircraft in the same airspace: Communication data from other aircraft in the same airspace as the pilot (i.e., the manned aircraft). Generally, this communication data includes the pilot's GPS (Global Positioning System) longitude, latitude, altitude, attitude, flight speed, acceleration, flight direction, and other flight data.

[0053] In this application, the manned aircraft and other aircraft in the same field can establish a data connection via long-distance wireless communication and periodically broadcast their own position information using a common data format and transmission rules. Furthermore, the manned aircraft and other aircraft in the same field can also receive position information broadcast by other aircraft through a listening communication channel, thereby achieving position information sharing.

[0054] It is worth noting that the aircraft operating in the same field are cooperative aircraft. Cooperative aircraft refer to aircraft that actively share their own flight status information, have communication equipment, and follow a unified communication protocol to provide their own position information to other aircraft or control systems through broadcast communication.

[0055] In actual low-altitude flight environments, in the same airspace, in addition to cooperative aircraft, there may also be non-cooperative aircraft and birds that cannot communicate with manned aircraft. This application classifies both non-cooperative aircraft and birds that cannot communicate with manned aircraft as flight obstacles.

[0056] Non-cooperative aircraft refer to aircraft that do not actively share their own location information. They may lack communication equipment or have not enabled broadcast functions, making it impossible for other aircraft to detect their presence through conventional communication methods.

[0057] As an example, within the same airspace, manned aircraft A broadcasts its own position information via long-range wireless communication while simultaneously monitoring the communication channel to receive broadcast information from cooperative aircraft B. Similarly, cooperative aircraft B can also learn the position information of manned aircraft A by monitoring the communication channel.

[0058] It should be noted that the above is only an example using two aircraft. In actual applications, there are often two or more aircraft in the same airspace at the same time. They also use the above communication methods to share location information, so that each aircraft can receive communication data from other aircraft in the vicinity. This can effectively avoid mutual interference and reduce the risk of conflict.

[0059] Step 202: Generate obstacle avoidance reference information for manned aircraft based on multi-source data.

[0060] In this embodiment of the application, the obstacle avoidance reference information includes the surrounding environment information of the manned aircraft and the flight trajectory of other aircraft in the same field.

[0061] The surrounding environment information is obtained by combining digital terrain database data and image data: DSM data is extracted from the digital terrain database data, and environmental point cloud model is obtained by combining DSM data and image data. Based on the environmental point cloud model, vector information fusion is performed on DSM data and image data to obtain the surrounding environment information of the manned aircraft.

[0062] In this application, DSM (Digital Surface Model) data refers to the height information of the ground surface and all objects above it, generated through remote sensing technology. For example, the height information of tall buildings and trees.

[0063] An environmental point cloud model refers to a three-dimensional data model used to represent the structure of the surrounding environment. It consists of a large number of "points", each of which has a clear position coordinate in three-dimensional space. Each point corresponds to information such as the shape, distance, and surface features of objects in the low-altitude flight environment. It can help the system construct an effective environmental structure and locate dynamic obstacles.

[0064] Surrounding environment information refers to the various environmental features and obstacle information around a manned aircraft after data fusion processing. It can be directly used as input data for obstacle avoidance planning to help the system make correct decisions.

[0065] As an example, after extracting DSM data from a digital terrain database, the DSM data needs to undergo preprocessing such as denoising and filtering to obtain preprocessed DSM data, which is the height information of static obstacles. Then, the height information of the DSM data is mapped one-to-one with the position of each pixel in the image data to obtain point cloud data containing spatial coordinates, height, texture, color, and other information. The point cloud data is then rasterized to obtain an environmental point cloud model. Finally, based on the environmental point cloud model, the DSM data and image data are converted to vector format and fused in the same coordinate system to obtain the surrounding environment information of the manned aircraft.

[0066] As another example, in the process of vector information fusion of DSM data and image data based on environmental point cloud models, dynamic obstacle data can be fused with DSM data and image data to obtain more complete and comprehensive information about the surrounding environment. Here, dynamic obstacle data includes communication data from co-flying aircraft and flight data from flying obstacles.

[0067] Furthermore, as described in step 201 regarding co-flying vehicles and obstacles, during the generation of surrounding environment information, if there are co-flying vehicles (cooperative vehicles) around the manned aircraft, the communication data of the co-flying vehicles can be directly received. The communication data of the co-flying vehicles can then be fused with DSM data and image data using vector information to obtain more complete and comprehensive surrounding environment information.

[0068] If there are flying obstacles (non-cooperative aircraft, birds) around the manned aircraft, the image data of the flying obstacles can be identified by sensors (such as lidar, high-definition cameras), the flight data of the flying obstacles can be analyzed from the image data, and the flight data of the aircraft in the same field can be fused with DSM data and image data to obtain more complete and comprehensive information about the surrounding environment.

[0069] The surrounding environment information obtained at this time not only integrates the characteristics of static obstacles such as tall buildings and terrain, but also integrates the communication data of the aircraft in the same field and the flight data such as the position and speed of flying obstacles, realizing comprehensive environmental perception and improving the accuracy of subsequent collision detection and obstacle avoidance decisions.

[0070] The flight trajectory is predicted based on the communication data and image data of the co-flying aircraft. The prediction process may include: detecting the 3D pose information of dynamic obstacles around the manned aircraft through image data and environmental point cloud models, wherein the 3D pose information of the dynamic obstacles includes at least the 3D pose information of the co-flying aircraft, and generating the predicted flight trajectory of the co-flying aircraft based on the 3D pose information and communication data of the co-flying aircraft.

[0071] In addition, the dynamic obstacle 3D pose information also includes the 3D pose information of flying obstacles. The flight trajectory is predicted based on the communication data and image data of the aircraft in the same field. The prediction process also includes generating the predicted flight trajectory of the flying obstacle based on the 3D pose information of the flying obstacle.

[0072] Furthermore, image data provides two-dimensional image information of obstacles, while environmental point cloud models provide 3D pose information of obstacles. The low-altitude manned flight avoidance system identifies and locates dynamic obstacles through image data and environmental point cloud models, and extracts the pose information of dynamic obstacles in the low-altitude flight environment, such as position, speed, and direction.

[0073] In addition to the 3D pose information of the aircraft in the same field, dynamic obstacles may also include the 3D pose information of the flying obstacles.

[0074] After acquiring the 3D pose information of the co-flying aircraft, historical motion data is extracted from the communication data of the co-flying aircraft. The 3D pose information and historical motion data of the co-flying aircraft are used as input data, and the future motion trajectory of the co-flying aircraft, i.e., the flight trajectory, is predicted by machine learning or kinematic models (such as constant velocity model and constant acceleration model).

[0075] Similarly, after obtaining the 3D pose information of the flying obstacle, the 3D pose information of the flying obstacle is directly used as input data, and the future motion trajectory of the flying obstacle, i.e., the flight trajectory, is obtained by using machine learning or kinematic models (such as constant velocity model, constant acceleration model).

[0076] As an example, a low-altitude manned flight obstacle avoidance system can use a combination of 2D static obstacle detection and 3D dynamic obstacle detection to detect dynamic obstacles around the manned aircraft, or it can use a variety of purely visual dynamic obstacle detection and tracking algorithms to detect dynamic obstacles around the manned aircraft.

[0077] Among them, various purely visual dynamic obstacle detection and tracking algorithms include: Simultaneous Localization and Mapping (SLAM) technology based on binocular cameras is used to achieve high-precision environmental perception and obstacle tracking.

[0078] The YOLO (You Only Look Once) series of target detection algorithms for obstacle detection are used to achieve efficient real-time perception.

[0079] The Single Shot MultiBox Detector (SSD) algorithm is used to achieve rapid identification and detection of obstacles.

[0080] Faster Region-based Convolutional Neural Network (Faster R-CNN) is used for high-precision obstacle detection and tracking.

[0081] As another example, the way a low-altitude manned flight avoidance system extracts the 3D pose information of dynamic obstacles in a low-altitude flight environment can be: by using multi-target tracking algorithms, such as Kalman filtering and particle filtering, to track the detected dynamic obstacles and update their position and motion state in real time.

[0082] Step 203: Obtain the attitude data of the manned spacecraft.

[0083] In this embodiment of the application, after steps 201 and 202, the low-altitude manned flight avoidance system has obtained the surrounding environment information of the manned aircraft and the flight trajectory of the aircraft in the same field.

[0084] In decision-making and planning, in addition to considering the surrounding environment information of the manned aircraft and the flight trajectory of other aircraft in the same field, the low-altitude manned flight avoidance system also needs to comprehensively consider the relative position and motion status of the manned aircraft and other dynamic obstacles to ensure that the complex low-altitude flight environment is fully taken into account during the decision-making and planning process.

[0085] To this end, this application further obtained the pose data of the manned aircraft and comprehensively analyzed it with the surrounding environment information, the flight trajectory of other aircraft in the same field and the motion state of dynamic obstacles. Combined with the collision risk detection algorithm, the optimal flight planning trajectory was generated.

[0086] As an example, the low-altitude manned flight avoidance system acquires the position and attitude data of the manned aircraft in real time, and uses the position and attitude data to calculate the relative distance between the manned aircraft and dynamic obstacles, as well as the direction of motion of the manned aircraft itself.

[0087] Step 204: Perform obstacle avoidance risk detection based on pose data and obstacle avoidance reference information.

[0088] In this embodiment of the application, the process of obstacle avoidance risk detection based on pose data and obstacle avoidance reference information can be as follows: the low-altitude manned flight avoidance system calculates the relative distance and direction of motion between the manned aircraft and the dynamic obstacle based on the pose data of the manned aircraft, the surrounding environment information, and the flight trajectory of dynamic obstacles (including co-flying aircraft, non-cooperative aircraft, and dynamic obstacles such as birds).

[0089] If the trajectory of a dynamic obstacle intersects with the trajectory of a manned aircraft, it indicates a collision risk. It is necessary to calculate the relative distance and collision time window between the manned aircraft and the dynamic obstacle, and initiate obstacle avoidance trajectory planning or braking trajectory planning to improve flight safety.

[0090] If the trajectory of a dynamic obstacle does not intersect with the trajectory of a manned aircraft, it indicates that there is no risk of collision, and the manned aircraft can continue to fly along its normal flight path, thereby avoiding unnecessary avoidance maneuvers.

[0091] As an example, methods for calculating the relative distance between a manned aircraft and dynamic obstacles include, but are not limited to, stereo vision ranging technology and combining multiple algorithms to achieve accurate ranging and tracking.

[0092] Examples include Semiglobal Block Matching (SGBM), YOLOv8 combined with SGBM stereo matching, binocular ranging based on feature point extraction and matching, and binocular ranging based on reprojection matrices.

[0093] In the process of collision risk detection based on pose data and obstacle avoidance reference information, the low-altitude manned flight avoidance system can also consider environmental constraints, such as whether the current airspace is no-fly zone or whether there is severe weather, to improve the reliability of collision risk detection.

[0094] Step 205: Based on the detection results of obstacle avoidance risk detection, generate the flight planning trajectory of the manned aircraft.

[0095] In the embodiments of this application, the low-altitude manned flight avoidance system generates the flight planning trajectory of the manned aircraft based on the detection results of obstacle avoidance risk detection, including at least three decision planning modes: normal flight mode, obstacle avoidance flight mode, and emergency braking mode, forming a multi-level obstacle avoidance strategy.

[0096] Normal Flight Mode: The obstacle avoidance risk detection result indicates a low collision risk, meaning the low-altitude flight environment is safe, and the manned aircraft can fly normally along the predetermined flight path. For example, there are no static obstacles such as tall buildings or power towers obstructing the manned aircraft's path in the surrounding environment, and the aircraft's flight altitude and position can avoid these static obstacles. Furthermore, there are no dynamic obstacles such as co-flying aircraft, non-cooperative aircraft, or birds in the surrounding environment on the aircraft's flight path, and the flight trajectories of these dynamic obstacles have been predicted and are guaranteed to pose no collision risk.

[0097] Obstacle Avoidance Flight Mode: If the obstacle risk detection result indicates a potential collision risk, the low-altitude manned flight avoidance system will execute an obstacle avoidance flight mission, obtaining an obstacle avoidance trajectory through obstacle avoidance trajectory planning. For example, if static obstacles such as buildings or power towers obstruct the flight path of the manned aircraft, or dynamic obstacles such as other aircraft, non-cooperative aircraft, or birds are in the flight path, these obstacles can be avoided by adjusting the manned aircraft's heading or altitude.

[0098] Emergency Braking Mode: When the obstacle avoidance risk detection indicates a high collision risk, the low-altitude manned flight avoidance system performs an emergency braking task. It obtains a braking trajectory through braking trajectory planning, and the manned aircraft flies according to this trajectory, enabling it to achieve emergency braking before colliding with the obstacle. For example, the system detects that the manned aircraft cannot avoid static or dynamic obstacles by adjusting its course or altitude.

[0099] Correspondingly, different decision-making and planning modes can produce different flight planning trajectories. In normal flight mode, the flight planning trajectory is the original planned flight trajectory. In obstacle avoidance flight mode, the flight planning trajectory is the obstacle avoidance trajectory. In emergency braking mode, the flight planning trajectory is the braking trajectory.

[0100] As an example, if a low-altitude manned flight obstacle avoidance system identifies a potential collision risk through collision risk detection, it initiates obstacle avoidance trajectory planning. The system calculates at least one obstacle avoidance trajectory that can safely bypass the obstacle using a path planning algorithm. If the calculated obstacle avoidance trajectory is infeasible, such as when the obstacle is too close, it indicates that the collision risk detection result is not merely a potential collision risk. The potential collision risk is then upgraded to a high-risk collision risk level. In this case, braking trajectory planning is initiated to plan a braking trajectory, which controls the manned aircraft to stop flying before colliding with the obstacle.

[0101] In practical use, low-altitude manned flight obstacle avoidance systems can flexibly select appropriate path planning algorithms according to flight requirements, thereby adapting to different flight environments and mission requirements, and optimizing obstacle avoidance performance and flight efficiency.

[0102] In this application, the path planning algorithms include not only the EGO (Elastic Guidance and Obstacle Avoidance) algorithm, but also artificial potential field method, timed elastic band, grid method, reinforcement learning algorithm, and Bézier curve-based obstacle avoidance path planning algorithm.

[0103] Taking the commonly used A* algorithm and RRT algorithm (Rapidly-exploring Random Tree) as examples: In scenarios where searching for the optimal path occurs in low-altitude flight environments, the A* algorithm can be chosen. The A* algorithm is mainly based on priority search and uses a heuristic evaluation function to determine the search direction. It has high computational efficiency and path quality in static environments or under known map conditions, and can quickly find the optimal path.

[0104] In high-dimensional spaces and dynamically complex, irregular environments, the RRT algorithm can be chosen. Unlike the A* algorithm, the RRT algorithm explores the search space using random sampling, gradually expanding a tree until a portion of the tree reaches the target point or meets certain termination conditions. Therefore, the RRT algorithm is suitable for more complex environments and can accurately generate feasible paths.

[0105] As mentioned earlier, obstacle avoidance trajectory planning refers to the process by which a low-altitude manned flight avoidance system calculates and generates at least one obstacle avoidance trajectory that can safely bypass an obstacle when an obstacle is detected in the flight path of a manned aircraft, based on the manned aircraft's pose data, surrounding environmental information, and the flight trajectory of the dynamic obstacle, using a path planning algorithm.

[0106] As an example, in obstacle avoidance trajectory planning, the low-altitude manned flight avoidance system can take into account factors such as the aircraft's flight status, weather conditions, and airspace management to plan a flight trajectory that can avoid both static and dynamic obstacles while meeting environmental constraints, thus comprehensively ensuring the stability and safety of manned aircraft flight.

[0107] Among them, path planning algorithms are divided into global path planning algorithms and local path planning algorithms.

[0108] Global path planning algorithms can take into account environmental constraints such as terrain height and no-fly zones, and meet flight mission requirements such as shortest path and lowest energy consumption. If a low-altitude manned flight obstacle avoidance system needs to plan an optimal path from the starting point to the target point based on the flight mission objective, a global path planning algorithm, such as the A* algorithm or Dijkstra's algorithm, can be used to generate a global obstacle avoidance path.

[0109] Local path planning algorithms can take into account the motion trajectory of dynamic obstacles, ensuring flight safety. If a low-altitude manned flight avoidance system needs to adjust the global path in real time based on the flight data of dynamic obstacles to generate short-term effective local obstacle avoidance paths, local path planning algorithms can be used to generate local obstacle avoidance paths, such as the DWA algorithm (Dynamic Window Approach) and the RRT algorithm (Rapidly-exploring Random Tree).

[0110] After obtaining the global obstacle avoidance path and the local obstacle avoidance path, this application can combine the two and then generate a smooth preliminary flight trajectory based on the dynamic constraints of the manned aircraft (such as maximum speed and maximum acceleration). The preliminary flight trajectory is then optimized using trajectory optimization algorithms such as spline interpolation and Bézier curves to obtain the final executable flight planning trajectory of the manned aircraft.

[0111] As mentioned earlier, braking trajectory planning refers to the need for the low-altitude manned flight avoidance system to activate emergency braking trajectory planning when an obstacle is too close to the manned aircraft, or when the aircraft's flight state makes effective detour impossible. In this situation, the low-altitude manned flight avoidance system generates a braking trajectory and controls the aircraft's speed and direction in real time to ensure the aircraft can stop safely before a collision, thus avoiding an accident.

[0112] This application combines collision risk detection with different decision-making and planning modes to dynamically identify potential collisions between manned aircraft and obstacles and make timely obstacle avoidance decisions, which significantly improves the flight safety of manned aircraft.

[0113] As an optional example of this application, the manned aircraft also includes a flight control system and a display and control system that are communicatively connected to a low-altitude manned flight avoidance system.

[0114] The flight control system is the core control unit of a manned aircraft, responsible for executing flight control commands and adjusting the flight status through real-time feedback. Its main tasks include attitude control, heading adjustment, and obstacle avoidance.

[0115] The display and control system provides the human-machine interface for manned aircraft, mainly used to display flight status and obstacle avoidance information, allowing operators to input control commands or display control commands automatically generated by the system in the human-machine interface.

[0116] In this application, the display screen of the display system is not limited to the existing screens on manned aircraft. In fact, multiple screens can be displayed simultaneously by calling and switching, thus enriching the ways of information display.

[0117] Furthermore, the display system supports discrete displays, for example, displaying and interacting with offline electronic devices through independently developed applications.

[0118] As an example, when a low-altitude manned flight avoidance system detects a collision risk between the manned aircraft and other aircraft, it generates control commands corresponding to the planned flight trajectory. These commands include position adjustments, speed control, and steering maneuvers to ensure the aircraft can fly smoothly and safely along the planned path. The control commands are then transmitted to the flight control system, which executes them, while a feedback system refines the control effect. Simultaneously, the display and control system can show the manned aircraft's flight status information and obstacle avoidance reference information.

[0119] If the flight control system malfunctions due to communication interference during flight avoidance, the low-altitude manned flight avoidance system can automatically trigger a warning and display alarm messages such as "Flight control system malfunction, please check connection" on the display and control system, allowing users to understand the system's piloting intent and improving the user experience.

[0120] It is worth noting that the above-mentioned implementation process of the low-altitude manned flight obstacle avoidance system is not only applicable to manned aircraft and unmanned aircraft, but can also be extended to various transportation and autonomous navigation devices such as automobiles, ships, and robots by appropriately adjusting the communication method. This provides intelligent path planning and obstacle avoidance support for various transportation and autonomous navigation devices, enhancing the autonomous driving capabilities in different scenarios.

[0121] To enable those skilled in the art to better understand the technical solutions of the embodiments of this application, an example is provided below, along with reference to... Figure 3 The illustrated perception and avoidance data flow diagram illustrates the embodiments of this application.

[0122] 1) Data Collection The manned aircraft is eVTOL (Electric Vertical Take-Off and Landing).

[0123] After system initialization, data acquisition is performed first. Data acquisition is the foundation of the low-altitude manned flight avoidance system, which mainly involves the collaborative work of multiple sensors and devices to obtain multi-source data.

[0124] Communication equipment: Receives broadcast target flight data (such as cooperative aircraft), which includes position, speed, heading, etc.

[0125] Image acquisition equipment: acquires image data through cameras or visual sensors to achieve environmental perception, target detection, and obstacle recognition.

[0126] Prior maps: Obtain terrain information from digital terrain databases (such as DSM data) to help build high-precision environmental point cloud models.

[0127] Inertial navigation system: Acquires pose data of eVTOL in real time, including position, velocity, attitude angle, etc., for trajectory prediction and obstacle avoidance decision-making.

[0128] 2) Time synchronization The timestamps of data collected by different sensors may have errors, so time synchronization is required to ensure the accuracy of data fusion.

[0129] The time synchronization methods used in this application include: using precise time information provided by the Global Positioning System to unify the time of each sensor, and synchronizing the clocks of different devices through a network time protocol to ensure data time alignment.

[0130] 3) Data fusion After data collection and time synchronization, the data is fused to obtain complete information about the surrounding environment.

[0131] In this application, data fusion includes vector information fusion and obstacle fusion.

[0132] Vector information fusion requires first performing point cloud raster mapping, that is, generating a 3D environmental point cloud model through image data and prior maps, using the 3D environmental point cloud model to describe the spatial structure of the environment around the eVTOL, and then fusing target flight data, dynamic obstacle information, terrain data and other data into unified vector information.

[0133] Obstacle fusion first uses image data and point cloud raster mapping to detect dynamic obstacles in the environment, such as other aircraft and birds, and extracts the 3D pose information of the dynamic obstacles. Then, it fuses the 3D pose information of the dynamic obstacles with the target flight data to obtain more accurate 3D pose information of the dynamic obstacles.

[0134] It should be noted that vector information fusion focuses on integrating multi-source data to provide global environmental perception, while obstacle fusion focuses on tracking and predicting dynamic obstacles to ensure the accuracy of obstacle avoidance decisions.

[0135] 4) Obstacle tracking Multi-target tracking algorithms (such as Kalman filtering and particle filtering) are used to track detected dynamic obstacles, thereby updating the position and motion status of the obstacles in real time.

[0136] 5) Trajectory Prediction Machine learning or kinematic models (such as constant velocity models and constant acceleration models) are used to predict the trajectory of obstacles.

[0137] 6) Dynamic obstacle avoidance Dynamic obstacle avoidance is the ultimate goal of low-altitude manned flight avoidance systems, achieved through collision risk detection, trajectory planning, and command control.

[0138] Among them, collision risk detection can combine the aircraft's own attitude data and the predicted trajectory of obstacles to calculate the relative distance and direction of motion, and then determine whether there is a collision risk. If there is a collision risk, it will enter the obstacle avoidance decision process; if there is no collision risk, it will continue to fly normally along the predetermined path.

[0139] Trajectory planning includes obstacle avoidance trajectory planning and braking trajectory planning.

[0140] Obstacle avoidance trajectory planning is a path planning algorithm (such as A* algorithm, RRT algorithm) that plans an obstacle avoidance trajectory based on the position of the obstacle and the predicted trajectory.

[0141] When the braking trajectory is planned, an emergency braking trajectory is planned to stop the eVTOL before a collision when it is detected that bypassing an obstacle is not feasible (such as when the obstacle is too close), thus avoiding potential risks.

[0142] 7) Obstacle Avoidance Command Control The planned obstacle avoidance trajectory or braking trajectory is converted into obstacle avoidance target commands, including speed, direction, attitude angle, etc., and the flight control system executes the obstacle avoidance target commands to control the eVTOL to fly or stop according to the trajectory.

[0143] 8) Pose update and feedback After executing the obstacle avoidance command, the eVTOL's pose information is updated and fed back to the low-altitude manned flight avoidance system for the next round of environmental perception and obstacle avoidance decision-making, ensuring that the aircraft always maintains the optimal flight path.

[0144] In this embodiment, by combining multi-sensor fusion technology, dynamic obstacle tracking and prediction, and path planning algorithms, eVTOL is equipped with intelligent and highly autonomous obstacle avoidance functions in scenarios such as urban air traffic or logistics transportation, which greatly improves the safety of eVTOL low-altitude flight.

[0145] It should be noted that the embodiments of this application include, but are not limited to, the examples described above. It is understood that, under the guidance of the ideas in the embodiments of this application, those skilled in the art can make settings according to the actual situation, and this application does not impose any restrictions on them.

[0146] In this embodiment, multi-source data is acquired, obstacle avoidance reference information for manned aircraft is generated based on the multi-source data, the pose data of the manned aircraft is acquired, obstacle avoidance risk detection is performed based on the pose data and obstacle avoidance reference information, and a flight planning trajectory for the manned aircraft is generated based on the detection results of the obstacle avoidance risk detection. By generating obstacle avoidance reference information for manned aircraft, the comprehensiveness and accuracy of environmental perception are ensured. Furthermore, obstacle avoidance risk detection is performed based on the pose data of the manned aircraft and the obstacle avoidance reference information to accurately identify potential collision risks and generate a flight planning trajectory that matches the detection results of the obstacle avoidance risk detection. Thus, the flight planning trajectory can avoid obstacles with high precision over long distances, improving the safety of low-altitude flight of manned aircraft.

[0147] Corresponding to the aforementioned application function implementation method embodiments, this application also provides a low-altitude manned flight avoidance system, electronic equipment, and corresponding embodiments.

[0148] Figure 4 This is a schematic diagram illustrating the structure of a low-altitude manned flight obstacle avoidance system according to an embodiment of this application. See also... Figure 4 The system includes a perception and avoidance processor; wherein the perception and avoidance processor includes: The perception module 401 is used to acquire multi-source data, which includes at least digital terrain database data and image data identified by the sensors of the manned aircraft; and to generate obstacle avoidance reference information for the manned aircraft based on the multi-source data. The obstacle avoidance planning module 402 is used to acquire the attitude data of the manned aircraft; and to generate the flight planning trajectory of the manned aircraft based on the attitude data and obstacle avoidance reference information, so as to control the flight according to the flight planning trajectory.

[0149] In this embodiment, the perception and avoidance processor refers to a processor that combines environmental perception and decision-making / planning functions. During the environmental perception and decision-making / planning process, it performs operations such as fusion tracking and prediction, collision risk assessment and detection, and obstacle avoidance path planning on multi-source data and the pose data of the manned aircraft. The specific implementation methods of these operations can be referred to the content described in the preceding methods, and will not be repeated here.

[0150] As an optional example of this application, the system also includes a multi-source data acquisition module, which includes: The image acquisition module is used to acquire image data identified by the sensors of the manned aircraft and transmit it to the perception module 401. The digital terrain database module is used to collect digital terrain database data and transmit it to the sensing module 401; The communication module is used to collect communication data from the co-flying aircraft and transmit it to the sensing module 401. The perception module 401 generates information about the surrounding environment of the manned aircraft based on digital terrain database data and image data, and generates a predicted flight trajectory of the co-flying aircraft based on communication data and image data of the co-flying aircraft.

[0151] The image acquisition module is used to collect image data of the surrounding environment through sensors (such as cameras, infrared sensors, lidar, etc.) on the manned aircraft, and transmit the image data to the perception module 401 for analysis and decision-making.

[0152] The digital terrain database module is used to acquire data from the digital terrain database, providing low-altitude elevation data for buildings and complex terrain, which is used for flight path planning and obstacle avoidance data input. The digital terrain database typically contains terrain information for the current flight airspace, including ground elevation, mountain ranges, and building locations.

[0153] The digital terrain database module can transmit digital terrain database data to the perception module 401, which can analyze the flight environment of the manned aircraft and assess obstacles or terrain limitations that may be encountered during flight.

[0154] The communication module is used to communicate and interact with other aircraft operating in the same field, collecting their communication data to achieve location information sharing and monitor the flight trajectories of cooperative aircraft. This communication data typically includes information such as the cooperative aircraft's location, flight trajectory, and flight status.

[0155] The communication module can transmit communication data to the perception module 401, which can then predict the flight trajectory of the co-flying aircraft.

[0156] In addition, the system includes a positioning module, which uses a combination of various navigation devices and sensors to achieve high-precision positioning, ensuring the accuracy of map matching and route planning. The navigation devices include RTK (Real-Time Kinematic Device) and IMU (Inertial Measurement Unit) devices, while the sensors include camera sensors, laser sensors, millimeter-wave sensors, and more.

[0157] In this application, the perception module 401 can integrate multi-source data provided by the image acquisition module, the digital terrain database module, the communication module, and the positioning module. By inputting multi-source data and combining map and positioning information, it can obtain the surrounding environment of the manned aircraft and predict the flight trajectory of other aircraft in the same field.

[0158] As an optional example of this application, the perception module 401 includes: The data acquisition submodule is used to acquire multi-source data, including digital terrain database data, image data identified by the sensors of the manned aircraft, and communication data of other aircraft in the same field. The data fusion submodule is used to extract DSM data from the digital terrain database; to obtain an environmental point cloud model by combining the DSM data and image data; and to perform vector information fusion on the DSM data and image data based on the environmental point cloud model to obtain the surrounding environment information of the manned aircraft. The tracking and prediction submodule is used to detect the 3D pose information of dynamic obstacles around the manned aircraft using image data and environmental point cloud models. The 3D pose information of the dynamic obstacles includes at least the 3D pose information of the aircraft in the same field. Based on the 3D pose information of the aircraft in the same field and the communication data of the aircraft in the same field, the submodule generates the predicted flight trajectory of the aircraft in the same field. The 3D pose information of the dynamic obstacles also includes the 3D pose information of the flying obstacles. The tracking and prediction submodule is also used to generate the predicted flight trajectory of the flying obstacles based on the 3D pose information of the flying obstacles.

[0159] It can be seen that the data fusion submodule mainly processes static obstacles to obtain information about the surrounding environment, while the tracking and prediction submodule mainly processes dynamic obstacles to obtain the flight trajectories of dynamic obstacles such as co-flying aircraft, non-cooperative aircraft, and birds.

[0160] The perception module 401 utilizes the data fusion submodule and the tracking and prediction submodule to achieve global environmental perception, thereby accurately outputting "static + dynamic" obstacle avoidance reference information to the obstacle avoidance planning module 402, ensuring the flight safety of the manned aircraft. The specific implementation methods of the data fusion submodule and the tracking and prediction submodule can be found in the descriptions above, and will not be repeated here.

[0161] As an optional example of this application, obstacle avoidance planning module 402 includes: The collision avoidance risk assessment and detection submodule is used to detect obstacle avoidance risks based on pose data and obstacle avoidance reference information. The obstacle avoidance path planning submodule is used to generate the flight path planning trajectory of the manned aircraft based on the detection results of obstacle avoidance risk detection.

[0162] Among them, the obstacle avoidance planning module 402 uses the collision risk assessment and detection submodule to analyze the flight environment and identify collision risks, and uses the obstacle avoidance path planning submodule to generate the flight planning trajectory of the manned aircraft to ensure the safe flight of the aircraft.

[0163] To enable those skilled in the art to better understand the technical solutions of the embodiments of this application, an example is provided below, along with reference to... Figure 5 The illustrated perception and avoidance system block diagram illustrates an embodiment of this application.

[0164] The low-altitude manned flight avoidance system of the manned aircraft is a perception avoidance system. The system includes a multi-source data acquisition module and a perception avoidance processor. The perception avoidance processor includes a perception module and an obstacle avoidance planning module. The perception avoidance processor is also connected to the flight control system, navigation and attitude system and display and alarm system of the manned aircraft.

[0165] The multi-source data acquisition module can collect intrusion obstacle data using the image acquisition module, digital terrain database module, and communication module. Intrusion obstacle data includes cooperative obstacles, non-cooperative obstacles, unknown terrain, and known terrain.

[0166] In simple terms, cooperative obstacles are those that can share and interact with manned aircraft, while non-cooperative obstacles are those that cannot. For example, cooperative obstacles can be the aforementioned cooperative aircraft and other communication electronic devices, while non-cooperative obstacles can be the aforementioned non-cooperative aircraft, birds, and other electronic devices that cannot communicate.

[0167] The perception and avoidance processor can receive multi-source data from the multi-source data acquisition module and pose status from the navigation pose system. It uses the perception module to perform fusion tracking and prediction, and the obstacle avoidance planning module to perform collision risk assessment and obstacle avoidance path planning. Finally, it outputs obstacle avoidance reference information and the flight trajectory of the aircraft in the same field.

[0168] Among them, fusion tracking and prediction can fuse data from different sensors, combine the target's historical and current states, estimate the target's current position, velocity, orientation, and other states, and predict its future trajectory or behavior. Collision risk assessment and detection can perform collision detection on the future trajectories of different objects, promptly identifying and warning of potential collision hazards.

[0169] The obstacle avoidance planning module generates at least one flight planning trajectory based on obstacle avoidance reference information and the flight trajectories of other aircraft in the same field. At this time, the perception and avoidance processor can convert the flight planning trajectory into control commands and send the control commands to the flight control system. The flight control system then controls the manned aircraft to fly at low altitude and sends the obstacle avoidance reference information to the display and alarm system to alert the operator.

[0170] During low-altitude flight, the positions and attitudes of the manned aircraft and other aircraft in the same field are updated in real time, which is used as input data for the next low-altitude manned flight avoidance maneuver.

[0171] In this embodiment, the perception and obstacle avoidance processor includes: a perception module 401, used to acquire multi-source data, wherein the multi-source data includes at least digital terrain database data and image data identified by the sensors of the manned aircraft; generating obstacle avoidance reference information for the manned aircraft based on the multi-source data; and an obstacle avoidance planning module 402, used to acquire the pose data of the manned aircraft; and generating a flight planning trajectory for the manned aircraft based on the pose data and obstacle avoidance reference information, so as to control flight according to the flight planning trajectory.

[0172] The perception and obstacle avoidance processor in this application integrates a perception module 401 and an obstacle avoidance planning module 402. The perception module 401 can acquire multi-source data in real time, including digital terrain database data and manned aircraft sensor image data, and generate comprehensive obstacle avoidance reference information to support obstacle avoidance decisions. The obstacle avoidance planning module 402 accurately calculates the flight trajectory of the manned aircraft based on the manned aircraft's pose data and the obstacle avoidance reference information generated by the perception module. This not only ensures that the manned aircraft can safely and efficiently complete its flight mission in a low-altitude flight environment, but also significantly enhances the manned aircraft's adaptability.

[0173] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated further here.

[0174] Figure 6 This is a schematic diagram of the structure of a low-altitude manned aircraft shown in an embodiment of this application.

[0175] See Figure 6 The low-altitude manned aircraft 600 includes a memory 610 and a processor 620.

[0176] The processor 620 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0177] Memory 610 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by the processor 620 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 610 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, memory 610 may include a removable storage device that is readable and / or writable, such as a laser disc (CD), a read-only digital multifunction optical disc (e.g., DVD-ROM, dual-layer DVD-ROM), a read-only Blu-ray disc, an ultra-high density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not contain carrier waves or transient electronic signals transmitted wirelessly or via wired connections.

[0178] The memory 610 stores executable code, which, when processed by the processor 620, can cause the processor 620 to execute part or all of the methods described above.

[0179] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0180] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.

[0181] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for low-altitude manned flight obstacle avoidance, characterized in that, The method includes: Acquire multi-source data, wherein the multi-source data includes at least digital terrain database data and image data identified by the sensors of the manned aircraft; Based on the multi-source data, obstacle avoidance reference information is generated for the manned aircraft. Acquire the pose data of the manned spacecraft; The flight planning trajectory of the manned aircraft is generated based on the pose data and the obstacle avoidance reference information, so as to control the flight according to the flight planning trajectory.

2. The method according to claim 1, characterized in that, The multi-source data also includes communication data from other aircraft operating in the same field. The step of generating obstacle avoidance reference information for the manned aircraft based on the multi-source data includes: The surrounding environment information of the manned aircraft is generated based on the digital terrain database data and the image data, and the flight trajectory of the co-flying aircraft is predicted based on the communication data of the co-flying aircraft and the image data.

3. The method according to claim 2, characterized in that, The step of generating the surrounding environment information of the manned aircraft based on the digital terrain database data and the image data includes: Extract DSM data from the digital terrain database; An environmental point cloud model is obtained by combining the DSM data and the image data. Based on the environmental point cloud model, vector information fusion is performed on the DSM data and the image data to obtain the surrounding environment information of the manned aircraft.

4. The method according to claim 3, characterized in that, The step of predicting the flight trajectory of the co-flying aircraft based on the communication data and image data of the co-flying aircraft includes: The image data and the environmental point cloud model are used to detect the 3D pose information of dynamic obstacles around the manned aircraft, wherein the 3D pose information of the dynamic obstacles includes at least the 3D pose information of the aircraft in the same field. Based on the 3D pose information and communication data of the co-flying aircraft, a predicted flight trajectory of the co-flying aircraft is generated.

5. The method according to claim 4, characterized in that, The dynamic obstacle 3D pose information also includes the 3D pose information of flying obstacles, and the method further includes: The predicted flight trajectory of the flying obstacle is generated based on the 3D pose information of the flying obstacle.

6. The method according to any one of claims 1 to 5, characterized in that, The step of generating the flight planning trajectory of the manned aircraft based on the pose data and the obstacle avoidance reference information includes: Obstacle avoidance risk detection is performed based on the pose data and the obstacle avoidance reference information; Based on the obstacle avoidance risk detection results, the flight planning trajectory of the manned aircraft is generated.

7. A low-altitude manned flight obstacle avoidance system, characterized in that, The system includes a perception and avoidance processor; wherein the perception and avoidance processor includes: A perception module is used to acquire multi-source data, wherein the multi-source data includes at least digital terrain database data and image data identified by the sensors of the manned aircraft; and to generate obstacle avoidance reference information for the manned aircraft based on the multi-source data. The obstacle avoidance planning module is used to acquire the position and attitude data of the manned aircraft; generate the flight planning trajectory of the manned aircraft based on the position and attitude data and the obstacle avoidance reference information, and control the flight according to the flight planning trajectory.

8. The system according to claim 7, characterized in that, The system also includes a multi-source data acquisition module, which comprises: The image acquisition module is used to acquire image data identified by the sensors of the manned aircraft through sensors and transmit it to the perception module; The digital terrain database module is used to collect digital terrain database data and transmit it to the sensing module; The communication module is used to collect communication data from the co-flying aircraft and transmit it to the sensing module; The perception module generates the surrounding environment information of the manned aircraft based on the digital terrain database data and the image data, and generates the predicted flight trajectory of the co-flying aircraft based on the communication data of the co-flying aircraft and the image data.

9. The system according to claim 7, characterized in that, The sensing module includes: The data acquisition submodule is used to acquire multi-source data, which includes digital terrain database data, image data identified by the sensors of the manned aircraft, and communication data of the co-flying aircraft. The data fusion submodule is used to extract DSM data from the digital terrain database; obtain an environmental point cloud model by combining the DSM data and the image data; and perform vector information fusion on the DSM data and the image data based on the environmental point cloud model to obtain the surrounding environment information of the manned aircraft. The tracking and prediction submodule is used to detect the 3D pose information of dynamic obstacles around the manned aircraft using the image data and the environmental point cloud model, wherein the 3D pose information of the dynamic obstacles includes at least the 3D pose information of the co-flying aircraft; based on the 3D pose information of the co-flying aircraft and the communication data of the co-flying aircraft, it generates a predicted flight trajectory of the co-flying aircraft; wherein the 3D pose information of the dynamic obstacles also includes the 3D pose information of flying obstacles, and the tracking and prediction submodule is also used to generate a predicted flight trajectory of the flying obstacles based on the 3D pose information of the flying obstacles.

10. The system according to any one of claims 7 to 9, characterized in that, The obstacle avoidance planning module includes: The collision avoidance risk assessment and detection submodule is used to perform obstacle avoidance risk detection based on the pose data and the obstacle avoidance reference information. The obstacle avoidance path planning submodule is used to generate the flight planning trajectory of the manned aircraft based on the detection results of obstacle avoidance risk detection.

11. A low-altitude manned aircraft, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-6.

Images