Control method, device and equipment for intelligent unmanned aerial vehicle with body and storage medium

Abstract

The embodiment of the invention provides a control method, device and equipment for an intelligent unmanned aerial vehicle with a body and a storage medium. The method comprises the steps of receiving a target task issued by an intelligent platform; in the process of controlling the flight of the unmanned aerial vehicle main body according to the target task, acquiring current environment information collected by a sensor; fusing the current environment information collected by different sensors to obtain a target environment model, wherein the target environment model is used for describing the current environment of the intelligent unmanned aerial vehicle; generating a behavior instruction according to the target environment model and the target task; and controlling the unmanned aerial vehicle body and the execution mechanism to execute the target task according to the behavior instruction. According to the embodiment of the invention, through the intelligent decision-making capability of the intelligent system and the intelligent platform and the efficient cooperation of the unmanned aerial vehicle main body and the execution mechanism, the real-time environment information collected by various sensors is fused, so that the intelligent unmanned aerial vehicle with the body has the capability of coping with a complex environment; and closed-loop control of real-time sensing, decision making and execution of the intelligent unmanned aerial vehicle with the body is realized in an unstructured environment.

Description

Technical Field

[0001] The present application relates to the technical field of unmanned aerial vehicles, and particularly to a control method and device for an embodied intelligent unmanned aerial vehicle, an electronic device, and a storage medium. Background Art

[0002] When a traditional unmanned aerial vehicle system operates, it usually relies on a preset path and manual intervention, lacking the ability of autonomous decision-making. This mode shows significant limitations in complex environments or dynamic tasks. For example, in complex environments such as mountainous areas, forests, or urban high-rise areas, it is difficult for manually controlled unmanned aerial vehicles to effectively avoid obstacles, and collision accidents occur frequently. In addition, the perception and control of traditional unmanned aerial vehicle systems are mostly designed in a separated manner, lacking an integrated intelligent system architecture. This separated design leads to bottlenecks in information interaction and task execution efficiency of the system, and it is difficult to achieve efficient collaborative work.

[0003] Although the prior art has improved the traditional unmanned aerial vehicle system to a certain extent and enhanced its autonomy and intelligence level, the existing unmanned aerial vehicle systems still cannot achieve a real-time cognitive, decision-making, and execution closed-loop in unstructured environments, thus limiting the application effect and reliability of unmanned aerial vehicles in complex environments. Summary of the Invention

[0004] Embodiments of the present application provide a control method for an embodied intelligent unmanned aerial vehicle to solve the problem that the existing unmanned aerial vehicle systems still cannot achieve a real-time cognitive, decision-making, and execution closed-loop in unstructured environments, thus limiting the application effect and reliability of unmanned aerial vehicles in complex environments.

[0005] Correspondingly, embodiments of the present application also provide a control device for an embodied intelligent unmanned aerial vehicle, an electronic device, and a storage medium to ensure the implementation and application of the above method.

[0006] To solve the above problems, embodiments of the present application disclose a control method for an embodied intelligent unmanned aerial vehicle, which is applied to an embodied intelligent unmanned aerial vehicle. The embodied intelligent unmanned aerial vehicle includes an intelligent system, an unmanned aerial vehicle body, and an execution mechanism. The intelligent system is communicatively connected to an intelligent platform, and the intelligent system is connected to a sensor. The method includes:

[0007] Receiving a target task issued by the intelligent platform;

[0008] During the process of controlling the flight of the unmanned aerial vehicle body according to the target task, acquiring the current environmental information collected by the sensor;

[0009] Fusing the current environmental information collected by different sensors to obtain a target environmental model, where the target environmental model is used to describe the environment where the embodied intelligent unmanned aerial vehicle is currently located;

[0010] Generate a behavior instruction based on the target environment model and the target task;

[0011] Control the drone body and the actuator to execute the target task according to the behavior instruction.

[0012] Optionally, the intelligent system includes a navigation module. The receiving of the target task sent by the intelligent platform includes:

[0013] Obtain the function information of the embodied intelligent drone and the position information of the drone body in the navigation module;

[0014] Upload the function information and the position information to the intelligent platform;

[0015] Receive the target task allocated by the intelligent platform according to the function information and the position information.

[0016] Optionally, the intelligent system includes a decision-making module. The generating of the behavior instruction according to the target environment model and the target task includes:

[0017] Determine the current reasoning task according to the target environment model and the target task;

[0018] Use the decision-making module to reason about the current reasoning task to obtain a reasoning result;

[0019] Generate the behavior instruction according to the reasoning result;

[0020] Or,

[0021] If an exception occurs when the decision-making module reasons about the current reasoning task, transmit the current reasoning task to the intelligent platform;

[0022] Receive the reasoning result obtained by the intelligent platform's reasoning about the current reasoning task;

[0023] Generate the behavior instruction according to the reasoning result.

[0024] Optionally, the embodied intelligent drone further includes a gimbal. The controlling of the drone body and the actuator to execute the target task according to the behavior instruction includes:

[0025] Obtain the gimbal steering instruction, flight path, and action instruction in the behavior instruction. The action instruction has a corresponding target action;

[0026] Control the gimbal to turn according to the gimbal steering instruction; and control the drone body to fly along the flight path; and control the actuator to execute the target action according to the action instruction to complete the target task.

[0027] Optionally, the intelligent system includes a navigation module, and controlling the drone body to fly along the flight path includes:

[0028] Obtaining navigation information of the navigation module;

[0029] Controlling the drone body to fly along the flight path according to the navigation information.

[0030] Optionally, the decision-making module is deployed with an intelligent reasoning model, and using the decision-making module to perform reasoning on the current reasoning task to obtain a reasoning result includes:

[0031] Performing reasoning on the current reasoning task using the intelligent reasoning model to obtain a reasoning result;

[0032] Uploading the reasoning result to the intelligent platform;

[0033] After the intelligent platform updates the intelligent reasoning model according to the reasoning result, obtaining the updated intelligent reasoning model;

[0034] Deploying the updated intelligent reasoning model in the decision-making module.

[0035] Optionally, the method further includes:

[0036] Collecting target image data and target point cloud data according to the target task;

[0037] Using the intelligent system to perform image recognition on the target image data and preprocess the target point cloud data to obtain an image recognition result and preprocessed target point cloud data;

[0038] Uploading the image recognition result and the preprocessed target point cloud data to the intelligent platform.

[0039] An embodiment of the present application also discloses a control device for an embodied intelligent drone, which is applied to an embodied intelligent drone. The embodied intelligent drone includes an intelligent system, a drone body, and an execution mechanism. The intelligent system is communicatively connected to an intelligent platform, and the intelligent system is connected to a sensor. The device includes:

[0040] A task receiving module, configured to receive a target task sent by the intelligent platform;

[0041] An information obtaining module, configured to obtain current environment information collected by the sensor during the process of controlling the drone body to fly according to the target task;

[0042] An information fusion module, configured to fuse the current environment information collected by different sensors to obtain a target environment model, where the target environment model is used to describe the environment where the intelligent unmanned aerial vehicle is currently located;

[0043] An instruction generation module, configured to generate a behavior instruction according to the target environment model and the target task;

[0044] A task execution module, configured to control the unmanned aerial vehicle body and the execution mechanism to execute the target task according to the behavior instruction.

[0045] An embodiment of the present application also discloses an electronic device, including: a processor; and a memory, on which executable code is stored, and when the executable code is executed, the processor is caused to execute the control method of one or more of the embodied intelligent unmanned aerial vehicles in the embodiments of the present application.

[0046] An embodiment of the present application also discloses one or more machine-readable media, on which executable code is stored, and when the executable code is executed, the processor is caused to execute the control method of one or more of the embodied intelligent unmanned aerial vehicles in the embodiments of the present application.

[0047] Compared with the prior art, the embodiments of the present application have the following advantages:

[0048] In the embodiments of the present application, a target task sent by an intelligent platform is received; during the process of controlling the flight of the unmanned aerial vehicle body according to the target task, the current environment information collected by sensors is obtained; the current environment information collected by different sensors is fused to obtain a target environment model, where the target environment model is used to describe the environment where the intelligent unmanned aerial vehicle is currently located; a behavior instruction is generated according to the target environment model and the target task; and the unmanned aerial vehicle body and the execution mechanism are controlled to execute the target task according to the behavior instruction. In the embodiments of the present application, the intelligent system is combined with the unmanned aerial vehicle body, and a communication connection is established with the intelligent platform. Through the intelligent decision-making capabilities of the intelligent system and the intelligent platform, as well as the efficient cooperation of the unmanned aerial vehicle body and the execution mechanism, the efficiency of the unmanned aerial vehicle in executing the target task is significantly improved. The intelligent unmanned aerial vehicle can also fuse the real-time environment information collected by multiple sensors, enabling it to have the ability to cope with complex environments, thereby realizing the closed-loop control of real-time perception, decision-making, and execution of the intelligent unmanned aerial vehicle in an unstructured environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flowchart of the steps of an embodiment of the control method of an embodied intelligent unmanned aerial vehicle of the present application;

[0050] Figure 2 is a schematic diagram of the communication connection of an embodiment of the control method of an embodied intelligent unmanned aerial vehicle of the present application;

[0051] Figure 3Schematic diagram of the hardware module of an embodiment of the control method for an embodied intelligent unmanned aerial vehicle of the present application;

[0052] Figure 4 Schematic diagram of the software module of an embodiment of the control method for an embodied intelligent unmanned aerial vehicle of the present application;

[0053] Figure 5 Flowchart of system initialization of an embodiment of the control method for an embodied intelligent unmanned aerial vehicle of the present application;

[0054] Figure 6 Schematic diagram of task execution of an embodiment of the control method for an embodied intelligent unmanned aerial vehicle of the present application;

[0055] Figure 7 Block diagram of the structure of an embodiment of the control device for an embodied intelligent unmanned aerial vehicle of the present application;

[0056] Figure 8 Schematic diagram of the structure of a device provided by an embodiment of the present application. Detailed implementation manners

[0057] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0058] When a traditional unmanned aerial vehicle system works, it usually relies on a preset path and manual intervention, lacking the ability of autonomous decision-making. This mode shows significant limitations in complex environments or dynamic tasks. For example, in complex environments such as mountainous areas, forests, or urban high-rise areas, it is difficult for manually controlled unmanned aerial vehicles to effectively avoid obstacles, and collision accidents occur frequently. In addition, the perception and control of traditional unmanned aerial vehicle systems are mostly designed in a separated manner, lacking an integrated intelligent system architecture. This separated design leads to bottlenecks in information interaction and task execution efficiency of the system, and it is difficult to achieve efficient collaborative work.

[0059] Although the prior art has improved the traditional unmanned aerial vehicle system to a certain extent and enhanced the autonomy and intelligence level of traditional unmanned aerial vehicles, there are still the following disadvantages:

[0060] 1. Limited perception ability: The perception systems of existing unmanned aerial vehicles are mostly designed in a separated manner, lacking an integrated intelligent system, resulting in insufficient perception accuracy and environmental adaptability. For example, in complex environments, it is difficult for unmanned aerial vehicles to effectively avoid obstacles and collisions are likely to occur.

[0061] 2. Low decision-making efficiency: The decision-making systems of existing unmanned aerial vehicles are mostly simple logics based on rules, lacking the ability to dynamically adapt to complex environments. For example, in a dynamic traffic environment, it is difficult for traditional autonomous driving systems to make optimal decisions in real time.

[0062] 3. Insufficient execution accuracy: The control accuracy and response speed of the actuators of existing drones need to be improved. For example, in collaborative transportation tasks, the collaborative accuracy and motion control accuracy of multi-drone systems are insufficient, affecting the task execution efficiency.

[0063] 4. Low data processing and analysis efficiency: Processing and analyzing the large amounts of image, video, etc. data collected by drones manually is time-consuming and laborious, and it is difficult to obtain valuable information in a timely manner.

[0064] To this end, the embodiments of the present application provide a control method for an embodied intelligent drone. By using an intelligent system as the decision-making system and the drone body and actuators as the perception and motion systems, the separation of the perception system and the decision-making system is realized. Based on the perception function of the drone body, various sensors (such as visual sensors, lidar, millimeter-wave radars, etc.) are used to obtain environmental information, and an environmental model is constructed through data fusion technology. Further, the target task is decomposed into high-level behavioral decisions and low-level motion controls, which are collaboratively completed by the intelligent system and the drone body and actuators respectively. Among them, high-level behavioral decisions are responsible for formulating the task objectives and strategies of the drone, which may include task planning, path planning, behavior selection, collaborative decision-making, etc.; low-level motion control is responsible for converting high-level decisions into specific drone motion instructions to ensure that the drone executes tasks stably and accurately. The embodiments of the present application significantly improve the perception accuracy, decision-making efficiency, and execution accuracy of intelligent drones, effectively solve the problem that existing drone systems are difficult to achieve real-time perception, decision-making, and execution closed-loop in unstructured environments, and thus break through the limitations of the application effect and reliability of drones in complex environments.

[0065] Refer to Figure 1 , which is a flowchart of the steps of an embodiment of the control method for an embodied intelligent drone of the present application, including the following steps:

[0066] Step 101, receive the target task sent by the intelligent platform.

[0067] The control method for an embodied intelligent drone shown in the embodiments of the present application is applied to an embodied intelligent drone. An embodied intelligent drone refers to a drone system with high autonomy and intelligence, which can complete specific tasks in complex environments through capabilities such as perception, decision-making, and execution. The number of embodied intelligent drones can be one or more. An embodied intelligent drone includes three parts: an intelligent system, a drone body, and actuators. Among them, the intelligent system is communicatively connected to the intelligent platform, and the intelligent system is connected to the sensors.

[0068] Refer to Figure 2 , which is a schematic diagram of the communication connection of an embodiment of the control method for an embodied intelligent drone of the present application.

[0069] As Figure 2As shown in the figure, the embodied intelligent drone shown in the embodiment of the present application can be divided into three parts, namely the drone body, the intelligent system, and the actuator. The intelligent drone can be communicatively connected to the intelligent platform through a communication network. Specifically, the intelligent system in the embodied intelligent drone is communicatively connected to the intelligent platform through a communication network. In one embodiment, the communication network may include communication networks such as 4G (the 4th Generation mobile communication technology), 5G (the 5th Generation mobile communication technology), 5G Redcap (5G Reduced Capability), and satellite communication.

[0070] In an alternative embodiment, the intelligent system can be an embodied intelligent X-BRAIN edge system, the intelligent platform can be an X-BRAIN intelligent platform, and an intelligent actuator is provided on the embodied intelligent drone. Embodied Intelligence is a form of intelligence that emphasizes that an intelligent agent realizes perception, learning, and decision-making through interaction with the physical environment. Different from traditional intelligence based on pure algorithms or data, embodied intelligence emphasizes the interaction between the "body" of an intelligent agent (such as a robot, a drone, etc.) and its environment in the real world, so as to achieve more natural and efficient intelligent behavior.

[0071] In this embodiment, the embodied intelligent X-BRAIN edge system is a highly integrated intelligent platform integrated in the embodied intelligent drone, and has three core functions: self-perception, environment perception, and strategic planning. The self-perception module monitors the state and position of the drone in real time through a cerebellum-like mechanism to ensure precise control; the environment perception module uses vision-like and auditory capabilities to identify surrounding objects and changes; the strategic planning module conducts task planning and decision-making through a brain-like mechanism to help the drone achieve strategic actions in complex environments. The embodied intelligent X-BRAIN edge system also has powerful AI (Artificial Intelligence) inference and image and video processing capabilities, supporting functions such as multi-channel video processing, neural network inference, model function upgrade, remote control and monitoring of the drone body, and communication and collaboration of multiple intelligent drones.

[0072] The embodied intelligent X-BRAIN edge system is communicatively connected to the X-BRAIN intelligent platform. The X-BRAIN intelligent platform consists of a training module and an inference module, and supports various algorithms such as deep learning and reinforcement learning. The training module is responsible for model training and function upgrade, and the inference module processes complex inference tasks such as multi-modal data fusion and advanced decision-making.

[0073] The embodied intelligent drone integrated with the X-BRAIN edge system can optimize its behavior decision-making and task execution capabilities in complex environments through the cycle of real-time perception, feedback, and action execution, combined with the learning and adaptation capabilities of embodied intelligence. It can be widely applied to scenarios such as highway inspection, water sampling, offshore vessel landing, logistics container loading, and disaster rescue. It has comprehensive capabilities of flight, perception, computing, decision-making, and operation, supports real-time inference tasks such as target detection and path planning, and communicates with the X-BRAIN intelligent platform through 4G / 5G networks to achieve efficient data processing and task execution.

[0074] The intelligent execution mechanism on the embodied intelligent drone integrates vision, tactile, torque, and inertial sensors, as well as servo motors such as frameless torque motors and coreless motors, combined with reducers such as precision planetary reducers and harmonic reducers, and multi-degree-of-freedom manipulators, etc., to achieve complex actions such as positioning, classification, picking, and placing. This mechanism has the ability of autonomous decision-making and can complete high-precision task operations in complex environments, providing strong execution support for the embodied intelligent drone.

[0075] In the embodiment of this application, in step 101, after the intelligent platform issues a target task, the intelligent system on the embodied intelligent drone receives the target task. For example, on highways, abnormal events such as traffic accidents, vehicle illegal parking, or spillage often occur. Due to factors such as complex traffic conditions and long distances, the disposal personnel often cannot reach the scene in the first place, which is likely to cause secondary accidents. The intelligent platform can issue a target task related to the handling of abnormal events to the embodied intelligent drone, and the embodied intelligent drone can quickly reach the scene and execute the task, thereby improving the emergency disposal efficiency and reducing the risk of secondary accidents.

[0076] Step 102, during the process of controlling the flight of the drone body according to the target task, obtain the current environmental information collected by the sensors.

[0077] In step 102, after receiving the target task, the embodied intelligent drone needs to fly to the target location to execute the target task. Under the control of the intelligent system according to the target task, the drone body flies towards the target location. During this process, the intelligent system can obtain the environmental information of the current environment where the embodied intelligent drone is located through various sensors, such as image information, climate information, distance information from surrounding objects, and terrain information where it is currently located. In one embodiment, the sensors can include cameras, millimeter-wave radars, lidar, etc., and sensors such as cameras can be set on the drone body.

[0078] Step 103: Integrate the current environmental information collected by different sensors to obtain a target environmental model, which is used to describe the environment where the embodied intelligent drone is currently located.

[0079] In step 103, after obtaining the current environmental information collected by multiple different sensors, perform multimodal data fusion on the current environmental information, so as to obtain a target environmental model for describing the environment where the embodied intelligent drone is currently located.

[0080] Specifically, first, the current environmental information of different sensors can be sorted out through data preprocessing, such as methods of time alignment, coordinate transformation, filtering, and standardization. Then, key information can be obtained from the preprocessed current environmental information through feature extraction. For example, image edges can be extracted from the camera, geometric shapes can be extracted from LiDAR (Light Detection And Ranging), and distance and speed can be extracted from the radar. Then, a data fusion method is used to integrate multi-source data. The data fusion method can include rule-based fusion, probability-based fusion, deep learning-based fusion, or graph model-based fusion, etc. After that, a target environmental model is constructed on this basis, such as a three-dimensional map, an obstacle map, a semantic map, or a dynamic environmental model, etc., and the accuracy and real-time performance of the environmental model are improved through model optimization and update. The embodiments of the present application do not impose any restrictions on the specific manner of data fusion, and those skilled in the art can select a suitable data fusion method according to the actual situation.

[0081] Step 104: Generate a behavior instruction according to the target environmental model and the target task.

[0082] In step 104, the intelligent system can reason and generate a behavior instruction according to the target environmental model and the target task. The behavior instruction is used to control the flight motion of the drone body and the actions of the execution mechanism, such as instructions for flight motions like translation, rotation, path planning, and speed control, as well as action instructions for the execution mechanism such as gimbal stabilization and landing gear operation. In one embodiment, when the intelligent system is unable to reason according to the target environmental model and the target task, the intelligent platform can intervene for reasoning to obtain a behavior instruction.

[0083] Step 105: Control the drone body and the execution mechanism to execute the target task according to the behavior instruction.

[0084] In step 105, after obtaining the behavior instruction, the drone body and the execution mechanism execute the corresponding behavior instruction, thereby completing the target task.

[0085] In the embodiment of the present application, an intelligent system is combined with the UAV body, and a communication connection is established with an intelligent platform. Through the intelligent decision-making capabilities of the intelligent system and the intelligent platform, as well as the efficient coordination of the UAV body and the actuator, the efficiency of the UAV in performing target tasks is significantly improved. The embodied intelligent UAV can also integrate real-time environmental information collected by multiple sensors, enabling it to have the ability to cope with complex environments, thereby realizing closed-loop control of real-time perception, decision-making, and execution of the embodied intelligent UAV in an unstructured environment.

[0086] In an alternative embodiment of the present application, the intelligent system includes a navigation module. The receiving of the target task sent by the intelligent platform includes:

[0087] Obtain the function information of the embodied intelligent UAV and the position information of the UAV body in the navigation module;

[0088] Upload the function information and the position information to the intelligent platform;

[0089] Receive the target task allocated by the intelligent platform according to the function information and the position information.

[0090] In this embodiment, the intelligent system may further include a navigation module, which can realize the navigation and positioning of the embodied intelligent UAV based on GPS (Global Positioning System), RTK (Real-Time Kinematic), or Beidou.

[0091] Among them, when performing task allocation, tasks can be allocated according to the distance between the embodied intelligent UAV and the target location of the target task, as well as the functions supported by the embodied intelligent UAV. When allocating tasks, the function information of the embodied intelligent UAV and the position information of the UAV body in the navigation module can be obtained first. After uploading the function information and the position information to the intelligent platform, the intelligent platform allocates the target task to the corresponding embodied intelligent UAV according to the function information and the position information. In one embodiment, the target task can be allocated to the embodied intelligent UAV that is closer to the target location and has equipment supporting the execution of the target task. The embodied intelligent UAV receives the target task allocated by the intelligent platform according to the function information and the position information through the intelligent system and executes the target task.

[0092] The embodiment of the present application performs dynamic task allocation based on the intelligent platform to achieve efficient task allocation, preferentially selects the embodied intelligent UAV that is close to the target and has the corresponding functions, and improves the task execution efficiency and resource utilization rate.

[0093] In an alternative embodiment of the present application, the intelligent system includes a decision-making module, and generating a behavior instruction according to the target environment model and the target task includes:

[0094] Determine the current inference task according to the target environment model and the target task;

[0095] Use the decision-making module to perform inference on the current inference task to obtain an inference result;

[0096] Generate the behavior instruction according to the inference result;

[0097] Or,

[0098] If an exception occurs when the decision-making module performs inference on the current inference task, then transmit the current inference task to the intelligent platform;

[0099] Receive the inference result obtained by the intelligent platform performing inference on the current inference task;

[0100] Generate the behavior instruction according to the inference result.

[0101] In this embodiment, the intelligent system may further include a decision-making module, and the decision-making module is responsible for task planning and strategy generation, that is, the decision-making module can process the current inference task during the process of the embodied intelligent drone executing the target task, and drive the embodied intelligent drone to continue executing the target task according to the inference result. The inference result may include changing the flight path, adjusting the altitude, adjusting the flight speed, starting a specific device (such as a camera), etc.

[0102] Specifically, determine the current inference task according to the target environment model and the target task. In one embodiment, the target environment model can be analyzed, such as analyzing the terrain, obstacle distribution, and climate conditions; decomposing the task requirements of the target task, and decomposing the target task into specific subtasks or action sequences. For example, if the target task is to "inspect a certain area", it can be decomposed into subtasks such as "fly to a certain area", "take a photo", and "return to the starting point". Further, determine the current inference task according to the target environment model and the target task. For example, if the current target task is executed to the stage of the subtask "fly to a certain area", and after analyzing the target environment model, it is found that there is an obstacle in front of the preset flight path, then the current inference task can be determined as "how to avoid the obstacle".

[0103] After determining the current inference task, the local inference model integrated in the decision-making module is used to perform inference on the current inference task to obtain an inference result. In one embodiment, if the decision-making module fails to perform inference on the current inference task, such as when the inference times out, it is determined that an abnormality has occurred, and the current inference task is transmitted to the intelligent platform. After the intelligent platform completes the inference, the intelligent system then receives the inference result obtained by the intelligent platform for the current inference task. After obtaining the inference result, a behavior instruction for the embodied intelligent drone is generated according to the inference result, so that the embodied intelligent drone can continue to execute the task.

[0104] In one embodiment, the intelligent system may also have a real-time decision correction mechanism, enabling the embodied intelligent drone to quickly correct the original decision according to real-time data and environmental changes during the task execution process to ensure the smooth completion of the task. For example, when detecting a sudden strong wind, adjust the flight altitude and speed to maintain stable flight; when detecting an obstacle, re-plan the flight path to avoid the obstacle; when the battery power is insufficient, adjust the task priority or plan a return route.

[0105] Through the intelligent decision-making capabilities of the intelligent system and the intelligent platform, the embodiments of this application perform inference on the current inference task of the embodied intelligent drone to execute the target task, thereby driving the embodied intelligent drone to complete the target task and improving the efficiency of task completion of the embodied intelligent drone.

[0106] In an alternative embodiment of this application, the embodied intelligent drone further includes a gimbal. The controlling the drone body and the actuating mechanism to execute the target task according to the behavior instruction includes:

[0107] Obtain the gimbal steering instruction, flight path, and action instruction in the behavior instruction, and the action instruction has a corresponding target action;

[0108] Control the gimbal to turn according to the gimbal steering instruction; and control the drone body to fly along the flight path; and control the actuating mechanism to execute the target action according to the action instruction to complete the target task.

[0109] In this embodiment, the embodied intelligent drone is also equipped with a gimbal. A gimbal is a mechanical device used to stabilize and control a camera or other sensor devices, usually composed of motors, sensors, and control systems, and can achieve stable rotation and precise positioning of the device in multiple axes.

[0110] The behavior instructions can at least include pan-tilt steering instructions, flight paths, and action instructions, which are used to control the pan-tilt steering, the flight direction of the UAV body, and the actions of the actuators respectively. Obtain the pan-tilt steering instructions, flight paths, and action instructions from the behavior instructions, control the pan-tilt steering according to the pan-tilt steering instructions, simultaneously control the UAV body to fly along the flight path, and control the actuators to perform the target actions according to the action instructions, so that each part of the embodied intelligent UAV can cooperate with each other to complete the target task.

[0111] For example, if the embodied intelligent UAV performs a building top inspection task, it needs to fly to the top of the building and take pictures. First, the embodied intelligent UAV takes off from the starting point. According to the behavior instructions, the UAV body flies towards the target location (the top of the building) along the preset flight path. The pan-tilt adjusts its attitude according to the pan-tilt steering instructions, so that the camera (i.e., the actuator) is aligned with the target device to be photographed (such as the solar panel on the top of the building). After reaching the target location, the UAV body hovers, and the pan-tilt further fine-tunes to ensure that the camera is accurately aligned with the target. Subsequently, the camera takes pictures (i.e., performs the target action).

[0112] After converting the inference result into behavior instructions in the embodiments of the present application, the pan-tilt steering, the flight path of the UAV body, and the execution of the target actions of the actuators are controlled according to the behavior instructions, which significantly improves the automation degree and accuracy of task completion.

[0113] In an alternative embodiment of the present application, the intelligent system includes a navigation module, and the control of the UAV body to fly along the flight path includes:

[0114] Obtain the navigation information of the navigation module;

[0115] Control the UAV body to fly along the flight path according to the navigation information.

[0116] In this embodiment, the intelligent system further includes a navigation module. During the flight of the UAV body, it is necessary to obtain the navigation information of the navigation module in real time, and control the UAV body to fly along the flight path according to the navigation information.

[0117] In one example, if the embodied intelligent UAV is performing an inspection task, the navigation module can obtain GPS or RTK positioning data in real time to determine the current position of the embodied intelligent UAV. If the embodied intelligent UAV deviates from the preset flight path, the intelligent system can dynamically adjust the flight direction according to the navigation information to ensure that it flies towards the target location along the planned flight path.

[0118] In the embodiments of the present application, the navigation module is used to obtain navigation information in real time and dynamically adjust the flight direction, ensuring that the embodied intelligent drone accurately flies along the preset path, improving the accuracy and reliability of task execution, and enhancing the adaptability of the drone in complex environments.

[0119] In an alternative embodiment of the present application, an intelligent reasoning model is deployed in the decision-making module. Using the decision-making module to perform reasoning on the current reasoning task to obtain a reasoning result, including:

[0120] Using the intelligent reasoning model to perform reasoning on the current reasoning task to obtain a reasoning result;

[0121] Uploading the reasoning result to the intelligent platform;

[0122] After the intelligent platform updates the intelligent reasoning model according to the reasoning result, obtaining the updated intelligent reasoning model;

[0123] Deploying the updated intelligent reasoning model in the decision-making module.

[0124] In this embodiment, an intelligent reasoning model is deployed in the decision-making module. The intelligent reasoning model is used to perform reasoning on the current reasoning task, and the intelligent reasoning model can be updated to the intelligent system after being trained on the intelligent platform. The algorithms for training the model can be algorithms such as deep learning and reinforcement learning. The intelligent reasoning model can include models such as an image recognition model.

[0125] Specifically, after the intelligent system uses the intelligent reasoning model to perform reasoning on the current reasoning task and obtains a reasoning result, the reasoning result is uploaded to the intelligent platform for updating the intelligent physical model. After the intelligent platform updates the intelligent reasoning model according to the reasoning result, the updated intelligent reasoning model is synchronized to the intelligent system, and the updated intelligent reasoning model is deployed in the decision-making module. In one embodiment, the reasoning result data of the intelligent system can be synchronized to the intelligent platform in real time, and the intelligent platform can regularly synchronize the updated intelligent reasoning model to the intelligent system.

[0126] Through the real-time update and synchronization of the intelligent reasoning model in the embodiments of the present application, the intelligent system of the embodied intelligent drone can continuously optimize its decision-making ability to adapt to complex task requirements. At the same time, by using the centralized training and update mechanism of the intelligent platform, the accuracy of the model and the intelligent level of the system are improved.

[0127] In an alternative embodiment of the present application, the method further includes:

[0128] Collecting target image data and target point cloud data according to the target task;

[0129] The intelligent system is used to perform image recognition on the target image data and preprocess the target point cloud data to obtain an image recognition result and preprocessed target point cloud data;

[0130] The image recognition result and the preprocessed target point cloud data are uploaded to the intelligent platform.

[0131] In this embodiment, the target image data collected by the camera and the target point cloud data collected by the millimeter-wave radar / lidar can be preprocessed and preliminarily analyzed, so as to reduce the amount of data uploaded to the intelligent platform and solve the problem of low data processing and analysis efficiency of traditional unmanned aerial vehicle systems. Specifically, the embodied intelligent unmanned aerial vehicle collects target image data using a camera and collects target point cloud data using a millimeter-wave radar / lidar according to the target task. The image recognition model in the intelligent system is used to perform image recognition on the target image data, and at the same time, the corresponding preprocessing model is used to preprocess the target point cloud data, which may include preprocessing models such as downsampling, denoising, normalization, and feature extraction. After obtaining the image recognition result and the preprocessed target point cloud data, the image recognition result and the preprocessed target point cloud data are directly uploaded to the intelligent platform. In one embodiment, not only the target image data and target point cloud data, but also other types of data can be preprocessed and preliminarily analyzed in the intelligent system before being uploaded to the intelligent platform.

[0132] In the embodiment of the present application, by first preprocessing and preliminarily analyzing the collected target image data and target point cloud data, the amount of data uploaded to the intelligent platform is reduced, and the problem of low data processing and analysis efficiency of traditional unmanned aerial vehicle systems is solved.

[0133] To make the intelligent unmanned aerial vehicle and the specific implementation method in the control method of an embodied intelligent unmanned aerial vehicle provided by the embodiment of the present application clearer, the following will Figures 3 - 6 be explained in detail.

[0134] Referring to Figure 3 , it is a schematic diagram of the hardware module of an embodiment of the control method of an embodied intelligent unmanned aerial vehicle of the present application.

[0135] Figure 3 The hardware module structure of the embodied intelligent unmanned aerial vehicle in the embodiment of the present application is shown, including the main control of the unmanned aerial vehicle body (i.e., the core control system of the unmanned aerial vehicle body), the main control of the intelligent execution mechanism (i.e., the core control system of the execution mechanism), and the embodied intelligent X-BRAIN edge system. The main control of the unmanned aerial vehicle body and the main control of the intelligent execution mechanism are connected to the embodied intelligent X-BRAIN edge system through USB (Universal Serial Bus, universal serial bus), network port or other data transmission interfaces.

[0136] The embodied intelligence X-BRAIN edge system includes the X-BRAIN management unit as the core control unit of the intelligent system. The X-BRAIN management unit includes a decision-making module, as well as a 4G / 5G / 5G Redcap / satellite communication module, a GPS / RTK / Beidou module, a control module, a power module, and a gigabit switch.

[0137] Among them, the 4G / 5G / 5G Redcap / satellite communication module is communicatively connected to the X-BRAIN intelligent platform through an antenna, and is connected to the X-BRAIN management unit through a USB or other data transmission interface for data transmission; the GPS / RTK / Beidou module obtains navigation information, position information, etc. through an antenna, and is connected to the X-BRAIN management unit through a serial port or other data transmission interface for data transmission; the control module is connected to the X-BRAIN management unit through a USB or other data transmission interface for data transmission, converts the inference result into a control instruction, and controls the pan-tilt rotation; the power module powers the intelligent drone through an on-board power supply; the gigabit switch obtains information collected by sensors such as cameras / millimeter-wave radars / lidar through network ports, and is connected to the X-BRAIN management unit through network ports for data transmission.

[0138] Refer to Figure 4 , which is a schematic diagram of software modules of an embodiment of the control method for an embodied intelligence drone of the present application.

[0139] Figure 4 The software module structure of the embodied intelligence drone of the embodiment of the present application is shown. The embodied intelligence X-BRAIN edge system includes a communication module, a navigation module, an X-BRAIN main control, a perception module, a decision-making module, and a control module.

[0140] Among them, the communication module realizes a wireless communication link with the X-BRAIN intelligent platform by obtaining data from the 4G / 5G / 5G Redcap / satellite communication module; the navigation module provides navigation information, position information, etc. for the intelligent drone by obtaining data from the GPS / RTK / Beidou module; the X-BRAIN main control is the core control unit of the intelligent system; the perception module performs data fusion by obtaining information collected by sensors such as cameras / millimeter-wave radars / lidar; the decision-making module is responsible for task planning and strategy generation; the control module converts the decision result into a control instruction to control the pan-tilt rotation, the flight path of the drone body, and the execution action instructions of the intelligent mechanism.

[0141] Refer to Figure 5 , which is a system initialization flowchart of an embodiment of the control method for an embodied intelligence drone of the present application.

[0142] As Figure 5As shown in the figure, when an embodied intelligent drone needs to be used, the embodied intelligent drone including an intelligent system needs to be started first. After the system is powered on, the GPS / RTK / Beidou module, network interface, pan-tilt control module, and 4G / 5G / 5G Redcap / satellite communication module are initialized in sequence. Then, it connects to the X-BRAIN intelligent platform through the 4G / 5G / 5G Redcap / satellite communication module (i.e., the wireless module) and enters the waiting task mode, waiting to receive the target task.

[0143] If abnormal events such as traffic accidents, vehicle illegal parking, and spillage occur on the highway, the embodied intelligent drone receives the target task and flies to the scene from the air with a proprietary device according to the target task. The target task may include: after the embodied intelligent drone arrives at the accident scene, providing real-time on-site images and data; remote personnel guiding the vehicle to move to a safe area through the loudspeaker of the embodied intelligent drone to reduce the risk of traffic congestion and secondary accidents caused by traffic accidents; the embodied intelligent drone deploying warning devices and cones around the abnormal event to improve the safety and handling efficiency of the accident scene; for the spillage scattered on the highway, the inspection embodied intelligent drone can accurately identify and report the location of the spillage, instruct the remote embodied intelligent drone to arrive at the scene, warn the following vehicles and grab the spillage to a safe area to reduce traffic hazards, etc. The embodiments of the present application do not impose any restrictions on the specific task types or task contents.

[0144] Refer to Figure 6 , which is a schematic diagram of task execution for an embodiment of the control method of an embodied intelligent drone of the present application.

[0145] Taking the cone deployment task as an example, after an abnormal event occurs on the highway, the intelligent platform issues the cone deployment task to the embodied intelligent drone in the waiting task mode. After receiving the task, the embodied intelligent drone carries facilities such as cones to the specified location. During the flight, the embodied intelligent drone obtains environmental information through visual sensors and lidar, and the perception module generates an environmental model after fusing multi-source data. The decision-making module generates a behavior instruction according to the task requirements and the environmental model, and the control module converts the behavior instruction into a motor control signal to drive the embodied intelligent drone to complete the cone deployment task. As Figure 6 shown, Figure 6 is the scene where the embodied intelligent drone needs to deploy cones. The dotted line box is the safe operation range of the embodied intelligent drone, and the embodied intelligent drone needs to deploy cones within the safe operation range. After the task is completed, the embodied intelligent drone returns.

[0146] In the embodiments of the present application, a target task sent by an intelligent platform is received; during the process of controlling the flight of the UAV body according to the target task, the current environmental information collected by sensors is obtained; the current environmental information collected by different sensors is fused to obtain a target environmental model, and the target environmental model is used to describe the environment where the embodied intelligent UAV is currently located; a behavior instruction is generated according to the target environmental model and the target task; the UAV body and the execution mechanism are controlled according to the behavior instruction to execute the target task. The embodiments of the present application combine the intelligent system with the UAV body, establish a communication connection with the intelligent platform, and through the intelligent decision-making capabilities of the intelligent system and the intelligent platform, as well as the efficient coordination of the UAV body and the execution mechanism, significantly improve the efficiency of the UAV in executing the target task. The embodied intelligent UAV can also fuse the real-time environmental information collected by multiple sensors, enabling it to have the ability to cope with complex environments, thereby realizing the closed-loop control of real-time perception, decision-making, and execution of the embodied intelligent UAV in an unstructured environment.

[0147] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of the present application are not limited by the described action sequence, because according to the embodiments of the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present application.

[0148] Based on the above embodiments, the present embodiment further provides a control device for an embodied intelligent UAV, which is applied to electronic devices such as terminal devices and servers.

[0149] Referring to Figure 7 , a structural block diagram of an embodiment of a control device for an embodied intelligent UAV according to the present application is shown, which may specifically include the following modules:

[0150] A task receiving module 701, configured to receive the target task sent by the intelligent platform;

[0151] An information obtaining module 702, configured to obtain the current environmental information collected by the sensors during the process of controlling the flight of the UAV body according to the target task;

[0152] An information fusion module 703, configured to fuse the current environmental information collected by different sensors to obtain a target environmental model, and the target environmental model is used to describe the environment where the embodied intelligent UAV is currently located;

[0153] An instruction generating module 704, configured to generate a behavior instruction according to the target environmental model and the target task;

[0154] The task execution module 705 is configured to control the drone body and the execution mechanism to execute a target task according to the behavior instruction.

[0155] Optionally, the intelligent system includes a navigation module, and the task receiving module 701 includes:

[0156] An allocation information acquisition sub-module, configured to acquire the function information of the embodied intelligent drone and the position information of the drone body in the navigation module;

[0157] An allocation information upload sub-module, configured to upload the function information and the position information to the intelligent platform;

[0158] A task allocation sub-module, configured to receive the target task allocated by the intelligent platform according to the function information and the position information.

[0159] Optionally, the intelligent system includes a decision module, and the instruction generation module 704 includes:

[0160] An inference task determination sub-module, configured to determine a current inference task according to the target environment model and the target task;

[0161] A first instruction generation sub-module, configured to use the decision module to perform inference on the current inference task to obtain an inference result; generate the behavior instruction according to the inference result; or,

[0162] A second instruction generation sub-module, configured to, if an exception occurs when the decision module performs inference on the current inference task, transmit the current inference task to the intelligent platform; receive the inference result obtained by the intelligent platform performing inference on the current inference task; generate the behavior instruction according to the inference result.

[0163] Optionally, the embodied intelligent drone further includes a gimbal, and the task execution module 705 includes:

[0164] A task instruction acquisition sub-module, configured to acquire the gimbal steering instruction, the flight path, and the action instruction in the behavior instruction, and the action instruction has a corresponding target action;

[0165] A task instruction execution sub-module, configured to control the gimbal to turn according to the gimbal steering instruction; and control the drone body to fly according to the flight path; and control the execution mechanism to execute the target action according to the action instruction to complete the target task.

[0166] Optionally, the intelligent system includes a navigation module, and controlling the drone body to fly according to the flight path includes:

[0167] Obtain the navigation information of the navigation module;

[0168] Control the main body of the drone to fly along the flight path according to the navigation information.

[0169] Optionally, the decision-making module is deployed with an intelligent reasoning model, and the first instruction generation sub-module is further configured to:

[0170] Perform reasoning on the current reasoning task by using the intelligent reasoning model to obtain a reasoning result;

[0171] Upload the reasoning result to the intelligent platform;

[0172] After the intelligent platform updates the intelligent reasoning model according to the reasoning result, obtain the updated intelligent reasoning model;

[0173] Deploy the updated intelligent reasoning model in the decision-making module.

[0174] Optionally, the device further includes:

[0175] A to-be-processed data collection module, configured to collect target image data and target point cloud data according to the target task;

[0176] A data preprocessing module, configured to perform image recognition on the target image data and preprocess the target point cloud data by using the intelligent system to obtain an image recognition result and preprocessed target point cloud data;

[0177] A preprocessing result upload module, configured to upload the image recognition result and the preprocessed target point cloud data to the intelligent platform.

[0178] An embodiment of the present application further provides a non-volatile readable storage medium, in which one or more modules (programs) are stored, and when the one or more modules are applied to a device, the device can be caused to execute instructions (instructions) of each method step in the embodiment of the present application.

[0179] An embodiment of the present application provides one or more machine-readable media, on which instructions are stored, and when executed by one or more processors, cause an electronic device to execute one or more of the methods as described in the above embodiments. In the embodiments of the present application, the electronic device includes various types of devices such as a terminal device and a server (cluster).

[0180] Embodiments of the present disclosure can be implemented as a device configured with any suitable hardware, firmware, software, or any combination thereof, and the device may include electronic devices such as a terminal device and a server (cluster). Figure 8Exemplary device 800 that can be used to implement the various embodiments described in the present application is schematically shown.

[0181] For one embodiment, Figure 8 Exemplary device 800 is shown, which has one or more processors 802, a control module (chipset) 804 coupled to at least one of the (one or more) processors 802, a memory 806 coupled to the control module 804, a non-volatile memory (NVM) / storage device 808 coupled to the control module 804, one or more input / output devices 810 coupled to the control module 804, and a network interface 812 coupled to the control module 804.

[0182] Processor 802 may include one or more single-core or multi-core processors, and processor 802 may include any combination of general-purpose processors or dedicated processors (such as graphics processors, application processors, baseband processors, etc.). In some embodiments, device 800 can function as devices such as the terminal device, server (cluster), etc. described in the embodiments of the present application.

[0183] In some embodiments, device 800 may include one or more computer-readable media (such as memory 806 or NVM / storage device 808) having instructions 814, and one or more processors 802 combined with the one or more computer-readable media and configured to execute the instructions 814 to implement modules so as to perform the actions described in the present disclosure.

[0184] For one embodiment, control module 804 may include any suitable interface controller to provide any suitable interface to at least one of the (one or more) processors 802 and / or any suitable device or component communicating with control module 804.

[0185] Control module 804 may include a memory controller module to provide an interface to memory 806. The memory controller module can be a hardware module, a software module, and / or a firmware module.

[0186] Memory 806 can be used, for example, to load and store data and / or instructions 814 for device 800. For one embodiment, memory 806 may include any suitable volatile memory, such as suitable DRAM. In some embodiments, memory 806 may include double data rate type four synchronous dynamic random access memory (DDR4 SDRAM).

[0187] For one embodiment, control module 804 may include one or more input / output controllers to provide an interface to NVM / storage device 808 and the (one or more) input / output devices 810.

[0188] For example, the NVM / storage device 808 can be used to store data and / or instructions 814. The NVM / storage device 808 can include any suitable non-volatile memory (e.g., flash memory) and / or can include any suitable (one or more) non-volatile storage devices (e.g., one or more hard disk drives (HDDs), one or more compact disc (CD) drives, and / or one or more digital versatile disc (DVD) drives).

[0189] The NVM / storage device 808 can include storage resources that are physically part of a device on which the device 800 is mounted, or it can be accessed by the device without being part of the device. For example, the NVM / storage device 808 can be accessed via a network through the (one or more) input / output devices 810.

[0190] (One or more) input / output devices 810 can provide an interface for the device 800 to communicate with any other suitable device. The input / output devices 810 can include communication components, audio components, sensor components, etc. The network interface 812 can provide an interface for the device 800 to communicate through one or more networks. The device 800 can wirelessly communicate with one or more components of a wireless network according to any of one or more wireless network standards and / or protocols, such as accessing a communication standard-based wireless network, such as WiFi, 2G, 3G, 4G, 5G, etc., or a combination thereof for wireless communication.

[0191] For one embodiment, at least one of the (one or more) processors 802 can be logically packaged together with one or more controllers of the control module 804 (e.g., a memory controller module). For one embodiment, at least one of the (one or more) processors 802 can be logically packaged together with one or more controllers of the control module 804 to form a system-in-package (SiP). For one embodiment, at least one of the (one or more) processors 802 can be logically integrated with one or more controllers of the control module 804 on the same die. For one embodiment, at least one of the (one or more) processors 802 can be logically integrated with one or more controllers of the control module 804 on the same die to form a system-on-chip (SoC).

[0192] In various embodiments, the device 800 may be, but is not limited to: a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.) and other terminal devices. In various embodiments, the device 800 may have more or fewer components and / or a different architecture. For example, in some embodiments, the device 800 includes one or more cameras, a keyboard, a liquid crystal display (LCD) screen (including a touch screen display), a non-volatile memory port, multiple antennas, a graphics chip, an application specific integrated circuit (ASIC), and a speaker.

[0193] Among them, a main control chip can be used as a processor or a control module in the detection device, sensor data, location information, etc. are stored in a memory or an NVM / storage device, the sensor group can be used as an input / output device, and the communication interface can include a network interface.

[0194] For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiments.

[0195] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0196] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable control terminal devices of drones to generate a machine, so that the instructions executed by the processor of the computer or other programmable control terminal devices of drones generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0197] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable control terminal devices of drones to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0198] These computer program instructions can also be loaded onto a computer or the control terminal device of other programmable drones, so that a series of operation steps are executed on the computer or other programmable terminal devices to generate computer-implemented processing, thereby the instructions executed on the computer or other programmable terminal devices provide for implementing in the process Figure 1 a process or multiple processes and / or boxes Figure 1 steps for the functions specified in a box or multiple boxes.

[0199] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0200] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the element.

[0201] The above has introduced in detail a control method and device for an embodied intelligent drone, an electronic device and a storage medium provided by the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A control method for an embodied intelligent unmanned aerial vehicle, characterized in that, Applied to an embodied intelligent unmanned aerial vehicle (UAV), the embodied intelligent UAV includes an intelligent system, a UAV body, and an actuator. The intelligent system is communicatively connected to an intelligent platform and is connected to sensors. The method includes: Receiving a target task issued by the intelligent platform; During the process of controlling the flight of the UAV body according to the target task, acquiring the current environmental information collected by the sensors; Fusing the current environmental information collected by different sensors to obtain a target environmental model, which is used to describe the environment where the embodied intelligent UAV is currently located; Generating a behavior instruction according to the target environmental model and the target task; Controlling the UAV body and the actuator to execute the target task according to the behavior instruction.

2. The method according to claim 1, wherein The intelligent system includes a navigation module. The receiving of the target task issued by the intelligent platform includes: Acquiring the function information of the embodied intelligent UAV and the position information of the UAV body in the navigation module; Uploading the function information and the position information to the intelligent platform; Receiving the target task allocated by the intelligent platform according to the function information and the position information.

3. The method according to claim 1, characterized in that, The intelligent system includes a decision-making module. The generating of the behavior instruction according to the target environmental model and the target task includes: Determining the current reasoning task according to the target environmental model and the target task; Using the decision-making module to reason about the current reasoning task to obtain a reasoning result; Generating the behavior instruction according to the reasoning result; Or, If an exception occurs when the decision-making module reasons about the current reasoning task, transmitting the current reasoning task to the intelligent platform; Receiving the reasoning result obtained by the intelligent platform's reasoning about the current reasoning task; Generating the behavior instruction according to the reasoning result.

4. The method according to claim 3, wherein The embodied intelligent UAV further includes a gimbal. The controlling of the UAV body and the actuator to execute the target task according to the behavior instruction includes: Acquiring the gimbal steering instruction, flight path, and action instruction in the behavior instruction, and the action instruction has a corresponding target action; Controlling the gimbal to turn according to the gimbal steering instruction; and controlling the UAV body to fly along the flight path; and controlling the actuator to execute the target action according to the action instruction to complete the target task.

5. The method according to claim 4, wherein The intelligent system includes a navigation module. The controlling of the UAV body to fly along the flight path includes: Acquiring the navigation information of the navigation module; Controlling the UAV body to fly along the flight path according to the navigation information.

6. The method according to claim 3, wherein The decision-making module is deployed with an intelligent reasoning model. The using of the decision-making module to reason about the current reasoning task to obtain a reasoning result includes: Using the intelligent reasoning model to reason about the current reasoning task to obtain a reasoning result; Uploading the reasoning result to the intelligent platform; After the intelligent platform updates the intelligent reasoning model according to the reasoning result, acquiring the updated intelligent reasoning model; Deploying the updated intelligent reasoning model in the decision-making module.

7. The method according to claim 1, wherein The method further includes: Collecting target image data and target point cloud data according to the target task; Using the intelligent system to perform image recognition on the target image data and preprocess the target point cloud data to obtain an image recognition result and preprocessed target point cloud data; Uploading the image recognition result and the preprocessed target point cloud data to the intelligent platform.

8. A control device for an embodied intelligent unmanned aerial vehicle, characterized in that, Applied to an embodied intelligent drone, the embodied intelligent drone includes an intelligent system, a drone body, and an actuator. The intelligent system is communicatively connected to the intelligent platform, and the intelligent system is connected to sensors. The device includes: A task receiving module, configured to receive a target task issued by the intelligent platform; An information acquisition module, configured to acquire current environment information collected by the sensors during the process of controlling the flight of the drone body according to the target task; An information fusion module, configured to fuse the current environment information collected by different sensors to obtain a target environment model, where the target environment model is used to describe the environment where the embodied intelligent drone is currently located; An instruction generation module, configured to generate a behavior instruction according to the target environment model and the target task; A task execution module, configured to control the drone body and the actuator to execute the target task according to the behavior instruction.

9. An electronic device, characterized in that, Comprising: A processor; And A memory, on which executable code is stored. When the executable code is executed, the processor is caused to execute the control method of the embodied intelligent drone according to one or more of claims 1-7.

10. One or more machine-readable media, on which executable code is stored. When the executable code is executed, the processor is caused to execute the control method of the embodied intelligent drone according to one or more of claims 1-7.

Images