Inspection path determination method and device, equipment, storage medium and program product

Abstract

The invention discloses an inspection path determination method, device and equipment, a storage medium and a program product, relates to the technical field of pipeline detection, and is used for effectively controlling flight inspection operation of an unmanned aerial vehicle and realizing effective detection of an oil and gas long-distance pipeline. The method comprises the following steps: acquiring pipeline environment information and pipeline image information collected by an unmanned aerial vehicle for a to-be-inspected pipeline area; based on the pipeline image information, determining a threat identification result of the to-be-inspected pipeline area; the threat identification result is used for indicating whether a threat object threatening normal operation of the pipeline exists in the to-be-inspected pipeline area; according to the pipeline environment information and the threat identification result, determining an inspection path of the unmanned aerial vehicle for the to-be-inspected pipeline area; the inspection path comprises a pre-specified path or a current planned path; the current planned path is used for enabling the unmanned aerial vehicle to avoid obstacles in the environment and / or approach a threat object.

Description

Technical Field

[0001] The present application relates to the technical field of pipeline detection, and in particular, to a method, device, equipment, storage medium, and program product for determining an inspection path. Background Art

[0002] At present, drone inspection has gradually become one of the mainstream methods for inspecting long-distance oil and gas pipelines. However, long-distance oil and gas pipelines span a wide area and have a complex deployment environment, which is likely to affect the flight inspection operation of drones, making it difficult to effectively detect long-distance oil and gas pipelines. Summary of the Invention

[0003] The purpose of the present application is to provide a method, device, equipment, storage medium, and program product for determining an inspection path, aiming to effectively control the flight inspection operation of drones and achieve effective detection of long-distance oil and gas pipelines.

[0004] To achieve the above objective, the present application adopts the following technical solutions: In a first aspect, an embodiment of the present application provides a method for determining an inspection path, including: obtaining pipeline environment information and pipeline image information collected by a drone for a pipeline area to be inspected. Based on the pipeline image information, determining a threat recognition result for the pipeline area to be inspected. The threat recognition result is used to indicate whether there is a threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected. According to the pipeline environment information and the threat recognition result, determining the inspection path of the drone for the pipeline area to be inspected. The inspection path includes a pre-specified path or a currently planned path. The currently planned path is used to enable the drone to avoid obstacles in the environment and / or approach the threat object.

[0005] Based on this, the present application can accurately identify the information (such as obstacles or threat objects) that need to be concerned when inspecting the pipeline area to be inspected by combining the pipeline environment information and the pipeline image information, so as to accurately determine the inspection path of the drone, enabling the drone to avoid obstacles in the environment and / or approach the threat object during the inspection process, adapting to the complex and changeable pipeline environment, and improving the inspection efficiency.

[0006] In some embodiments, according to the pipeline environment information and the threat recognition result, determining the inspection path of the unmanned aerial vehicle (UAV) for the pipeline area to be inspected includes: determining the flight mode of the UAV according to the pipeline environment information. The flight mode includes a normal mode, an anti-interference mode, or a no-fly mode. The normal mode is a mode of inspecting according to a pre-specified path. The anti-interference mode is a mode of inspecting according to the currently planned path. The no-fly mode is a mode that prohibits inspection. If the flight mode is the normal mode and there are no threat objects in the pipeline area to be inspected, the pre-specified path is determined as the inspection path. If the flight mode is the anti-interference mode and there are no threat objects in the pipeline area to be inspected, based on the pipeline environment information, the currently planned path is determined, and the currently planned path is determined as the inspection path. If the flight mode is not the no-fly mode and there are threat objects in the pipeline area to be inspected, based on the pipeline environment information and the pipeline image information, the currently planned path is determined, and the currently planned path is determined as the inspection path.

[0007] In some embodiments, based on the pipeline environment information and the pipeline image information, determining the currently planned path includes: obtaining the image position of the threat object in the pipeline image and the pose information of the UAV when collecting the pipeline image information. According to the image position and the pose information, determining the geographical location of the threat object. Based on the pipeline environment information and the geographical location, determining the currently planned path.

[0008] In some embodiments, based on the pipeline environment information and the geographical location, determining the currently planned path includes: determining the currently planned path according to the reinforcement learning algorithm, the operating state of the UAV, the pipeline environment information, and the geographical location.

[0009] In some embodiments, the pipeline environment information includes at least one of the following: the terrain information of the pipeline area to be inspected, the weather information, the height of the obstacle ahead, the size of the obstacle ahead, and the current distance between the UAV and the object. The current distance between the UAV and the object is the distance of the UAV relative to the obstacle ahead.

[0010] In some embodiments, determining the flight mode of the UAV according to the pipeline environment information includes: determining the environmental complexity of the pipeline area to be inspected according to the pipeline environment information. If the environmental complexity is less than or equal to the first threshold, the flight mode is determined to be the normal mode. If the environmental complexity is greater than the first threshold and less than or equal to the second threshold, the flight mode is determined to be the anti-interference mode. If the environmental complexity is greater than the second threshold, the flight mode is determined to be the no-fly mode.

[0011] In some embodiments, determining the flight mode of the UAV according to the pipeline environment information includes: inputting the pipeline environment information and the current flight height of the UAV into the first environment recognition model for processing to obtain the flight mode.

[0012] In some embodiments, before obtaining the pipeline environment information and pipeline image information collected by the drone for the pipeline area to be inspected, the method further includes: obtaining the historical environment information of the pipeline area to be inspected. The historical environment information includes at least one of the following: terrain information, weather information, number of obstacles, minimum obstacle height, and minimum obstacle spacing. Input the historical environment information into the second environment recognition model for processing to obtain the initial flight mode of the drone.

[0013] In some embodiments, the method further includes: obtaining various inspection data collected by the drone. Different inspection data have different data formats. Perform denoising processing and fusion processing on the various inspection data to obtain processed data in the same format. Analyze the data in the processed data where the collection location is around the threat object to determine the abnormal recognition result. The abnormal recognition result is used to indicate whether there is an abnormal operation of the pipeline. When the abnormal recognition result indicates that there is an abnormal operation of the pipeline, an early warning message is output.

[0014] In a second aspect, a device for determining an inspection path is provided, including: an acquisition unit and a processing unit.

[0015] The acquisition unit is configured to obtain the pipeline environment information and pipeline image information collected by the drone for the pipeline area to be inspected.

[0016] The processing unit is configured to determine a threat recognition result for the pipeline area to be inspected based on the pipeline image information. The threat recognition result is used to indicate whether there is a threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected.

[0017] The processing unit is further configured to determine an inspection path of the drone for the pipeline area to be inspected according to the pipeline environment information and the threat recognition result. The inspection path includes a pre-specified path or a currently planned path. The currently planned path is used to enable the drone to avoid obstacles in the environment and / or approach the threat object.

[0018] In some embodiments, the processing unit is specifically configured to: determine the flight mode of the drone according to the pipeline environment information. The flight mode includes a normal mode, an anti-interference mode, or a no-fly mode. The normal mode is a mode of inspecting according to a pre-specified path. The anti-interference mode is a mode of inspecting according to the currently planned path. The no-fly mode is a mode of prohibiting inspection. If the flight mode is the normal mode and there is no threat object in the pipeline area to be inspected, the pre-specified path is determined as the inspection path. If the flight mode is the anti-interference mode and there is no threat object in the pipeline area to be inspected, the currently planned path is determined based on the pipeline environment information, and the currently planned path is determined as the inspection path. If the flight mode is not the no-fly mode and there is a threat object in the pipeline area to be inspected, the currently planned path is determined based on the pipeline environment information and the pipeline image information, and the currently planned path is determined as the inspection path.

[0019] In some embodiments, the processing unit is specifically configured to: obtain the image position of the threat object in the pipeline image and the pose information when the drone collects the pipeline image information. Determine the geographical location of the threat object according to the image position and the pose information. Determine the current planned path based on the pipeline environment information and the geographical location.

[0020] In some embodiments, the processing unit is specifically configured to: determine the current planned path according to the reinforcement learning algorithm, the operating state of the drone, the pipeline environment information, and the geographical location.

[0021] In some embodiments, the pipeline environment information includes at least one of the following: the terrain information of the pipeline area to be inspected, the weather information, the height of the obstacle ahead, the size of the obstacle ahead, and the current distance between the drone and the object. The current distance between the drone and the object is the distance between the drone and the obstacle ahead.

[0022] In some embodiments, the processing unit is specifically configured to: determine the environmental complexity of the pipeline area to be inspected according to the pipeline environment information. If the environmental complexity is less than or equal to the first threshold, determine the flight mode as the normal mode. If the environmental complexity is greater than the first threshold and less than or equal to the second threshold, determine the flight mode as the anti-interference mode. If the environmental complexity is greater than the second threshold, determine the flight mode as the no-fly mode.

[0023] In some embodiments, the processing unit is specifically configured to: input the pipeline environment information and the current flight altitude of the drone into the first environment recognition model for processing to obtain the flight mode.

[0024] In some embodiments, the acquisition unit is further configured to acquire the historical environment information of the pipeline area to be inspected. The historical environment information includes at least one of the following: terrain information, weather information, the number of obstacles, the minimum obstacle height, and the minimum obstacle distance. The processing unit is further configured to input the historical environment information into the second environment recognition model for processing to obtain the initial flight mode of the drone.

[0025] In some embodiments, the acquisition unit is further configured to acquire a variety of inspection data collected by the drone. Different inspection data have different data formats. The processing unit is further configured to perform denoising processing and fusion processing on the variety of inspection data to obtain the processed data in the same format. The processing unit is further configured to analyze the data in the processed data whose acquisition location is around the threat object to determine the abnormal recognition result. The abnormal recognition result is used to indicate whether there is an abnormal operation of the pipeline. The processing unit is further configured to output a warning message when the abnormal recognition result indicates that there is an abnormal operation of the pipeline.

[0026] In a third aspect, an embodiment of the present application provides a computer device, including: a processor, the processor is connected to a memory, the memory is used to store computer execution instructions, and the processor executes the computer execution instructions stored in the memory so that the computer device executes the inspection path determination method according to any one of the first aspect.

[0027] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing computer execution instructions. When the computer execution instructions run on a computer device, the computer device is caused to execute the inspection path determination method according to any one of the first aspect.

[0028] In a fifth aspect, an embodiment of the present application provides a computer program product, including computer execution instructions. When the computer execution instructions run on a computer device, the computer device is caused to execute the inspection path determination method according to any one of the first aspect.

[0029] It should be understood that the technical effects brought by any implementation manner in the second aspect to the fifth aspect can refer to the technical effects brought by the corresponding implementation manner in the first aspect, which will not be elaborated here. Description of the Drawings

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0031] Figure 1 It is a schematic structural diagram of an inspection path determination system provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of a computing device provided by an embodiment of the present application; Figure 3 It is a schematic flowchart of an inspection path determination method provided by an embodiment of the present application; Figure 4 It is a schematic diagram of an inspection path determination process provided by an embodiment of the present application; Figure 5 It is a schematic structural diagram of an inspection path determination device provided by an embodiment of the present application. Detailed Embodiments

[0032] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0033] The terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. Unless otherwise stated in the following description, the meaning of "a plurality" is two or more.

[0034] In the embodiments of the present application, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the existence of additional identical elements in the process, article or device comprising such element.

[0035] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0036] In the description of this specification, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0037] First, a brief introduction to the application scenarios involved in the present application will be given.

[0038] At present, drone patrol has gradually become one of the mainstream methods for patrolling oil and gas long-distance pipelines. When using drones for patrol, although the requirements for long-distance patrol of drones in oil and gas long-distance pipeline patrol can be met by setting up drone hangars, there are still many problems in the actual application of oil and gas long-distance pipeline patrol.

[0039] First, the areas spanned by oil and gas long-distance pipelines are wide, and the involved environments are complex and changeable. It is required that drones can overcome problems such as bad weather and signal interference during patrol, and achieve large-scale, continuous and stable flight patrol operations.

[0040] Secondly, due to the wide variety of ground features and complex scenarios around the long-distance oil and gas pipelines, there are often a large amount of interference information in the information such as videos and images collected during the drone inspection, resulting in the inability to accurately identify and locate the threat objects that endanger the pipeline operation.

[0041] Moreover, the massive heterogeneous data collected during the drone inspection brings huge pressure to storage, computing, and analysis. How to design a reasonable inspection flight strategy to achieve accurate attention to threat objects such as third-party occupation and excavation beside the long-distance oil and gas pipelines, improve the efficiency and accuracy of the path during the drone inspection, and establish a timely and efficient data processing mechanism are urgent problems to be solved during the drone inspection of long-distance oil and gas pipelines.

[0042] In a related method, the power line distance and light sensing recognition module can be trained with labeled data to recognize the distance and light changes of the transmission line, and the position and flight path of the drone can be adjusted according to the real-time collected environmental data and the distance of the transmission line, so as to further generate the optimal inspection route based on the environmental data and the position and flight path of the drone, and perform autonomous inspection according to the optimal inspection route during the inspection process. This method realizes the adaptive optimization of the drone autonomous inspection to a certain extent.

[0043] In another related method, the drone is equipped with an autonomous decision-making module, which dynamically adjusts the inspection path according to the real-time detection data and the prediction results of artificial intelligence (AI), focusing on problems such as obstacle avoidance and multi-aircraft cooperation faced in the inspection of urban sewage pipelines.

[0044] In another related method, the image information and sound information of the inspection area of the power system are designed to be collected in real time and transmitted to the ground control station, so that the ground control station uses the dynamic path planning algorithm to generate the inspection route based on the sound information and image information and send it to the drone.

[0045] In another related method, the flight site information can be established according to the actual environment of the port, the flight starting point and all task points are set, and then the improved A-star algorithm is used for path planning between the starting point and the task points. Secondly, the drone is positioned in real time to make the drone fly along the planned path. After that, local path planning is carried out for the new obstacles that appear during the flight of the drone.

[0046] When implementing the drone inspection of long-distance oil and gas pipelines, the similarities and differences with the above various related methods are as follows. The drone inspection of long-distance oil and gas pipelines also pays attention to the obstacle information along the way and needs to achieve dynamic obstacle avoidance. However, the drone inspection of long-distance oil and gas pipelines pays more attention to the influence of regional climate conditions, topography, strong winds, electromagnetic interference, etc. The light change and distance information are not the core concerns. Moreover, it is necessary to implement an adaptive inspection path planning model that is more suitable for the drone inspection of long-distance oil and gas pipelines.

[0047] Based on this, when using drones for inspection of long-distance oil and gas pipelines, it is necessary to combine the environmental information around the pipeline to design a reasonable inspection flight strategy to ensure the safety and efficiency of flight under the condition of minimizing the design logic. Moreover, it is necessary to design an effective drone inspection path planning model to support the implementation of inspection path planning under various different conditions, making the drone inspection path have the characteristics of high quality, high efficiency, and accuracy. Moreover, it is necessary to implement inspection path planning and threat object recognition at the edge of the drone, and be able to comprehensively consider the attention to threat objects when planning the inspection path.

[0048] To achieve the above objectives, the present application provides a method for determining an inspection path, which can accurately identify the information (such as obstacles or threat objects) that need to be concerned when inspecting the pipeline area to be inspected by combining the pipeline environment information and the pipeline image information, so as to accurately determine the inspection path of the drone, enabling the drone to avoid obstacles in the environment and / or approach threat objects during the inspection process to adapt to the complex and changeable pipeline environment and improve the inspection efficiency.

[0049] Next, a brief introduction to the implementation environment (implementation architecture) involved in the present application is given.

[0050] As Figure 1 shown, it is a schematic structural diagram of an inspection path determination system provided by an embodiment of the present application. The inspection path determination system may include a drone 101 and a computing device 102. Among them, the drone 101 and the computing device 102 are connected.

[0051] It should be noted that the numbers of the drone 101 and the computing device 102 included in the above inspection path determination system are only examples, and the embodiments of the present application do not limit this.

[0052] The unmanned aerial vehicle 101 can be configured with various types of sensors, capable of supporting the collection of pipeline environment information, pipeline image information, pipeline inspection data, etc. For example, a lidar for collecting terrain information and obstacle information. Another example is a six-component meteorological instrument for collecting weather information. Another example is an optical imaging sensor such as a high-resolution visible light camera or a multispectral imager for collecting image information. Another example is a laser methane telemeter for detecting whether there is gas leakage in the pipeline. Another example is a high-precision inclination sensor for detecting whether the pipeline is deformed. The embodiments of the present application do not limit this.

[0053] The computing device 102 is used to plan the inspection path of the unmanned aerial vehicle for the pipeline area to be inspected based on the pipeline environment information and pipeline image information collected by the unmanned aerial vehicle for the pipeline area to be inspected. Moreover, the computing device 102 can process and analyze various inspection data collected by the unmanned aerial vehicle to determine whether there is an abnormal operation of the pipeline, and output a warning message when there is an abnormal operation of the pipeline.

[0054] Optionally, Figure 1 the computing device 102 in can be mounted on the unmanned aerial vehicle 101 and implemented as a functional module integrated within the unmanned aerial vehicle 101, thereby supporting real-time data processing at the edge of the unmanned aerial vehicle 101. In this way, by deploying an edge computing node at the unmanned aerial vehicle end, multi-source data collected by the unmanned aerial vehicle can be processed in real time, enabling efficient processing of determining the inspection mode of the unmanned aerial vehicle, identifying threat objects, dynamically adjusting the inspection path of the unmanned aerial vehicle, determining pipeline anomalies, etc. And when the data processing volume of a single edge computing node exceeds its computing capacity, a distributed computing mode can be started. The multi-source data is split to adjacent computing nodes for collaborative processing. Through the parallel computing of multiple computing nodes, the speed and efficiency of data processing are improved. The adjacent computing nodes can be computing devices carried by adjacent unmanned aerial vehicles.

[0055] Or, Figure 1 the computing device 102 in can also be a device independently set up from the unmanned aerial vehicle 101. The present application does not limit this.

[0056] Optionally, the computing device 102 can be a terminal or a server. The terminal can be a personal computer such as a desktop computer, a tablet computer, and a laptop computer, or can also be a remote terminal, a user terminal (terminal equipment, TE), and a mobile device, etc. The server can be a single server, or can also be a server cluster composed of multiple servers. The server cluster can also be called a computing device cluster. In some implementation manners, the server cluster can also be a distributed cluster. The present application does not limit the forms of the terminal and the server.

[0057] In terms of hardware implementation, the above-mentioned computing device can be through such as Figure 2Implemented by the structure shown. As Figure 2 shown, it is a schematic structural diagram of a computing device provided by an embodiment of the present application. Figure 2 The computing device shown may include: a processor 201, a memory 202, a communication interface 203, and a bus 204. The processor 201, the memory 202, and the communication interface 203 may be connected through the bus 204.

[0058] The processor 201 is the control center of the computing device, and may be a general-purpose central processing unit (CPU), or other general-purpose processors, etc. Among them, the general-purpose processor may be a microprocessor or any conventional processor, etc.

[0059] As an example, the processor 201 may include one or more CPUs, such as Figure 2 the CPU0 and CPU1 shown in

[0060] The memory 202 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a magnetic disk storage medium, or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0061] In a possible implementation, the memory 202 may exist independently of the processor 201. The memory 202 may be connected to the processor 201 through the bus 204, and is used to store data, instructions, or program code. When the processor 201 calls and executes the instructions or program code stored in the memory 202, it can implement the recognition of the object to be recognized.

[0062] In another possible implementation, the memory 202 may also be integrated with the processor 201.

[0063] A communication interface 203 is used for a computing device to connect with other devices via a communication network, which can be an Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc. The communication interface 203 may include a receiving unit for receiving data and a transmitting unit for transmitting data.

[0064] A bus 204 can be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity in representation, Figure 2 it is only represented by a thick line in the figure, but it does not mean that there is only one bus or one type of bus.

[0065] It should be noted that, Figure 2 the structure shown in the figure does not constitute a limitation on the computing device. Except Figure 2 for the components shown, the computing device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0066] For the sake of easy understanding, the inspection path determination method provided by the present application will be specifically introduced below in conjunction with the accompanying drawings.

[0067] As Figure 3 shown, it is a schematic flowchart of an inspection path determination method provided by an embodiment of the present application. Figure 3 The shown inspection path determination method can be applied to the computing device shown in the above Figures 1 to 2 figure, and the method includes: S301 - S303.

[0068] S301. Obtain the pipeline environment information and pipeline image information collected by the unmanned aerial vehicle (UAV) for the pipeline area to be inspected.

[0069] Among them, the pipeline area to be inspected can be the pipeline deployment area in front of the current inspection path of the UAV. The pipeline can be an oil and gas transmission pipeline.

[0070] The pipeline environment information includes at least one of the following: the terrain information of the pipeline area to be inspected, the weather information, the height of the obstacle in front, the size of the obstacle in front, and the current distance between the UAV and the obstacle. The current distance between the UAV and the obstacle is the distance between the UAV and the obstacle in front currently.

[0071] The pipeline image information refers to the images taken by the UAV of the pipeline area to be inspected in front of the flight path, which may include pipelines.

[0072] In one possible implementation, the UAV can start the inspection according to a pre-set inspection path. During the inspection according to the pre-set inspection path, the UAV can collect various pipeline environment information such as the terrain, climate, and obstacle data of the pipeline area to be inspected in real time through various sensors such as a radar rangefinder or a laser rangefinder, a visible light camera or an infrared camera, and a real-time weather observer.

[0073] S302. Based on the pipeline image information, determine the threat recognition result of the pipeline area to be inspected.

[0074] Among them, the threat recognition result is used to indicate whether there is a threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected. For example, the threat object can be a construction object such as a third-party construction machine or an excavator. Another example is that the threat object can be a heavy object that illegally occupies the soil above the pipeline.

[0075] In one possible implementation, the computing device can input the pipeline image information into an object detection model for processing to obtain the threat recognition result. The object detection model can be a detection model implemented based on a deep convolutional neural network, and the model parameters it has can be pre-trained. Or further, the object detection model can also identify the image position of the threat object in the pipeline image information.

[0076] S303. According to the pipeline environment information and the threat recognition result, determine the inspection path of the UAV for the pipeline area to be inspected.

[0077] Among them, the inspection path includes a pre-specified path or a current planned path. The pre-specified path refers to the UAV inspection path manually edited in advance by the staff according to the deployment position of the long-distance oil and gas pipeline in the pipeline area to be inspected and the distribution of obstacles. The current planned path refers to the UAV inspection path planned in real time by the computing device based on the pipeline environment information collected by the UAV and the identified threat recognition result.

[0078] For example, the computing device can determine the environmental complexity of the pipeline area to be inspected based on the pipeline environment information collected by the UAV. The environmental complexity is a comprehensive index used to reflect the complexity of the pipeline surrounding environment, including factors such as the undulation of the terrain, the density of obstacles, and the severity of the climate.

[0079] If the environmental complexity of the pipeline area to be inspected is less than or equal to the set complexity threshold and the threat recognition result indicates that there is no threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected, the computing device can determine the pre-specified path as the inspection path of the UAV for the pipeline area to be inspected.

[0080] If the environmental complexity of the pipeline area to be inspected is greater than the set complexity threshold, it can indicate that the current environment may have an adverse impact on the flight safety or inspection effect of the UAV. Then, the computing device can generate the current planned path based on the pipeline environment information collected by the UAV, enabling the UAV to avoid obstacles in the environment and achieve stable inspection operations.

[0081] Alternatively, if the threat recognition result indicates that there is a threat object in the pipeline area to be inspected that threatens the normal operation of the pipeline, the computing device can generate the current planned path based on the location of the threat object, enabling the UAV to approach the threat object to collect more accurate inspection data, thereby facilitating the identification of pipeline anomalies.

[0082] Alternatively, if the environmental complexity of the pipeline area to be inspected is greater than the set complexity threshold and the threat recognition result indicates that there is a threat object in the pipeline area to be inspected that threatens the normal operation of the pipeline, the computing device can generate the current planned path based on the pipeline environment information collected by the UAV and the location of the threat object, enabling the UAV to avoid obstacles in the environment and approach the threat object, achieving stable inspection operations and collecting more accurate inspection data, thereby facilitating the identification of pipeline anomalies.

[0083] In one embodiment, in S303 above, that is, when the computing device determines the inspection path of the UAV for the pipeline area to be inspected according to the pipeline environment information and the threat recognition result, the present application embodiment provides an optional implementation manner, including: S3031 - S3034.

[0084] S3031. Determine the flight mode of the UAV according to the pipeline environment information.

[0085] Among them, the flight mode includes a normal mode, an anti-interference mode, or a no-fly mode.

[0086] The normal mode is a mode of inspecting according to a pre-specified path. When the UAV flies according to the pre-specified path, the height, speed, and attitude are all pre-set, thus avoiding the problem of low efficiency caused by multiple meaningless dynamic flight path planning calculations when the UAV flies in a simple environment.

[0087] The anti-interference mode is a mode of inspecting according to the current planned path. In this mode, the computing device can automatically adjust parameters such as the flight height, speed, and attitude of the UAV according to the path planning algorithm and the current pipeline environment information, enabling the UAV to maintain a stable flight state in a complex environment and maximizing the inspection effect.

[0088] The no-fly mode is a mode of prohibiting inspection. In this mode, the UAV can immediately return or land at a nearby hangar to avoid collisions or other anomalies with obstacles.

[0089] S3032. If the flight mode is the normal mode and there are no threat objects in the pipeline area to be inspected, the pre-specified path is determined as the inspection path.

[0090] S3033. If the flight mode is the anti-interference mode and there are no threat objects in the pipeline area to be inspected, based on the pipeline environment information, the current planned path is determined and the current planned path is determined as the inspection path.

[0091] S3034. If the flight mode is not the no-fly mode and there are threat objects in the pipeline area to be inspected, based on the pipeline environment information and the pipeline image information, the current planned path is determined and the current planned path is determined as the inspection path.

[0092] Based on this, the computing device can flexibly determine the inspection path of the UAV based on the current flight mode of the UAV and whether there are threat objects in the pipeline area to be inspected, so that the UAV can adapt to the complex pipeline environment and thus perform the inspection task safely and effectively.

[0093] In one embodiment, in the above S3034, that is, when the computing device determines the current planned path based on the pipeline environment information and the pipeline image information, the embodiment of the present application provides an optional implementation method, including: S30341 - S30343.

[0094] S30341. Obtain the image position of the threat object in the pipeline image and the pose information of the UAV when collecting the pipeline image information.

[0095] The image position of the threat object in the pipeline image can be the pixel coordinates of the threat object in the pipeline image. The pose information of the UAV when collecting the pipeline image information can include the position information and orientation information of the UAV, etc.

[0096] For example, the computing device can process the pipeline image information based on the object detection model to identify the image position of the threat object in the pipeline image. And the computing device can determine the pose information of the UAV when collecting the pipeline image information based on the position and orientation system (POS) information of the UAV when collecting the pipeline image information.

[0097] S30342. Determine the geographical location of the threat object according to the image position and the pose information.

[0098] For example, the computing device determines the absolute position in the global geographical coordinate system based on the pose information of the UAV, and performs coordinate transformation on the image position of the threat object to obtain the geographical location of the threat object in the global geographical coordinate system.

[0099] S30343. Determine the current planned path based on pipeline environment information and geographical location.

[0100] For example, the computing device can use the pipeline environment information to establish the three-dimensional terrain information of the pipeline area to be inspected. Among them, the three-dimensional terrain information of the pipeline area to be inspected can mark the positions of obstacles and threat objects, etc. Furthermore, the computing device can generate the current planned path based on the three-dimensional terrain information of the pipeline area to be inspected, so that when the drone flies along the current planned path, it can avoid obstacles in the environment and approach the threat object, realizing stable inspection operations and collecting more accurate inspection data, thereby facilitating the identification of pipeline anomalies.

[0101] In one embodiment, in the above S30343, that is, when the computing device determines the current planned path based on the pipeline environment information and geographical location, the embodiment of the present application provides an optional implementation method, including: Step A.

[0102] Step A. Determine the current planned path according to the reinforcement learning algorithm, the operating state of the drone, the pipeline environment information, and the geographical location.

[0103] Reinforcement learning takes actions in the environment to maximize the cumulative reward. The state-action function of reinforcement learning, denoted as Q π (s,a), represents the expected return that can be obtained by performing action a in state s and then performing subsequent actions according to the path policy π. Using the Bellman equation for decomposition, the calculation formula is as follows: .

[0104] Among them, s is the current state information, including the operating state of the drone, the pipeline environment information, and the geographical location of the threat object at the current moment. s' represents the next state information, including the operating state of the drone, the pipeline environment information, and the geographical location of the threat object at the next moment. a is the action of the drone in state s. a' is the action of the drone in state s'. D is the action space, including forward, backward, left, right, up, down, accelerate, decelerate, maintain altitude, hover. R(s,a,s') is the reward after performing the action, γ represents the decay coefficient of future rewards, π(a'|s') is the probability distribution of selecting action a' in state s', and P(s'|s,a) represents the state transition probability. Q π (s',a') represents the expected return that can be obtained by performing action a' in state s' and then performing subsequent actions according to the path policy π.

[0105] The operating state of the drone refers to the current position of the drone, the speed and direction of the drone, and the battery state of the drone, etc. The pipeline environment information is the position, size, and shape of obstacles, wind speed and direction, pipeline position and orientation in the pipeline area to be inspected.

[0106] The action space is in discrete form, where moving forward, backward, left, and right are directional movements, accelerating and decelerating are speed adjustments, ascending, descending, and maintaining altitude are altitude adjustments, and hovering is to simplify the comprehensive maintenance of actions.

[0107] R(s,a,s') is the reward after executing an action. For example, if the drone is close to a threat object and far from an obstacle, a positive reward is given; if the drone is far from a threat object and close to an obstacle, the reward is negative. Based on this, the best path can be guided to sequentially determine the highest-value action that the drone can execute at each next step from the action space, generate the current planned path, and achieve a balance in dimensions such as path safety, drone flight efficiency, and attention to threat objects.

[0108] Alternatively, flight constraints of the drone can be set, such as flying above a pipeline, not colliding with obstacles, maximum flight speed, maximum acceleration, and turning ability.

[0109] Optionally, the reinforcement learning algorithm can be implemented using a double deep Q network (DDQN), or a dueling network (DN), or a dueling double deep Q network (D3QN), or other algorithms. The embodiments of the present application do not limit this.

[0110] In one embodiment, the pipeline environment information includes at least one of the following: terrain information of the pipeline area to be inspected, weather information, height of the obstacle ahead, size of the obstacle ahead, and current distance between the drone and the object. The current distance between the drone and the object is the distance between the drone and the obstacle ahead.

[0111] In S3031 above, that is, when the computing device determines the flight mode of the drone according to the pipeline environment information, an optional implementation manner provided by the embodiments of the present application includes: S3031 - S3034.

[0112] S3031: Determine the environmental complexity of the pipeline area to be inspected according to the pipeline environment information.

[0113] For example, the computing device can assign values to each parameter in the pipeline environment information according to a preset scoring rule. Further, the computing device can use the weighted scoring method to sum the values of each parameter in the pipeline environment information to obtain the environmental complexity of the pipeline area to be inspected.

[0114] S3032: If the environmental complexity is less than or equal to the first threshold, determine the flight mode as the normal mode.

[0115] S3033. If the environmental complexity is greater than the first threshold and less than or equal to the second threshold, determine the flight mode as the anti-interference mode.

[0116] S3034. If the environmental complexity is greater than the second threshold, determine the flight mode as the no-fly mode.

[0117] Among them, the first threshold is less than the second threshold. The first threshold and the second threshold can be reasonably set manually according to experience.

[0118] In the above S3031, that is, when the computing device determines the flight mode of the drone according to the pipeline environment information, another optional implementation manner provided by the embodiment of the present application includes: S3035.

[0119] S3035. Input the pipeline environment information and the current flight altitude of the drone into the first environment recognition model for processing to obtain the flight mode.

[0120] Among them, the input values of the first environment recognition model are the terrain information, wind information, rainfall probability, obstacle height, obstacle size, current distance between the aircraft and objects, and current flight altitude of the pipeline area to be inspected ahead, and the output value is the predicted flight mode of the drone.

[0121] The first environment recognition model can be pre-trained based on the pipeline environment information collected during the historical inspection of the drone to support accurate prediction of the flight mode.

[0122] Optionally, the first environment recognition model can be implemented using a C4.5 decision tree model, or can also be implemented using a random forest or a gradient boosting tree. The embodiment of the present application does not limit this.

[0123] In one embodiment, before obtaining the pipeline environment information and pipeline image information collected by the drone for the pipeline area to be inspected, the inspection path determination method provided by the embodiment of the present application further includes: S401 - S402.

[0124] S401. Obtain the historical environment information of the pipeline area to be inspected.

[0125] Among them, the historical environment information includes at least one of the following: terrain information, weather information, number of obstacles, minimum obstacle height, minimum obstacle distance.

[0126] Before the drone executes the inspection task, the computing device can obtain the historical environment information of the pipeline area to be inspected collected historically, so as to preliminarily judge the environmental complexity of the pipeline area to be inspected in advance.

[0127] S402. Input the historical environment information into the second environment recognition model for processing to obtain the initial flight mode of the drone.

[0128] Among them, the input values of the second environment recognition model are the terrain information, wind information, rainfall probability, number of obstacles, minimum obstacle height, and minimum obstacle spacing collected historically, and the output value is the predicted initial flight mode of the drone.

[0129] The second environment recognition model can be pre-trained based on the pipeline environment information collected during the historical inspection of the drone to support accurate prediction of the flight mode.

[0130] Optionally, the second environment recognition model can be implemented using a C4.5 decision tree model, or can also be implemented using a random forest or gradient boosting tree. The embodiments of the present application do not limit this.

[0131] Based on this, in the case of facing a complex pipeline deployment environment, the computing device can accurately predict the flight mode of the drone at different time periods through the pipeline environment information collected historically and the pipeline environment information collected currently, support the drone to perform inspection tasks safely and effectively in a complex environment, and improve the flexibility and adaptability of drone inspection.

[0132] In one embodiment, the inspection path determination method provided by the embodiments of the present application further includes: S501 - S504.

[0133] S501. Obtain various inspection data collected by the drone.

[0134] Among them, different inspection data have different data formats.

[0135] During the inspection process of the drone, a large amount of inspection data can be collected based on various sensors carried. The formats of these inspection data are often different, including the environmental data around the pipeline, the flight parameters of the drone, the inspection video images, and the threat objects extracted through image recognition technology, etc., which are the basis for subsequent abnormal determination and early warning.

[0136] S502. Perform denoising processing and fusion processing on the various inspection data to obtain processed data in the same format.

[0137] Considering that the drone inspection process may be interfered by various noises such as environmental noise and equipment noise, and the various inspection data come from multiple different sensors and devices, resulting in differences in the format, unit, and accuracy of these inspection data. In order to support accurate identification of pipeline operation anomalies, the computing device can perform denoising processing and fusion processing on the various inspection data to obtain processed data in the same format.

[0138] For example, the computing device can use an adaptive filtering algorithm to perform denoising processing on the inspection data, and can integrate the various inspection data into a unified format through data fusion technology to obtain processed data with higher accuracy and reliability.

[0139] When the computing device adopts an adaptive filtering algorithm, it can automatically adjust the filter parameters according to the error between the input signal and the desired output signal, effectively removing the noise components in the inspection data and improving the signal-to-noise ratio of the data. Moreover, through data fusion technology, the computing device can integrate the data from the drone side, the ground station, and other relevant devices, remove the redundant information in the inspection data, associate and integrate the relevant information to form a complete, accurate, and uniformly formatted inspection data set, so as to obtain more comprehensive and accurate pipeline inspection data, effectively support the triggering of subsequent anomaly determination and early warning mechanisms, and provide strong guarantee for the safe operation of long-distance oil and gas pipelines.

[0140] S503. Analyze the data with the acquisition location around the threat object in the processed data to determine the anomaly recognition result.

[0141] Among them, the anomaly recognition result is used to indicate whether there is an abnormal operation of the pipeline.

[0142] For example, the computing device can select the data with the acquisition location around the threat object from the processed data based on the geographical location of the threat object, and input the selected data into the anomaly determination logic library to further match according to the preset rules, realize the analysis of the input data, and thus, based on the results of the matching and analysis, analyze multi-source data (such as pressure, temperature, flow rate, vibration, corrosion potential, etc.). Determine the anomaly recognition result and judge whether the pipeline is in an abnormal state.

[0143] Among them, the anomaly determination logic library can include matching rules for pipeline faults such as pressure, temperature, leakage, blockage, corrosion, stress concentration, etc.

[0144] For example, the anomaly matching rule for pipeline pressure can be that when the currently collected pipeline pressure is greater than the set pressure threshold, it is determined that the pipeline is abnormal.

[0145] Another example is that the anomaly matching rule for pipeline temperature can be that when the currently collected pipeline temperature is greater than the set temperature threshold, it is determined that the pipeline is abnormal.

[0146] Another example is that the anomaly matching rule for pipeline leakage can be that when the currently collected pipeline pressure is less than the set pressure threshold, the flow rate is less than the set flow rate threshold, and the acoustic wave detection is abnormal, it is determined that the pipeline is abnormal.

[0147] Another example is that the anomaly matching rule for pipeline leakage can be that when the laser methane telemeter detects a gas leakage point, it is determined that the pipeline is abnormal.

[0148] S504. Output a warning message when the anomaly recognition result indicates that there is an abnormal operation of the pipeline.

[0149] Among them, the warning information is used to prompt that there are abnormal operations in the pipeline and the location information of the abnormal pipeline.

[0150] Optionally, the computing device can output the warning information by means of real-time notification to relevant personnel via text messages, emails, phone calls, etc., or by displaying the abnormal location and relevant information on the monitoring screen so that relevant personnel can quickly understand and handle it.

[0151] Or further, the computing device can automatically generate an abnormal report, generate a processing task based on the content and location of the abnormal report to trigger a warning mechanism, and assign the processing task to maintenance personnel. Moreover, the computing device can also feedback the abnormal report to the dispatching system to optimize subsequent UAV inspection tasks.

[0152] In one embodiment, as Figure 4 shown, it is a schematic diagram of a patrol path determination process provided by an embodiment of the present application. Combining the descriptions in the above embodiments, before the UAV patrol inspection, the computing device can obtain historical environment information based on the methods of S401-S402, and process the historical environment information based on the second environment recognition model to determine the initial flight mode of the UAV as the normal mode, or the anti-interference mode, or the no-fly mode. Furthermore, during the UAV patrol inspection, the computing device can process the pipeline environment information currently collected by the UAV based on the method of S3035, and determine the current flight mode of the UAV as the normal mode, or the anti-interference mode, or the no-fly mode.

[0153] If it is determined that the current flight mode of the UAV is the normal mode, the computing device can pre-specify the path as the patrol path of the UAV.

[0154] If it is determined that the current flight mode of the UAV is the anti-interference mode, the computing device can dynamically generate the current planned path and determine the current planned path as the patrol path of the UAV.

[0155] If it is determined that the current flight mode of the UAV is the no-fly mode, the computing device can instruct the UAV to return immediately or land at a nearby hangar.

[0156] If it is determined that the current flight mode of the UAV is the normal mode or the anti-interference mode and it is determined that there is a threat object, the computing device can dynamically generate the current planned path and determine the current planned path as the patrol path of the UAV.

[0157] Furthermore, the computing device can obtain and process the patrol inspection data based on the methods in S501-S504 above to further realize abnormal recognition and warning.

[0158] In one embodiment, in a scenario where multiple drones form a drone cluster to achieve collaborative inspection, different drones can establish connections through a multi-drone collaborative communication protocol. Moreover, the computing device can dynamically adjust the inspection paths of each drone based on a reinforcement learning algorithm according to the pipeline environment information and inspection requirements (such as approaching a threat object) collected in real time, ensuring that each drone can efficiently complete the inspection task and avoiding overlapping of the inspection areas among different drones. Also, in the scenario of multiple drones conducting collaborative inspection, the current state information in the reinforcement learning algorithm described in the above embodiment may further include the positions of other drones and the areas already inspected by other drones, and constraints for reducing the number of task switches of the drones can be added simultaneously.

[0159] In the above embodiments of the present application, the computing device can dynamically adjust the flight path of the drone based on the fusion of multiple sensors carried by the drone, in combination with an environment recognition model, so that the drone can adapt to the complex and changeable pipeline environment, and perform data processing in combination with edge computing to improve the inspection efficiency and data processing ability. In this way, it is possible to support ensuring the safety and efficiency of the drone flight under the condition of minimizing the design logic, design an effective drone inspection path planning method, realize drone inspection under various different conditions, and dynamic path planning during the inspection of multiple drones, making the inspection path of the drone have characteristics such as optimization, efficiency, and accuracy. Combining logics such as threat object recognition, edge computing, pipeline anomaly recognition, and early warning can effectively support the intelligent inspection of oil and gas long-distance pipelines.

[0160] Moreover, in the above embodiments of the present application for the drone inspection of oil and gas long-distance pipelines, the environmental information around the pipeline is fully utilized, a dynamic path planning method is designed, ensuring the safety and efficiency of the drone flight, and enhancing the attention to pipeline threat objects. And the computing device can be deployed as an edge computing node on the drone, greatly reducing the data transmission pressure and improving the timeliness of dynamic path planning.

[0161] The above mainly introduced the solutions of the embodiments of the present application from the perspective of methods. It can be understood that in order for the computer device to implement the above functions, it includes at least one of the corresponding hardware structures and software modules for executing each function. Those skilled in the art should easily realize that, combining the units and algorithm steps of each example described in the embodiments disclosed in this article, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the way of hardware or computer software driving the hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0162] Embodiments of the present application can divide functional units of a computing device according to the above method examples. For example, each functional unit can be divided corresponding to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, merely a logical functional division, and there can be other division methods in actual implementation.

[0163] Exemplarily, as Figure 5 shown, it is a schematic structural diagram of an inspection path determination device provided by an embodiment of the present application. The inspection path determination device includes: an acquisition unit 601 and a processing unit 602.

[0164] The acquisition unit 601 is configured to acquire pipeline environment information and pipeline image information collected by a drone for a pipeline area to be inspected.

[0165] The processing unit 602 is configured to determine a threat recognition result for the pipeline area to be inspected based on the pipeline image information. The threat recognition result is used to indicate whether there is a threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected.

[0166] The processing unit 602 is further configured to determine an inspection path of the drone for the pipeline area to be inspected according to the pipeline environment information and the threat recognition result. The inspection path includes a pre-specified path or a currently planned path. The currently planned path is used to enable the drone to avoid obstacles in the environment and / or approach the threat object.

[0167] In some embodiments, the processing unit 602 is specifically configured to: determine the flight mode of the drone according to the pipeline environment information. The flight mode includes a normal mode, an anti-interference mode, or a no-fly mode. The normal mode is a mode of inspecting according to a pre-specified path. The anti-interference mode is a mode of inspecting according to the currently planned path. The no-fly mode is a mode of prohibiting inspection. If the flight mode is the normal mode and there is no threat object in the pipeline area to be inspected, the pre-specified path is determined as the inspection path. If the flight mode is the anti-interference mode and there is no threat object in the pipeline area to be inspected, the currently planned path is determined based on the pipeline environment information, and the currently planned path is determined as the inspection path. If the flight mode is not the no-fly mode and there is a threat object in the pipeline area to be inspected, the currently planned path is determined based on the pipeline environment information and the pipeline image information, and the currently planned path is determined as the inspection path.

[0168] In some embodiments, the processing unit 602 is specifically configured to: obtain the image position of the threat object in the pipeline image and the pose information when the drone collects the pipeline image information. Determine the geographical location of the threat object according to the image position and the pose information. Determine the current planned path based on the pipeline environment information and the geographical location.

[0169] In some embodiments, the processing unit 602 is specifically configured to: determine the current planned path according to the reinforcement learning algorithm, the operating state of the drone, the pipeline environment information, and the geographical location.

[0170] In some embodiments, the pipeline environment information includes at least one of the following: the terrain information of the pipeline area to be inspected, the weather information, the height of the obstacle ahead, the size of the obstacle ahead, and the current distance between the drone and the object. The current distance between the drone and the object is the distance between the drone and the obstacle ahead.

[0171] In some embodiments, the processing unit 602 is specifically configured to: determine the environmental complexity of the pipeline area to be inspected according to the pipeline environment information. If the environmental complexity is less than or equal to the first threshold, determine the flight mode as the normal mode. If the environmental complexity is greater than the first threshold and less than or equal to the second threshold, determine the flight mode as the anti-interference mode. If the environmental complexity is greater than the second threshold, determine the flight mode as the no-fly mode.

[0172] In some embodiments, the processing unit 602 is specifically configured to: input the pipeline environment information and the current flight height of the drone into the first environment recognition model for processing to obtain the flight mode.

[0173] In some embodiments, the acquisition unit 601 is further configured to obtain the historical environment information of the pipeline area to be inspected. The historical environment information includes at least one of the following: terrain information, weather information, the number of obstacles, the minimum obstacle height, and the minimum obstacle distance. The processing unit 602 is further configured to input the historical environment information into the second environment recognition model for processing to obtain the initial flight mode of the drone.

[0174] In some embodiments, the acquisition unit 601 is further configured to obtain a variety of inspection data collected by the drone. Different inspection data have different data formats. The processing unit 602 is further configured to perform denoising processing and fusion processing on the variety of inspection data to obtain the processed data in the same format. The processing unit 602 is further configured to analyze the data whose acquisition position is around the threat object in the processed data to determine the abnormal recognition result. The abnormal recognition result is used to indicate whether there is an abnormal operation of the pipeline. The processing unit 602 is further configured to output a warning message when the abnormal recognition result indicates that there is an abnormal operation of the pipeline.

[0175] For the specific description of the above optional manner, reference may be made to the foregoing method embodiments, which will not be elaborated herein. In addition, the explanations and beneficial effects of any of the above-provided computer devices can be referred to the corresponding method embodiments above, which will not be elaborated herein.

[0176] An embodiment of the present application further provides a readable storage medium, on which a computer program is stored. When the computer program runs on a computing device, the computing device is caused to execute any of the methods executed by the above-provided computing devices.

[0177] For the explanations and beneficial effects of the relevant content in any of the above-provided readable storage media, reference may be made to the corresponding embodiments above, which will not be elaborated herein.

[0178] An embodiment of the present application further provides a computer program product including instructions. When the instructions run on a computing device, the computing device is caused to execute any of the methods in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computing device, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computing device may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a readable storage medium or transmitted from one readable storage medium to another. For example, the computer instructions may be transmitted from a website, a computer, a server, or a data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The readable storage medium may be any available medium accessible by the computing device or a data storage device such as a server or a data center including one or more integrated media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), etc.

[0179] It should be noted that the above-mentioned devices for storing computer instructions or computer programs provided in the embodiments of the present application, such as but not limited to, the above-mentioned memory, readable storage medium, etc., are all non-transitory.

[0180] The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for determining an inspection path, characterized in that: include: Obtain pipeline environment information and pipeline image information collected by the drone for the pipeline area to be inspected; Based on the pipeline image information, determining a threat identification result of the pipeline area to be inspected; the threat identification result is used to indicate whether there is a threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected; Determining an inspection path of the drone for the pipeline area to be inspected according to the pipeline environment information and the threat identification result; The inspection path includes a pre-specified path or a currently planned path; The currently planned path is used to enable the drone to avoid obstacles in the environment and / or approach the threat object.

2. The method according to claim 1, characterized in that The step of determining the inspection path of the drone for the pipeline area to be inspected according to the pipeline environment information and the threat identification result includes: Determine the flight mode of the UAV according to the pipeline environment information; the flight mode includes a normal mode, an anti-interference mode or a no-fly mode; the normal mode is a mode for patrolling along a pre-specified path; the anti-interference mode is a mode for patrolling along the currently planned path; the no-fly mode is a mode for prohibiting patrolling; If the flight mode is a normal mode and the threat object does not exist in the pipeline area to be inspected, the pre-specified path is determined as the inspection path; If the flight mode is the anti-interference mode and the threat object does not exist in the pipeline area to be inspected, determine the current planned path based on the pipeline environment information, and determine the current planned path as the inspection path; If the flight mode is not the no-fly mode and the threat object exists in the pipeline area to be inspected, a current planned path is determined based on the pipeline environment information and the pipeline image information, and the current planned path is determined as the inspection path.

3. The method according to claim 2, characterized in that The determining the current planned path based on the pipeline environment information and the pipeline image information includes: Acquire the image position of the threat object in the pipeline image, and the position and posture information of the drone when collecting the pipeline image information; Determining the geographic location of the threat object according to the image position and the posture information; The currently planned path is determined based on the pipeline environment information and the geographical location.

4. The method according to claim 3, characterized in that The determining the current planned path based on the pipeline environment information and the geographical location includes: The currently planned path is determined according to the reinforcement learning algorithm, the operating status of the drone, the pipeline environment information and the geographical location.

5. The method according to claim 2, characterized in that: The pipeline environment information includes at least one of the following: terrain information, weather information, height of obstacles ahead, size of obstacles ahead, and current distance between the drone and the obstacle ahead of the pipeline area to be inspected; the current distance between the drone and the obstacle ahead is the current distance between the drone and the obstacle ahead.

6. The method according to claim 5, characterized in that The step of determining the flight mode of the UAV according to the pipeline environment information includes: Determining the environmental complexity of the pipeline area to be inspected according to the pipeline environmental information; If the environmental complexity is less than or equal to a first threshold, determining that the flight mode is a normal mode; If the environmental complexity is greater than a first threshold and less than or equal to a second threshold, determining that the flight mode is an anti-interference mode; If the environmental complexity is greater than a second threshold, the flight mode is determined to be a no-fly mode.

7. The method according to claim 5, characterized in that The step of determining the flight mode of the UAV according to the pipeline environment information includes: The pipeline environment information and the current flight altitude of the UAV are input into a first environment recognition model for processing to obtain the flight mode.

8. The method according to claim 1, characterized in that Before obtaining the pipeline environment information and pipeline image information collected by the drone for the pipeline area to be inspected, the method further includes: Acquire historical environmental information of the pipeline area to be inspected; the historical environmental information includes at least one of the following: terrain information, weather information, number of obstacles, minimum obstacle height, and minimum obstacle spacing; The historical environmental information is input into a second environmental recognition model for processing to obtain an initial flight mode of the UAV.

9. The method according to claim 1, characterized in that: The method further comprises: Acquire a variety of inspection data collected by the drone; different inspection data have different data formats; Performing denoising and fusion processing on the multiple inspection data to obtain processed data in the same format; Based on the data collected in the processed data and located around the threat object, an abnormality identification result is analyzed to determine the abnormality identification result; the abnormality identification result is used to indicate whether there is an operational abnormality in the pipeline; When the abnormality identification result indicates that the pipeline has an operational abnormality, an early warning message is output.

10. A patrol route determination device, characterized in that: include: Acquisition unit and processing unit; An acquisition unit, used to acquire pipeline environment information and pipeline image information collected by the drone for the pipeline area to be inspected; A processing unit, configured to determine a threat identification result of the pipeline area to be inspected based on the pipeline image information; the threat identification result is used to indicate whether there is a threat object that threatens the normal operation of the pipeline in the pipeline area to be inspected; The processing unit is further used to determine the inspection path of the drone for the pipeline area to be inspected according to the pipeline environment information and the threat identification result; The inspection path includes a pre-specified path or a currently planned path; The currently planned path is used to enable the drone to avoid obstacles in the environment and / or approach the threat object.

11. A computer device, characterized in that: include: A processor, wherein the processor is connected to a memory, the memory is used to store computer-executable instructions, and the processor executes the computer-executable instructions stored in the memory so that the computer device implements the inspection path determination method as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: Used to store computer execution instructions, when the computer execution instructions are executed on a computer device, the computer device implements the inspection path determination method according to any one of claims 1 to 9.

13. A computer program product, characterized in that It includes computer execution instructions, and when the computer execution instructions are executed on a computer device, the computer device implements the inspection path determination method according to any one of claims 1 to 9.

Images