Insulator cleaning method and device

Abstract

The invention relates to an insulator cleaning method and device. The method comprises the following steps: controlling a movable sensing device carried by a bearing platform to collect environment information of a current operation area; according to the environment information, pose data of sensing equipment corresponding to each insulator and water outlet parameters of cleaning equipment are determined; for each insulator, controlling the cleaning equipment to clean the insulator based on the corresponding water outlet parameter, and controlling the sensing equipment to carry out first obstacle detection on the insulator; under the condition that the sensing equipment detects the first obstacle, the sensing equipment and the cleaning equipment are controlled to move horizontally according to the position relation between the first obstacle and the insulator, and water outlet parameters corresponding to the insulator are adjusted according to the position relation between the horizontally-moved cleaning equipment and the insulator; and based on the adjusted water outlet parameters, returning to execute the operation of controlling the cleaning equipment to clean the insulator based on the corresponding water outlet parameters until it is detected that cleaning of the insulator is completed. By adopting the method, the insulator cleaning efficiency can be improved.

Description

Technical Field

[0001] This application relates to the field of railway catenary maintenance technology, and in particular to an insulator cleaning method and apparatus. Background Technology

[0002] Railway overhead contact line insulators operate outdoors for extended periods, which can lead to the accumulation of dirt and grime, resulting in decreased insulation performance, flashover accidents, and threats to railway operation safety.

[0003] Traditional cleaning methods mainly rely on manual operation using high-pressure water guns or cleaning vehicles at fixed points. However, this method is inefficient in complex cleaning environments such as high altitudes and areas with many obstacles due to its poor operational flexibility and limited operation. Summary of the Invention

[0004] Therefore, it is necessary to provide an insulator cleaning method and apparatus that can improve cleaning efficiency in response to the above-mentioned technical problems.

[0005] Firstly, this application provides an insulator cleaning method, comprising:

[0006] When the carrier platform travels to the scanning position corresponding to the current work area, the mobile sensing device carried by the carrier platform is controlled to collect environmental information of the current work area; the carrier platform is also carried with a mobile cleaning device, and the environmental information includes at least one insulator in the current work area;

[0007] Based on the environmental information, determine the pose data of the sensing device corresponding to each insulator and the water output parameters of the cleaning device;

[0008] For each insulator, the cleaning equipment is controlled to clean the insulator based on the corresponding water output parameters, and the sensing equipment is controlled to detect the first obstacle on the insulator.

[0009] When the sensing device detects the first obstacle, it controls the sensing device and the cleaning device to translate according to the positional relationship between the first obstacle and the insulator, and adjusts the water discharge parameters of the insulator according to the positional relationship between the cleaning device and the insulator after translation; wherein the detection area of ​​the sensing device and the water discharge flow area of ​​the cleaning device after translation both avoid the first obstacle.

[0010] Based on the adjusted effluent parameters, the system returns to control the cleaning equipment to perform cleaning operations on the insulators according to the corresponding effluent parameters until the insulator cleaning is detected as complete.

[0011] In one embodiment, the method further includes:

[0012] Based on environmental information, determine the cleaning sequence for at least one insulator;

[0013] For each insulator, the cleaning equipment is controlled to clean the insulator based on the corresponding effluent parameters, including:

[0014] According to the cleaning sequence, traverse at least one insulator;

[0015] When each insulator is visited, the cleaning equipment is controlled to clean the insulator based on the corresponding water discharge parameters.

[0016] In one embodiment, before controlling the cleaning equipment to clean the insulator based on corresponding effluent parameters, the method further includes:

[0017] If it is detected that the carrier platform is not located in the preset working area corresponding to the insulator, control the carrier platform to travel to the preset working area corresponding to the insulator;

[0018] The preset work area is the area within the current work area that ensures the cleaning equipment can clean the insulators, and each preset work area corresponds to at least one insulator.

[0019] In one embodiment, detecting that the insulator cleaning is complete includes:

[0020] The degree of dirtiness of the insulators is detected;

[0021] If the detected level of dirt meets the preset cleaning standard, the insulator cleaning is considered complete.

[0022] In one embodiment, the degree of contamination of the insulator is detected, including:

[0023] Control the sensing device to collect the current image of the insulator;

[0024] The current image is analyzed to determine the degree of dirtiness of the insulator.

[0025] In one embodiment, the cleaning equipment includes an image acquisition device; controlling the cleaning equipment to clean the insulators based on corresponding effluent parameters includes:

[0026] The cleaning equipment is controlled to emit water streams based on corresponding water output parameters; the water streams are used to clean the insulators.

[0027] The current image of the water flow captured by the image acquisition device is obtained, and the cleaning effect of the water flow on the insulator is determined based on the current image.

[0028] Adjust the water discharge parameters corresponding to the insulators based on the cleaning effect;

[0029] Based on the adjusted effluent parameters, return to the execution control cleaning equipment to launch water flow based on the corresponding effluent parameters.

[0030] In one embodiment, based on environmental information, the pose data of the sensing device corresponding to each insulator and the water discharge parameters of the cleaning device are determined, including:

[0031] The environmental information is fused with the railway catenary map associated with the current work area to obtain the target semantic map;

[0032] Based on the target semantic map, pose coordination planning is performed on the sensing device and the cleaning device to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to each insulator.

[0033] In one embodiment, the target semantic map includes at least one insulator, location information and dirt information of each insulator, a second obstacle in the current work area, and spatial contour information of the second obstacle.

[0034] Based on the target semantic map, pose collaborative planning is performed between the sensing device and the cleaning device to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device for each insulator, including:

[0035] For each insulator, based on the insulator's position information and the spatial contour information of the second obstacle, the initial pose data of the sensing device corresponding to the insulator is determined; and,

[0036] Based on the insulator's location information, the insulator's dirt information, and the spatial contour information of the second obstacle, determine the initial water discharge parameters of the cleaning equipment corresponding to the insulator.

[0037] Based on the initial pose data and initial water discharge parameters, pose coordination planning is performed between the sensing device and the cleaning device to obtain the pose data of the sensing device corresponding to the insulator and the water discharge parameters of the cleaning device.

[0038] In one embodiment, the environmental information includes environmental point cloud data and environmental image data; the environmental information is fused with a railway catenary map associated with the current work area to obtain a target semantic map, including:

[0039] The environmental point cloud data is standardized to obtain the target point cloud data;

[0040] Identify dirt information for each insulator and spatial contour information for second obstacles in environmental image data;

[0041] The target point cloud data, the spatial contour information of the first obstacle, and the dirt information of each insulator are fused to obtain an initial semantic map.

[0042] Based on the railway catenary map associated with the current work area, the location information in the initial semantic map is calibrated to obtain the target semantic map.

[0043] Secondly, this application also provides an insulator cleaning device, comprising:

[0044] The data acquisition module is used to control the mobile sensing device mounted on the carrier platform to collect environmental information of the current work area when the carrier platform travels to the scanning position corresponding to the current work area; the carrier platform is also equipped with mobile cleaning equipment, and the environmental information includes at least one insulator in the current work area;

[0045] The processing module is used to determine the pose data of the sensing device corresponding to each insulator and the water output parameters of the cleaning device based on the environmental information.

[0046] The cleaning module is used to control the cleaning equipment to clean the insulator based on the corresponding water output parameters for each insulator, and to control the sensing equipment to perform the first obstacle detection on the insulator.

[0047] The processing module is also used to control the translation of the sensing device and the cleaning device according to the positional relationship between the first obstacle and the insulator when the sensing device detects the first obstacle, and to adjust the water discharge parameters corresponding to the insulator according to the positional relationship between the translated cleaning device and the insulator; wherein the detection area of ​​the sensing device and the water discharge flow area of ​​the cleaning device after translation both avoid the first obstacle.

[0048] The cleaning module is also used to return the control of the cleaning equipment to perform cleaning operations on the insulators based on the adjusted outlet water parameters, until the cleaning of the insulators is detected to be complete.

[0049] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described insulator cleaning method.

[0050] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described insulator cleaning method.

[0051] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described insulator cleaning method.

[0052] The aforementioned insulator cleaning method and apparatus utilize a platform carrying movable sensing and cleaning equipment. First, environmental information of the work area is collected at the scanning location, and the corresponding pose data and water discharge parameters for each insulator are planned. During the insulator cleaning process, the sensing equipment performs real-time obstacle detection. When an obstacle is detected, the sensing and cleaning equipment are synchronously adjusted according to the positional relationship between the obstacle and the insulator, ensuring that both the detection area and the water flow area avoid the obstacle. The water discharge parameters are also adaptively adjusted before cleaning continues until completion. This method enables autonomous obstacle avoidance and cleaning in complex working environments, reducing the risks of working at heights. Furthermore, dynamic obstacle avoidance and adaptive adjustment of water discharge parameters mitigate cleaning interruptions caused by obstacles, thereby improving the efficiency of insulator cleaning in complex environments. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a side view of the overall system structure of the insulator cleaning method in some embodiments of this application;

[0055] Figure 2 This is a top view schematic diagram of the overall system structure of the insulator cleaning method in some embodiments of this application;

[0056] Figure 3 This is a flowchart illustrating the insulator cleaning method in some embodiments of this application;

[0057] Figure 4 This is a flowchart illustrating the insulator cleaning method in some embodiments of this application;

[0058] Figure 5 This is a flowchart illustrating the insulator cleaning method in some other embodiments of this application;

[0059] Figure 6 This is a flowchart illustrating the insulator cleaning method in some embodiments of this application;

[0060] Figure 7 This is a schematic diagram of the collaborative planning process for the insulator cleaning method in some embodiments of this application;

[0061] Figure 8 This is a flowchart illustrating the collaborative operation of the insulator cleaning method in a complex station environment in some embodiments of this application;

[0062] Figure 9 This is a schematic diagram of a rinsing scenario for an insulator cleaning method in some embodiments of this application;

[0063] Figure 10 This is a structural block diagram of the insulator cleaning device in some embodiments of this application;

[0064] Figure 11 This is a diagram showing the internal structure of a computer device in some embodiments of this application. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0066] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments or any combination of multiple embodiments.

[0067] Railway contact network insulators operate outdoors for extended periods, easily accumulating dirt and grime, leading to decreased insulation performance, flashover accidents, and threatening railway operational safety. Traditional cleaning methods primarily rely on manual operation using high-pressure water guns or cleaning vehicles at fixed points, resulting in low efficiency, poor safety, high water consumption, and limitations imposed by maintenance windows. In recent years, automated cleaning solutions based on vehicle-mounted robotic arms or fixed water cannons have emerged. Vehicle-mounted robotic arm automated cleaning equipment is a mechanical contact cleaning method, generally operating at fixed points with a small cleaning area; while fixed water cannon automated cleaning equipment can perform dynamic cleaning with a larger cleaning area, it is only suitable for main lines and monotonous track conditions. Existing intelligent water flushing systems typically include radar or identification device systems and high-pressure water cannon systems. The workflow is as follows: the radar or camera identification device locates and identifies the insulators ahead, and then the control system guides the high-pressure water cannon to target the insulators for high-pressure flushing. However, these solutions have significant drawbacks in practical applications, especially in complex station and hub sections:

[0068] Perception limitations: Fixed sensors (such as cameras and lidar) have limited field of view (FOV), making it difficult to quickly identify and locate all targets to be cleaned in densely populated equipment areas (such as environments with multiple tracks, multiple insulators, garages, or signal towers that may obstruct the view). This is especially true in station and hub sections where there are many types of insulators, which can easily lead to overlap within the field of view.

[0069] Limitations of execution: The jet path of fixed or limited-range actuators (water cannons) is easily blocked by contact wire posts, cantilever structures, nearby equipment or the vehicle's own devices, making it impossible to form an effective cleaning trajectory.

[0070] Poor adaptability: It cannot dynamically adjust perception and execution strategies according to environmental complexity, lacks global optimization capabilities, resulting in blind spots or repetitive operations during cleaning.

[0071] In view of this, in order to solve the above-mentioned technical problems, this application provides an insulator cleaning method.

[0072] In one exemplary embodiment, the insulator cleaning method provided in this application can be applied to, for example, Figure 1 The insulator cleaning system shown is illustrated. The rail flatcar (1) serves as a mobile carrying platform, which can mount the entire water flushing equipment, including the movable sensing device (2), the movable cleaning device (3), the central control room (5), the water storage tank (6), the generator set (7), and the water pump station (8). The movable sensing device (2) is typically mounted at the end of the flatcar, offering advantages such as unobstructed operation and early identification. A top-view diagram of the insulator cleaning system is shown below. Figure 2As shown, the movable sensing device (2) is mounted on the first lifting column (22) of the first position adjustment mechanism, which can be adjusted up and down. The first lifting column (22) is also fixedly mounted on the sliding plate of the first transverse guide rail (21) of the first position adjustment mechanism, which can move left and right. Similarly, the movable cleaning device (3) is usually fixedly mounted in the middle of the flatcar, with the second lifting column (32) of the second position adjustment mechanism installed below it. The second lifting column (32) is also fixedly mounted on the sliding plate of the second transverse guide rail (31) of the second position adjustment mechanism, which can move left and right. Therefore, the movable sensing device (2) and the movable cleaning device (3) can move left and right and up and down independently. An execution module (3) such as a third position adjustment mechanism (4) can also be installed on the track flatcar (1). The third position adjustment mechanism (4) also has a third lifting column (42) that can be adjusted up and down and a third transverse guide rail (41) that can move left and right. The central intelligent control unit is an electrical control system composed of industrial computers and PLCs and other electrical control components, which is integrated and installed in the control room shed (5). Various sensors of the insulator (9) and the pollution detection system (10) can be installed on the top of the control room shed (5).

[0073] In one exemplary embodiment, such as Figure 3 As shown, an insulator cleaning method is provided, which is applied to... Figure 1 The central intelligent control unit in the insulator cleaning system is used as an example for explanation, including:

[0074] Step 302: When the carrier platform travels to the scanning position corresponding to the current work area, control the movable sensing device carried on the carrier platform to collect environmental information of the current work area.

[0075] The platform is also equipped with mobile cleaning equipment, and the environmental information includes at least one insulator in the current work area.

[0076] The carrying platform can be a general-purpose rail flatcar that carries a reinforced water washing platform or a special contact network insulator water washing vehicle to provide a mobile base platform; the current work area can be a designated railway contact network work area where insulator cleaning tasks need to be performed; the scanning position can be a pre-set initial stopping point within the work area for global environmental data acquisition.

[0077] The movable sensing device is mounted on a first position adjustment mechanism. This mechanism includes at least a lateral moving guide rail and a vertical lifting column, driving the sensing device to move a wide range of distances in both the horizontal and vertical directions. The sensing device is a multi-sensor fusion unit, including but not limited to: a lidar for 3D environment modeling, coarse target localization, and distance measurement; and a multispectral / high dynamic range camera for insulator pollution level identification, fine contour recognition, and visual evaluation of cleaning effects. It may also include: an infrared thermal imager for detecting abnormal heating points on insulators; millimeter-wave radar for penetrating rain and fog in adverse weather conditions; and an incremental encoder for detecting and measuring the moving speed and precise displacement of the mobile support platform.

[0078] Optionally, after the carrying platform travels and stops at the scanning position corresponding to the current work area, the control system activates the movable sensing device on the platform to collect environmental data of the current work area. The environmental information obtained includes at least the relevant information of the insulators to be cleaned in the area, such as the location information and degree of dirt of the insulators.

[0079] Step 304: Based on the environmental information, determine the pose data of the sensing device corresponding to each insulator and the water discharge parameters of the cleaning device.

[0080] Among them, the pose data is used to characterize the spatial attitude control parameters such as the position, orientation, and angle of the sensing device, and determines the observation range and viewing angle of the sensing device for the insulator; the water output parameters can be the working parameters that control the cleaning equipment to perform cleaning, including but not limited to water output pressure, flow rate, spray angle and water output time.

[0081] The movable execution module (i.e., the movable cleaning equipment) is mounted on the second position adjustment mechanism. The second position adjustment mechanism includes at least a horizontal moving guide rail and a vertical lifting column, driving the execution module to move independently of the sensing module. The execution module is a servo-driven high-pressure two-degree-of-freedom automatic water cannon with high-precision rotation in both pitch and horizontal directions, enabling precise positioning of the water jet to strike the target insulator. The water cannon also integrates a miniature vision sensor for real-time feedback of the jet trajectory, and the control system adjusts the water cannon's output flow rate and pressure parameters.

[0082] Optionally, the central intelligent control unit analyzes the position and spatial distribution of each insulator based on the environmental information of the work area collected by the sensing devices; for each insulator to be cleaned, it calculates and determines the pose data of the sensing devices that can be effectively observed, as well as the water discharge parameters of the cleaning equipment that can meet the cleaning requirements of the insulator.

[0083] Step 306: For each insulator, control the cleaning equipment to clean the insulator based on the corresponding water output parameters, and control the sensing equipment to perform the first obstacle detection on the insulator.

[0084] The first obstacle can be an obstruction on the observation path or jet path during the cleaning process.

[0085] Optionally, the central intelligent control unit processes the insulators in the work area one by one in a preset order. For the insulator currently being processed, the control unit controls the cleaning equipment to perform cleaning operations on the insulator using pre-matched water discharge parameters. At the same time, the control unit controls the sensing equipment to perform real-time detection on the insulator and the surrounding area, thereby determining whether there is a first obstacle on the observation path or the cleaning path.

[0086] Step 308: When the sensing device detects the first obstacle, the sensing device and the cleaning device are controlled to translate according to the positional relationship between the first obstacle and the insulator. The water discharge parameters corresponding to the insulator are adjusted according to the positional relationship between the cleaning device and the insulator after translation. Based on the adjusted water discharge parameters, the process returns to step 306 until the insulator cleaning is detected to be complete.

[0087] The detection area of ​​the translational sensing device and the water flow area of ​​the cleaning device both avoid the first obstacle.

[0088] Optionally, when the sensing device detects the first obstacle during the cleaning process, the system calculates the direction and distance that the sensing device and the cleaning device need to move according to the spatial relationship between the obstacle and the insulator, and controls the two devices to move and adjust synchronously so that the detection area of ​​the sensing device and the water flow area of ​​the cleaning device can bypass and avoid the obstacle; then, based on the new relative position between the cleaning device and the insulator after the translation, the original water discharge parameters corresponding to the insulator are adaptively adjusted to ensure that effective cleaning can still be achieved in the new position.

[0089] Optionally, after completing the obstacle avoidance and translation of the sensing and cleaning equipment and adaptively adjusting the water output parameters, the control system returns to the cleaning execution step and controls the cleaning equipment to continue cleaning the current insulator according to the updated water output parameters; this process is repeated until the system confirms through sensing detection that the insulator has been cleaned, and then the cleaning operation ends.

[0090] In the aforementioned insulator cleaning method, a mobile sensing and cleaning device is mounted on a support platform. First, environmental information of the work area is collected at the scanning location, and the corresponding pose data and water discharge parameters for each insulator are planned. During the insulator cleaning process, the sensing device performs real-time obstacle detection. When an obstacle is detected, the sensing and cleaning devices are synchronously adjusted according to the positional relationship between the obstacle and the insulator, ensuring that both the detection area and the water flow area avoid the obstacle. The water discharge parameters are also adaptively adjusted before cleaning continues until completion. This method enables autonomous obstacle avoidance and cleaning in complex working environments, reducing the risks of working at heights. Furthermore, dynamic obstacle avoidance and adaptive adjustment of water discharge parameters mitigate cleaning interruptions caused by obstacles, thereby improving the efficiency of insulator cleaning in complex environments.

[0091] In one exemplary embodiment, the insulator cleaning method further includes: determining a cleaning sequence for at least one insulator based on environmental information.

[0092] The cleaning sequence can be the order in which all insulators are cleaned.

[0093] Optionally, the central intelligent control unit first analyzes the number and distribution of insulators in the work area based on the collected environmental information, and plans a reasonable cleaning sequence. For example, it can follow the order from near to far to improve efficiency, clean the heavily soiled first to quickly reduce risks, or plan a specific path to avoid frequent reciprocating movement of equipment, thus determining the cleaning sequence.

[0094] For each insulator, the cleaning equipment is controlled to clean the insulator based on the corresponding water discharge parameters, including: traversing at least one insulator according to the cleaning sequence; and when each insulator is traversed, the cleaning equipment is controlled to clean the insulator based on the corresponding water discharge parameters.

[0095] Optionally, during cleaning, all insulators to be cleaned are traversed sequentially according to the cleaning sequence. For each insulator traversed, the cleaning equipment is controlled to use the water discharge parameters pre-matched for that insulator to perform targeted cleaning, and the cleaning operation of all insulators is completed in sequence.

[0096] In this embodiment, by determining the cleaning sequence of the insulators based on environmental information and cleaning the insulators one by one according to the cleaning sequence, the cleaning operation can be carried out in an orderly manner according to the planned order, reducing the problems of chaotic cleaning sequence, repeated cleaning or missed cleaning, thereby improving the operation efficiency of batch insulator cleaning.

[0097] In one exemplary embodiment, such as Figure 4 As shown, before the cleaning equipment cleans the insulators based on the corresponding effluent parameters, the following steps are also included:

[0098] Step 402: If it is detected that the carrying platform is not located in the preset working area corresponding to the insulator, control the carrying platform to travel to the preset working area corresponding to the insulator.

[0099] The preset work area is the area within the current work area that ensures the cleaning equipment can clean the insulators, and each preset work area corresponds to at least one insulator.

[0100] Optionally, the central intelligent control unit first detects whether the carrying platform is already within the preset working area corresponding to the insulator. For example, the central intelligent control unit compares the actual position with the preset working area (i.e., the planned optimal working point) corresponding to the insulator to determine whether the carrying platform is already within the preset working area corresponding to the insulator. If it is not in the area, the carrying platform is controlled to travel to the preset working area, which is part of the overall current working area and can ensure that the cleaning equipment can clean the insulator normally.

[0101] In this embodiment, by driving the carrier platform to the preset working area corresponding to the insulator before cleaning, it can be ensured that the cleaning equipment is always within the effective working range, thereby alleviating problems such as inadequate cleaning, ineffective jet or equipment interference caused by improper platform position, and thus improving the reliability of cleaning.

[0102] In one exemplary embodiment, such as Figure 5 As shown, the insulator cleaning is complete, including:

[0103] Step 502: Detect the degree of dirtiness of the insulator.

[0104] The degree of contamination can be the level of pollution on the surface of the insulator, which is a quantification of substances such as dust, salt spray and chemical pollutants that accumulate on its surface and cause a decline in insulation performance. For example, changes in the thickness, area coverage or optical properties (such as color and reflectivity) of the surface coating.

[0105] Optionally, after the cleaning equipment has cleaned an insulator for a period of time, the system instructs the movable sensing device to adjust its position and acquire an image of the current surface condition of the insulator. The system processes and analyzes the acquired image, for example, by comparing the features of the cleaned insulator with the baseline image, or by using multispectral information to analyze the surface material composition, to calculate or determine the current degree of dirtiness (such as pollution coverage and gray value changes).

[0106] Step 504: If the detected level of dirt meets the preset cleaning standard, it is determined that the insulator cleaning is complete.

[0107] The preset cleaning standard can be a pre-set threshold or condition for determining whether the insulator has been cleaned; for example, it can be a quantitative indicator (such as the proportion of dirty area being less than 1%) or a qualitative visual standard (such as the standard sample color of the surface color returning to a clean state).

[0108] Optionally, if the detected level of dirt meets the preset cleaning standard, it indicates that the cleaning of the detected insulator is complete; otherwise, the system determines that the cleaning is not complete, and triggers a new round of cleaning parameter adjustment (such as increasing water pressure and changing angle) and continues cleaning, and then returns to this step for detection, forming a cycle of cleaning-evaluation-re-cleaning, until the detected level of dirt meets the preset cleaning standard, and it is determined that the cleaning of the detected insulator is complete.

[0109] In this embodiment, the cleaning completion status is determined by comparing the degree of dirt with a preset standard, which realizes accurate quantitative judgment of the cleaning effect, alleviates the problems of strong subjectivity of manual judgment, over-cleaning or incomplete cleaning, and reduces the time spent on ineffective cleaning while ensuring the cleaning quality of insulators.

[0110] In one exemplary embodiment, detecting the degree of dirtiness of an insulator includes: controlling a sensing device to acquire a current image of the insulator; and identifying the current image to obtain the degree of dirtiness of the insulator.

[0111] The current image can be a real-time image captured by the sensing device during the cleaning process, taken directly at the target insulator, containing visual information about the distribution of dirt and the cleanliness of the insulator surface.

[0112] Optionally, during the cleaning process of the insulator, the system controls the sensing device to aim at the target insulator and collect its current on-site image; then the image is processed to extract the dirt features and distribution information on the surface of the insulator, thereby obtaining the degree of dirtiness of the insulator.

[0113] In this embodiment, the degree of dirtiness is detected by using a sensing device to collect images and perform identification, which improves the accuracy of cleaning quality evaluation. At the same time, it realizes real-time online monitoring of the cleaning effect, ensuring that the cleaning is only ended when the insulator reaches the preset cleaning standard, thereby improving the stability of cleaning quality.

[0114] In one exemplary embodiment, such as Figure 6 As shown, the control cleaning equipment cleans the insulators based on the corresponding effluent parameters, including:

[0115] Step 602: Control the cleaning equipment to emit water flow based on the corresponding water output parameters.

[0116] The cleaning equipment includes an image acquisition device; water is used to clean the insulators.

[0117] The image acquisition device can be a miniature vision sensor integrated near the water cannon nozzle, used to capture real-time images of water jets and the cleaning status of insulators.

[0118] Optionally, the central intelligent control unit sends control commands to the cleaning equipment based on the water outlet parameters matched for the current insulator. The cleaning equipment adjusts the water pressure, flow rate, and spray angle according to these parameters to spray high-pressure water outward. This water stream is sprayed directly onto the surface of the insulator, removing dirt through rinsing and thus cleaning the insulator.

[0119] Step 604: Obtain the current image of the water flow captured by the image acquisition device, and determine the cleaning effect of the water flow on the insulator based on the current image.

[0120] The current image of the water flow can be one or more frames of digital image captured in real time by an image acquisition device during the cleaning process, and the image includes the shape of the water flow and the local surface condition of the insulator where the water flow impacts; the cleaning effect can be an immediate assessment of the effectiveness of the water flow on the insulator. For example, whether the water column accurately hits and covers the preset cleaning area, and whether there is any deviation, scattering, or blind spots in the coverage.

[0121] Optionally, while controlling the water cannon to spray water to clean the insulators, the central intelligent control unit simultaneously drives a miniature vision sensor integrated on the water cannon nozzle to continuously capture images, obtaining real-time images of the backflow status and contact conditions. The central intelligent control unit then rapidly processes and analyzes the acquired images. Its core function is to assess the jet coverage. For example, by identifying the outline and brightness of the water column in the image and comparing it with the expected image position of the insulator target area, the system can determine whether the water flow accurately covers the target and the uniformity of the coverage. Based on the image analysis results, the system determines the current effectiveness of the water flow, such as: good coverage, a certain distance offset to the left, or a blind spot on the right.

[0122] Step 606: Adjust the water discharge parameters corresponding to the insulator according to the cleaning effect, and return to S602 based on the adjusted water discharge parameters.

[0123] Optionally, the system receives the cleaning effect from a miniature vision sensor, compares this effect with the expected target (e.g., the jet should completely cover the insulator disc), and decides how to adjust the parameters: if the coverage is off, it calculates the compensation required for the water cannon's pitch or yaw angle and generates a new angle command; if the coverage is poor, it determines that the water pressure or flow rate needs to be increased and generates a new power parameter command; finally, the calculated new parameters (i.e., the adjusted outlet parameters) are sent to the cleaning equipment. The water cannon's servo mechanism then fine-tunes the angle, and the water pump station adjusts the pressure and flow rate, thereby enabling the jet (i.e., the water flow) to correct deviations in real time.

[0124] Optionally, after adjusting the water output parameters based on the cleaning effect, the control system returns to the cleaning execution step and controls the cleaning equipment to re-spray water using the latest adjusted water output parameters to continue cleaning the insulators.

[0125] Optionally, the real-time tracking and control subsystem in the insulator cleaning system is the executor of the decision. On the one hand, it sends the pose command of the sensing module (i.e., the movable sensing device) to the servo controller of its underlying first position adjustment mechanism (lateral guide rail 21, lifting column 22, etc.) to drive the sensing module to move to the predetermined observation point. On the other hand, it receives the ultra-high precision real-time pose data stream (up to millimeter level) transmitted by the sensing module under the optimal pose for a specific target. The system combines these data with the unobstructed path planning results to generate dynamic water cannon servo control commands, accurately control the second position adjustment mechanism (lateral guide rail 31, lifting column 32, etc.) or the third position adjustment mechanism (4) and its water cannon's pitch angle, yaw angle, and water pressure and flow rate, to achieve dynamic tracking cleaning. The miniature vision sensor integrated near the water cannon nozzle provides real-time feedback on the jet coverage, forming a closed-loop control for the cleaning effect.

[0126] Optionally, the actions of the execution module (i.e., the movable cleaning equipment) alter the local environment (such as water splashing) while the work platform is also moving. These new states are continuously captured by the perception module (2) and input as new data into the global planning subsystem, thereby initiating the next "perception-planning-decision-execution" cycle, enabling the system to have true environmental adaptability.

[0127] In this embodiment, by integrating the image acquisition device into the cleaning equipment, while emitting water to clean the insulators, the water flow image is acquired in real time and the cleaning effect is judged. The water output parameters are dynamically and adaptively adjusted according to the effect, and the cleaning operation is performed cyclically based on the adjusted parameters. This ensures that the water output parameters always match the actual dirt condition and cleaning status of the insulators, thereby improving the accuracy and adequacy of the cleaning.

[0128] In an exemplary embodiment, the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to each insulator are determined based on environmental information, including: fusing environmental information with the railway catenary map associated with the current work area to obtain a target semantic map; and performing pose collaborative planning for the sensing device and the cleaning device based on the target semantic map to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to each insulator.

[0129] Among them, the railway catenary map can be a pre-loaded digital engineering base map, including but not limited to line GIS (Geographic Information System) data (such as macro-geographic information) and station map (such as the precise design coordinates of micro-equipment layout); the semantic map can be a structured environmental map formed after fusion, containing semantic information such as object category, location, occlusion relationship and workable area.

[0130] Optionally, the real-time collected environmental information is segmented, classified, and registered, and then precisely aligned and fused with a pre-set railway catenary map in a spatial coordinate system. A semantic map is output, in which all insulator targets and obstacles are accurately identified and labeled. The collaborative decision-making subsystem receives this semantic map and initiates parallel planning calculations. Then, with the goal of covering all insulators without blind spots and minimizing the number of movements, one or more optimal observation pose sequences are calculated for the sensing devices, and an optimal jet parabolic trajectory from the water cannon to its surface is calculated separately for each insulator in the map. This trajectory must avoid all obstacles in the map. Finally, the above two planning results are jointly verified and adjusted to resolve possible spatial or temporal interference between devices and form advanced collaborative tactics. The final output is the pose data of the sensing device corresponding to each insulator and the water discharge parameters of the cleaning equipment.

[0131] In this embodiment, by combining map fusion and pose collaborative planning, an accurate semantic map can be established in a complex railway catenary environment, thereby achieving coordinated and unified action of sensing and cleaning equipment. This alleviates problems such as equipment interference, blind spots in observation, or unreasonable cleaning paths, and improves planning accuracy and operational safety.

[0132] In an exemplary embodiment, pose collaborative planning is performed on the sensing device and the cleaning device based on the target semantic map to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to each insulator. This includes: for each insulator, determining the initial pose data of the sensing device corresponding to the insulator based on the insulator's position information and the spatial contour information of the second obstacle; and determining the initial water discharge parameters of the cleaning device corresponding to the insulator based on the insulator's position information, the insulator's dirt information, and the spatial contour information of the second obstacle; and performing pose collaborative planning on the sensing device and the cleaning device based on the initial pose data and the initial water discharge parameters to obtain the pose data of the sensing device corresponding to the insulator and the water discharge parameters of the cleaning device.

[0133] The target semantic map includes at least one insulator, the location and contamination information of each insulator, a second obstacle in the current work area, and the spatial outline information of the second obstacle.

[0134] The second obstacle can be a fixed or temporary obstruction (such as a support, tower, or equipment box) within the work area that may affect the observation or cleaning path; the spatial contour information can be the spatial geometric features of the obstacle, such as its external dimensions, boundary range, height, width, and protruding structures; the initial pose data can be the initial attitude parameters of the sensing equipment, such as its position, angle, and orientation, obtained from the preliminary planning of the insulator and the obstacle; and the initial water discharge parameters can be the initial cleaning parameters, such as the water discharge pressure, flow rate, and spray angle, pre-planned based on the insulator position, the degree of contamination, and the obstacle.

[0135] Optionally, for each insulator, the initial pose data of the sensing device is first determined based on its position information and the spatial contour of the obstacle. Then, the initial water discharge parameters of the cleaning device are determined by combining the insulator position, dirt information and obstacle contour. Finally, the pose of the sensing device and the cleaning device is coordinated based on the initial pose and water discharge parameters to obtain the final pose data and water discharge parameters adapted to the insulator.

[0136] Optionally, collaborative planning can begin after receiving the three-dimensional semantic map (i.e., the target semantic map) through the collaborative decision-making subsystem in the central intelligent control unit of the insulator cleaning system. This subsystem contains two parallel computing engines: one is the observation pose optimization engine, whose task is to calculate one or a series of optimal observation positions and angles for the mobile sensing module (i.e., the mobile sensing device). The evaluation criterion is whether it can cover all the targets to be cleaned in the map with the fewest number of moves and without blind spots; the other is the unobstructed path planning engine, which is dedicated to serving the mobile execution module (3). Based on the same map, it calculates the best jet parabolic trajectory from the water cannon nozzle to the surface of the insulator for each target insulator. This trajectory must strictly avoid all obstacles marked in the map and meet the mechanical requirements of the cleaning impact force. Subsequently, the system performs collaborative calculation to ensure that the moving paths and working poses planned for the sensing module and the execution module do not interfere with each other in space and time, and can actively form collaborative tactics such as "high and low misalignment" and "lateral insertion" in complex scenarios, fundamentally solving the occlusion problem. Finally, this layer outputs an optimized, synchronized set of module pose sequence instructions (i.e., pose data of the sensing device corresponding to each insulator) and a target cleaning queue (i.e., water discharge data of the cleaning device corresponding to each insulator).

[0137] In this embodiment, by utilizing the location, dirt, and obstacle contour information in the semantic map, obstacle interference can be avoided in advance during the planning stage. At the same time, the sensing pose and cleaning parameters are accurately matched with the insulator status, ensuring unobstructed observation by the sensing device and enabling the cleaning device to have targeted cleaning capabilities, thereby improving operational safety, planning rationality, and cleaning accuracy.

[0138] In an exemplary embodiment, fusing environmental information with a railway catenary map associated with the current work area to obtain a target semantic map includes: standardizing environmental point cloud data to obtain target point cloud data; identifying dirt information of each insulator and spatial contour information of a second obstacle in the environmental image data; fusing the target point cloud data, the spatial contour information of the second obstacle, and the dirt information of each insulator to obtain an initial semantic map; and calibrating the position information in the initial semantic map according to the railway catenary map associated with the current work area to obtain the target semantic map.

[0139] The environmental information includes environmental point cloud data and environmental image data.

[0140] Among them, environmental point cloud data can be a three-dimensional spatial point set of the work area, used to characterize the spatial position and shape structure of objects such as insulators and obstacles; target point cloud data can be normalized three-dimensional point cloud data after standardization processing for subsequent fusion; environmental image data can be collected two-dimensional images of the work area, used to identify visual features such as dirt and obstacle outlines.

[0141] Optionally, the collected environmental point cloud data is standardized by filtering, denoising, and coordinate normalization to obtain target point cloud data; image recognition is performed on the collected environmental image data, that is, each insulator in the image is identified and its surface condition is analyzed to output dirt information; all obstacles in the image are identified and their two-dimensional or three-dimensional spatial contour information is extracted; the target point cloud data, the spatial contour information of obstacles identified from the image, and the dirt information of each insulator are fused. For example, the insulator contours identified in the image are associated with the corresponding point cloud clusters in the point cloud, and dirt level labels are added to them to generate an initial semantic map. At this point, the initial semantic map has semantics, but the location may not be accurate enough. Then, the railway catenary map is loaded, and the initial semantic map is registered with it. The coordinate transformation relationship between the two is found through the algorithm, and the position information of all targets (such as insulators and obstacles) in the initial semantic map is corrected and adjusted to accurately align with the coordinate system of the railway catenary map, and finally a high-precision semantic map that can be used for global planning is output, that is, the target semantic map.

[0142] Optional, such as Figure 7 As shown, the central intelligent control unit of the insulator cleaning system in this embodiment is a hierarchical processing and closed-loop feedback intelligent hub. Its core lies in the progressive processing flow consisting of global planning, collaborative decision-making, and real-time control. Specifically, after the insulator cleaning system is started, the multi-sensor fusion perception module (2) (including lidar, high dynamic range camera, etc.) begins to collect raw point cloud (i.e., environmental point cloud data) and image data (i.e., environmental image data). At the same time, the preset GIS and station map data are loaded. These two sources of information are input to the global planning subsystem, which first segments, classifies, and registers the real-time point cloud, and merges it with the preset map to generate a dynamic three-dimensional operation map (i.e., target semantic map) containing semantic information. This map not only marks all identified insulator string targets and their initial dirt status, but also clearly marks the three-dimensional outlines of all static obstacles such as carports, signal towers, and existing equipment, providing a unique real environmental reference for subsequent decision-making.

[0143] In this embodiment, by fusing multi-source data and combining it with the railway catenary map for location calibration, the accuracy and reliability of environmental modeling are improved. This enables the semantic map to have both accurate spatial coordinates and rich semantic features, which not only alleviates point cloud noise and positioning deviation, but also fully preserves key information such as insulator dirt and obstacle outlines.

[0144] In one exemplary embodiment, a mobile sensing module (i.e., a mobile sensing device), a mobile execution module (i.e., a mobile cleaning device), and a central intelligent control unit are installed or modified on a mobile carrier platform. The mobile carrier platform is also equipped with a power system for water and electricity supply, providing pressurized water jets for cleaning. First, the mobile sensing module moves up and down and left and right over a wide range to identify as many insulators as possible in the station area. The central intelligent control unit forms a three-dimensional coordinate map information (i.e., a target semantic map) of all insulators and quickly calculates and plans the time, sequence, path, and flushing posture of the mobile execution module for flushing each insulator, ensuring that there are no obstructions when flushing each insulator. Then, rapid, precise, and orderly cleaning is performed, and the pressure and flow rate of the water pressure system can be adjusted in real time according to the distance of the insulators, achieving adaptive, intelligent, and efficient cleaning of multiple insulators in the complex railway contact network.

[0145] The central intelligent control unit includes: a global planning subsystem: based on line GIS data, pre-input station maps, and initial scanning by the sensing modules, it constructs a 3D semantic map of the work area, marking all potential insulator targets and obstacle areas; a collaborative decision-making subsystem: dynamically calculates the optimal observation pose of the sensing modules to maximize field-of-view coverage of all targets to be identified; and simultaneously plans the optimal jet path for the execution modules that is collision-free and unobstructed (avoiding known obstacles). In complex areas, the sensing and execution modules are controlled to perform collaborative pose adjustments, for example, by adopting strategies such as "sensing module probes for identification - execution module follows up with cleaning" or "staggered layout to avoid mutual obstruction"; and a real-time tracking and control subsystem: based on the real-time pose and contamination data of the insulators provided by the sensing modules, it dynamically adjusts the water cannon jet angle, pressure, and flow rate, and achieves closed-loop tracking cleaning while in motion.

[0146] The power system for water and electricity supply includes water storage tanks, water pressure pump stations, water circulation filtration devices, generator sets, control room sheds, etc.

[0147] Optionally, the insulator cleaning system is also scalable, with its first and second position adjustment mechanisms being expandable to integrate rotary joints, giving the sensing module and the execution module horizontal rotation and pitch degrees of freedom, further increasing flexibility.

[0148] Optionally, at least one to two sets of movable sensing modules and at least one to four sets of movable execution modules can be added as needed to work collaboratively and efficiently flush the insulators, enabling simultaneous identification, capture, and flushing of more target insulators. Further expansion includes third, fourth, fifth, and sixth position adjustment mechanisms.

[0149] Optionally, the global planning subsystem of the central intelligent control unit also integrates a "dirt prediction and priority cleaning algorithm", which predicts the dirt accumulation rate of insulators in different sections based on historical dirt data, weather data, and line environmental data, and dynamically generates a cleaning priority queue with the goal of maximizing safety benefits.

[0150] Optionally, the system can be expanded to include a rapid scanning drone deployed at the front of the mobile platform. Before entering highly complex areas, the drone can take off first to perform a wide-angle 3D scan of the area, transmitting point cloud data back to the central intelligent control unit in real time to assist the global planning subsystem in quickly generating a more accurate initial operational map.

[0151] Optionally, the insulator cleaning method may also include: S1: The mobile carrier platform travels along the line and the system is started. S2: Active perception and mapping of the surrounding environment. S2.1: The mobile perception module moves to the starting scanning position (i.e., the scanning position corresponding to the current work area) according to the preset path or the instructions of the global planning subsystem, and performs a large-scale dynamic scan and identification of the environment in front and to the sides. S2.2: The central intelligent control unit integrates the perception data and the preset map to build / update the three-dimensional semantic map of the current work area, identify all insulator targets and their dirt status, and mark static and dynamic obstacles. S3: The collaborative decision-making subsystem performs collaborative pose planning based on the current map: S3.1: Plan one or more observation point pose sequences for the perception module (i.e., the pose data of the perception device corresponding to each insulator) to ensure that there are no blind spots covering all insulators to be cleaned. S3.2: Plan the cleaning jet path corresponding to each insulator for the execution module (i.e., the water discharge data of the perception device corresponding to each insulator), ensuring that there are no fixed or temporary obstacles blocking the path and that there is no interference with the perception module itself. S4: Control the perception module and the execution module to move to the planned starting pose. S5: For the first target insulator, the sensing module tracks and locks onto it, while the execution module adjusts the water cannon aiming based on real-time pose data and activates the jet for cleaning. A miniature vision sensor provides feedback on the cleaning coverage. S6: Once the current target is cleaned (meeting visual or preset standards), the system controls the sensing and execution modules to move collaboratively to the next working position according to the planned sequence, repeating S5 until all targets in this batch are cleaned. S7: The mobile platform continues to advance, continuously updating the insulator target data, and repeating S2-S6 until all tasks are completed.

[0152] In one exemplary embodiment, a flowchart of collaborative operations in a complex station environment is shown below. Figure 8 As shown; a schematic diagram of a flushing scenario during collaborative operations in a complex station environment. Figure 9 As shown. This embodiment describes a collaborative adaptive operation process in a complex station environment, which is an intelligent process based on dynamic, cyclical environmental perception. Specifically:

[0153] Step S1, Job initialization:

[0154] The mobile carrier platform (1) moves along the track into the designated cleaning operation area (such as a large marshalling yard). The system starts up, and each module completes its self-check. The central intelligent control unit (5) loads the digital base map of the station (i.e., the railway catenary map).

[0155] Step S2, Active Environmental Perception and Mapping:

[0156] The control unit instructs the movable sensing module (2) to move to a starting pose with a wide field of view (e.g., the left front side of the flatcar raised) through its first position adjustment mechanism (21, 22) according to the preset strategy or initial plan. The sensing module performs high-speed three-dimensional scanning and image capture of the fan-shaped area in front. After the data is transmitted back, a three-dimensional semantic map (i.e., the target semantic map) of the current work area is constructed or updated. At this time, the map may show that there is a set of insulators (9) on the right side of the platform, but part of their line of sight is blocked by a device pole (11); at the same time, if the water cannon (i.e., the movable cleaning equipment) fires directly from the current position, the jet path will interfere with the first position adjustment mechanism (lifting column 22) or the device pole (11) on the line.

[0157] Step S3, Collaborative Decision Generation:

[0158] After analyzing the map, the collaborative decision-making subsystem determines that "static occlusion exists". The collaborative cracking algorithm is then activated. The algorithm outputs the decision: adopt the strategy of "sensing and probing for surveillance, and executing a lateral assault". The specific plan is as follows: 1. The instruction sensing module (2) moves to the left along the transverse guide rail (21) to the limit position and further raises the lifting column (22) to obtain an "eagle eye" view that can pass over the top of the generator set (7) from the left and overlook the insulator group on the right; 2. At the same time, the instruction execution module (i.e., the movable cleaning equipment) moves to the right front through its second position adjustment mechanism (31, 32) to find a firing position that can bypass the gap or edge of the side of the equipment pole (11) and directly shoot at the insulator with a smooth flow channel.

[0159] Step S4, Module Coordinated Maneuvering and Target Locking:

[0160] The two modules move synchronously but independently along the planned path to the target pose. The sensing module (2) stably locks onto the target insulator (9) in the new pose and sends the precise aiming data stream relative to the water cannon (3) after coordinate transformation to the real-time tracking and control subsystem in real time.

[0161] Step S5, Precise Tracking and Cleaning:

[0162] The water cannon (3) adjusts its pitch and yaw angles according to the real-time data stream and activates the high-pressure jet. Throughout the cleaning process, the sensing module (2) continuously provides visual tracking, the miniature visual sensor provides feedback on the water jet's impact on the target, and the control unit dynamically fine-tunes the water cannon's attitude and water flow parameters to ensure the cleaning effect.

[0163] Step S6, Batch Evaluation and Cycle:

[0164] After the current insulator cleaning meets the standard, the collaborative decision-making subsystem evaluates the current three-dimensional semantic map: if there are still other targets in the same batch (the range that can be collaboratively cleaned from the current platform pose), the system controls the perception and execution module to move to the next set of collaborative poses according to the pre-planned sequence, jumps to step S4, and starts cleaning the next target until all targets in the batch are completed. If the batch has been completed, the system instructs the mobile carrier platform (1) to move forward to the next work section (S7), the process jumps back to step S2, and starts a new round of "perception-planning-collaboration-cleaning" cycle until the entire task is completed.

[0165] To more comprehensively demonstrate this solution, this embodiment presents an insulator cleaning method, specifically including:

[0166] 1. When the carrier platform travels to the scanning position corresponding to the current work area, control the mobile sensing device carried by the carrier platform to collect the environmental information of the current work area; wherein, the carrier platform is also carried with a mobile cleaning device, and the environmental information includes at least one insulator in the current work area;

[0167] 2. Standardize the environmental point cloud data to obtain the target point cloud data;

[0168] 3. Identify the dirt information of each insulator and the spatial contour information of the second obstacle in the environmental image data;

[0169] 4. The target point cloud data, the spatial contour information of the first obstacle, and the dirt information of each insulator are fused to obtain the initial semantic map;

[0170] 5. Based on the railway catenary map associated with the current work area, calibrate the location information in the initial semantic map to obtain the target semantic map; the target semantic map includes at least one insulator, the location information and dirt information of each insulator, the second obstacle in the current work area, and the spatial outline information of the second obstacle;

[0171] 6. For each insulator, based on the insulator's position information and the spatial contour information of the second obstacle, determine the initial pose data of the sensing device corresponding to the insulator; and,

[0172] 7. Based on the insulator's location information, the insulator's dirt information, and the spatial contour information of the second obstacle, determine the initial water discharge parameters of the cleaning equipment corresponding to the insulator;

[0173] 8. Based on the initial pose data and initial water discharge parameters, perform pose collaborative planning between the sensing device and the cleaning device to obtain the pose data of the sensing device corresponding to the insulator and the water discharge parameters of the cleaning device.

[0174] 9. For each insulator, control the cleaning equipment to clean the insulator based on the corresponding water output parameters, and control the sensing equipment to perform the first obstacle detection on the insulator;

[0175] 10. When the sensing device detects the first obstacle, the sensing device and the cleaning device are controlled to translate according to the positional relationship between the first obstacle and the insulator, and the water discharge parameters of the insulator are adjusted according to the positional relationship between the cleaning device and the insulator after translation; wherein the detection area of ​​the sensing device and the water discharge flow area of ​​the cleaning device after translation both avoid the first obstacle.

[0176] 11. Based on the adjusted water discharge parameters, return to the execution control cleaning equipment to perform the operation of cleaning the insulator based on the corresponding water discharge parameters until the control sensing equipment collects the current image of the insulator; identify the current image to obtain the degree of dirtiness of the insulator; if the detected degree of dirtiness meets the preset cleaning standard, determine that the insulator cleaning is completed.

[0177] The specific process of the above steps can be found in the description of the above method embodiments. The implementation principle and technical effect are similar, and will not be repeated here.

[0178] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0179] Based on the same inventive concept, this application also provides an insulator cleaning device for implementing the insulator cleaning method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the insulator cleaning device provided below can be found in the limitations of the insulator cleaning method described above, and will not be repeated here.

[0180] In one exemplary embodiment, such as Figure 10 As shown, an insulator cleaning device is provided, comprising: a data acquisition module 101, a processing module 102, and a cleaning module 103, wherein:

[0181] The data acquisition module 101 is used to control the mobile sensing device mounted on the carrier platform to collect environmental information of the current work area when the carrier platform travels to the scanning position corresponding to the current work area; wherein, the carrier platform is also equipped with a mobile cleaning device, and the environmental information includes at least one insulator in the current work area.

[0182] Processing module 102 is used to determine the pose data of the sensing device corresponding to each insulator and the water discharge parameters of the cleaning device based on environmental information.

[0183] The cleaning module 103 is used to control the cleaning equipment to clean the insulator based on the corresponding water discharge parameters for each insulator, and to control the sensing equipment to detect the first obstacle on the insulator.

[0184] The processing module 102 is also used to control the sensing device and the cleaning device to translate according to the positional relationship between the first obstacle and the insulator when the sensing device detects the first obstacle, and to adjust the water discharge parameters corresponding to the insulator according to the positional relationship between the translated cleaning device and the insulator; wherein the detection area of ​​the translated sensing device and the water discharge flow area of ​​the cleaning device both avoid the first obstacle.

[0185] The cleaning module 103 is also used to return to the execution control cleaning equipment to perform cleaning operations on the insulator based on the adjusted water outlet parameters until the insulator cleaning is detected to be complete.

[0186] In one embodiment, the cleaning module 103 is further configured to:

[0187] Based on environmental information, determine the cleaning sequence for at least one insulator;

[0188] For each insulator, the cleaning equipment is controlled to clean the insulator based on the corresponding effluent parameters, including:

[0189] According to the cleaning sequence, traverse at least one insulator;

[0190] When each insulator is visited, the cleaning equipment is controlled to clean the insulator based on the corresponding water discharge parameters.

[0191] In one embodiment, the cleaning module 103 is further configured to:

[0192] If it is detected that the carrier platform is not located in the preset working area corresponding to the insulator, control the carrier platform to travel to the preset working area corresponding to the insulator;

[0193] The preset work area is the area within the current work area that ensures the cleaning equipment can clean the insulators, and each preset work area corresponds to at least one insulator.

[0194] In one embodiment, the cleaning module 103 is further configured to:

[0195] The degree of dirtiness of the insulators is detected;

[0196] If the detected level of dirt meets the preset cleaning standard, the insulator cleaning is considered complete.

[0197] In one embodiment, the cleaning module 103 is further configured to:

[0198] Control the sensing device to collect the current image of the insulator;

[0199] The current image is analyzed to determine the degree of dirtiness of the insulator.

[0200] In one embodiment, the cleaning module 103 is further configured to:

[0201] The cleaning equipment is controlled to emit water streams based on corresponding water output parameters; the water streams are used to clean the insulators.

[0202] The current image of the water flow captured by the image acquisition device is obtained, and the cleaning effect of the water flow on the insulator is determined based on the current image.

[0203] Adjust the water discharge parameters corresponding to the insulators based on the cleaning effect;

[0204] Based on the adjusted effluent parameters, return to the execution control cleaning equipment to launch water flow based on the corresponding effluent parameters.

[0205] In one embodiment, the processing module 102 is further configured to:

[0206] The environmental information is fused with the railway catenary map associated with the current work area to obtain a semantic map;

[0207] Based on the semantic map, pose coordination planning is performed between the sensing device and the cleaning device to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device for each insulator.

[0208] In one embodiment, the processing module 102 is further configured to:

[0209] For each insulator, based on the insulator's position information and the spatial contour information of the second obstacle, the initial pose data of the sensing device corresponding to the insulator is determined; and,

[0210] Based on the insulator's location information, the insulator's dirt information, and the spatial contour information of the second obstacle, determine the initial water discharge parameters of the cleaning equipment corresponding to the insulator.

[0211] Based on the initial pose data and initial water discharge parameters, pose coordination planning is performed between the sensing device and the cleaning device to obtain the pose data of the sensing device corresponding to the insulator and the water discharge parameters of the cleaning device.

[0212] In one embodiment, the processing module 102 is further configured to:

[0213] The environmental point cloud data is standardized to obtain the target point cloud data;

[0214] Identify dirt information for each insulator and spatial contour information for second obstacles in environmental image data;

[0215] The target point cloud data, the spatial contour information of the first obstacle, and the dirt information of each insulator are fused to obtain an initial semantic map.

[0216] Based on the railway catenary map associated with the current work area, the location information in the initial semantic map is calibrated to obtain the semantic map.

[0217] Each module in the aforementioned insulator cleaning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0218] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements an insulator cleaning method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0219] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0220] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0221] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0222] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0223] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0224] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0225] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0226] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for cleaning insulators, characterized in that, The method includes: When the carrier platform travels to the scanning position corresponding to the current work area, the mobile sensing device mounted on the carrier platform is controlled to collect the environmental information of the current work area; wherein, the carrier platform is also equipped with a mobile cleaning device, and the environmental information includes at least one insulator in the current work area; Based on the environmental information, determine the pose data of the sensing device corresponding to each insulator and the water output parameters of the cleaning device; For each insulator, the cleaning equipment is controlled to clean the insulator based on the corresponding water output parameters, and the sensing equipment is controlled to perform a first obstacle detection on the insulator; When the sensing device detects a first obstacle, it controls the sensing device and the cleaning device to translate according to the positional relationship between the first obstacle and the insulator, and adjusts the water discharge parameters corresponding to the insulator according to the positional relationship between the translated cleaning device and the insulator; wherein the detection area of ​​the sensing device and the water discharge flow area of ​​the cleaning device both avoid the first obstacle after translation. Based on the adjusted water outlet parameters, the system returns to control the cleaning equipment to clean the insulator based on the corresponding water outlet parameters until the cleaning of the insulator is detected to be complete.

2. The method according to claim 1, characterized in that, The method further includes: Based on the environmental information, determine the cleaning sequence for the at least one insulator; The step of controlling the cleaning equipment to clean each insulator based on corresponding water output parameters includes: According to the cleaning sequence, traverse the at least one insulator; When each insulator is visited, the cleaning equipment is controlled to clean the insulator based on the corresponding water discharge parameters.

3. The method according to claim 1, characterized in that, Before controlling the cleaning equipment to clean the insulator based on corresponding effluent parameters, the method further includes: If it is detected that the carrier platform is not located in the preset working area corresponding to the insulator, the carrier platform is controlled to travel to the preset working area corresponding to the insulator. The preset work area is the area within the current work area that ensures the cleaning equipment can clean the insulator, and each preset work area corresponds to at least one insulator.

4. The method according to claim 1, characterized in that, The detection that the insulator cleaning is complete includes: The degree of contamination of the insulator is detected; If the detected level of dirt meets the preset cleaning standard, it is determined that the insulator cleaning is complete.

5. The method according to claim 4, characterized in that, The detection of the degree of contamination of the insulator includes: The control and sensing device acquires the current image of the insulator; The current image is analyzed to determine the degree of dirtiness of the insulator.

6. The method according to claim 1, characterized in that, The cleaning equipment includes an image acquisition device; controlling the cleaning equipment to clean the insulator based on corresponding water outlet parameters includes: The cleaning equipment is controlled to emit water flow based on corresponding water output parameters; wherein the water flow is used to clean the insulator. The current image of the water flow acquired by the image acquisition device is obtained, and the cleaning effect of the water flow on the insulator is determined based on the current image. Based on the cleaning effect, the water outlet parameters corresponding to the insulator are adjusted; Based on the adjusted water output parameters, the system returns to control the cleaning equipment to emit water based on the corresponding water output parameters.

7. The method according to any one of claims 1-6, characterized in that, The step of determining the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to each insulator based on the environmental information includes: The environmental information is fused with the railway catenary map associated with the current work area to obtain a target semantic map; Based on the target semantic map, pose collaborative planning is performed on the sensing device and the cleaning device to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to each insulator.

8. The method according to claim 7, characterized in that, The target semantic map includes the at least one insulator, the location information and dirt information of each insulator, the second obstacle in the current work area, and the spatial outline information of the second obstacle; The step of performing pose collaborative planning for the sensing device and the cleaning device based on the target semantic map to obtain the pose data of the sensing device corresponding to each insulator and the water discharge parameters of the cleaning device includes: For each insulator, based on the insulator's position information and the spatial contour information of the second obstacle, the initial pose data of the sensing device corresponding to the insulator is determined; and, Based on the position information of the insulator, the dirt information of the insulator, and the spatial contour information of the second obstacle, the initial water discharge parameters of the cleaning equipment corresponding to the insulator are determined. Based on the initial pose data and the initial water discharge parameters, pose collaborative planning is performed on the sensing device and the cleaning device to obtain the pose data of the sensing device and the water discharge parameters of the cleaning device corresponding to the insulator.

9. The method according to claim 7, characterized in that, The environmental information includes environmental point cloud data and environmental image data; the process of fusing the environmental information with the railway catenary map associated with the current work area to obtain a target semantic map includes: The environmental point cloud data is standardized to obtain the target point cloud data; Identify the dirt information of each insulator and the spatial contour information of the second obstacle in the environmental image data; The target point cloud data, the spatial contour information of the second obstacle, and the dirt information of each insulator are fused to obtain an initial semantic map. Based on the railway catenary map associated with the current work area, the location information in the initial semantic map is calibrated to obtain the target semantic map.

10. A cleaning device for railway contact wire insulators, characterized in that, The device includes: The data acquisition module is used to control the mobile sensing device mounted on the carrier platform to collect environmental information of the current work area when the carrier platform travels to the scanning position corresponding to the current work area; wherein, the carrier platform is also equipped with a mobile cleaning device, and the environmental information includes at least one insulator in the current work area; The processing module is used to determine the pose data of the sensing device corresponding to each insulator and the water output parameters of the cleaning device based on the environmental information. The cleaning module is used to control the cleaning equipment to clean the insulator based on the corresponding water output parameters for each insulator, and to control the sensing device to perform a first obstacle detection on the insulator. The processing module is further configured to, when the sensing device detects a first obstacle, control the sensing device and the cleaning device to translate according to the positional relationship between the first obstacle and the insulator, and adjust the water discharge parameters corresponding to the insulator according to the positional relationship between the translated cleaning device and the insulator; wherein the detection area of ​​the sensing device and the water discharge flow area of ​​the cleaning device after translation both avoid the first obstacle. The cleaning module is also used to return to the operation of controlling the cleaning equipment to clean the insulator based on the adjusted water output parameters, until the cleaning of the insulator is detected to be complete.

Images