Intelligent dust cleaning robot with self-adaptive pipe diameter in dust removal pipeline

Abstract

This invention belongs to the field of robot dust removal control technology, specifically an intelligent dust removal robot that adapts to the pipe diameter in dust removal pipelines. By integrating a variable diameter walking mechanism into the robot body, it can sense local pipe diameter changes in real time and dynamically adjust the contact posture of its own support structure during its movement along the dust removal pipeline. This achieves reliable passage capability and operational stability throughout the entire pipeline. At the same time, when faced with the dual tasks of pipe diameter adjustment and dust removal, the robot defines the working window based on its current position. Combining the positional relationship between the dust accumulation area, the working window, and the current pipe diameter segment, control instructions containing the work type and execution order are generated. This achieves decoupling and orderly scheduling of multi-task requirements, ensuring that in complex pipeline environments, priority is given to ensuring robot pipe diameter adaptation, and dust removal actions are completed in advance when conditions permit, thereby improving the autonomy, safety, and continuity of the operation.

Description

Technical Field

[0001] This invention belongs to the field of robot dust removal control technology, specifically an intelligent dust removal robot with adaptive pipe diameter in dust removal pipelines. Background Technology

[0002] In industrial sectors such as mining, metallurgy, and thermal power generation, large amounts of dust are generated during production. To ensure a safe working environment, centralized dust collection systems are commonly used. These systems connect each dust-generating point to a central dust collector through a network of dust collection pipelines, enabling the transport and purification of dust-laden airflow. However, during operation, factors such as airflow velocity attenuation and temperature fluctuations cause some dust to settle and accumulate at the bottom of the pipelines, forming a dust layer.

[0003] Currently, the mainstream solution to this type of dust accumulation problem is to deploy a dust removal inspection trolley inside the pipeline. This device is equipped with a rotating dust removal structure, which identifies the dust accumulation area during its movement and re-raises the deposited dust through mechanical disturbance, allowing it to be transported to the dust collector with the main airflow, thereby achieving automatic dust removal.

[0004] However, due to the complexity of industrial site layouts, dust collection pipelines often contain irregular structures such as diameter changes and bends. Existing inspection trolleys mostly use fixed wheelbases or simple elastic support structures, which are prone to problems such as support instability and movement stagnation when passing through curved areas, making it difficult to guarantee continuous and stable passage within the entire pipeline.

[0005] Even if some equipment has pipe diameter self-adaptation function, when faced with the dual tasks of pipe diameter adaptation and dust removal at the same time, its control logic either mechanically executes according to a fixed program, such as forcibly removing dust in a bumpy diameter change section, resulting in poor fit of the dust removal structure and poor dust removal effect; or it relies on manual remote intervention and judgment, and cannot achieve autonomous continuous operation.

[0006] Furthermore, when assessing the ash accumulation status inside dust removal ducts, existing technologies mostly rely on a single sensor, such as a vision sensor. This sensor is susceptible to interference from factors such as insufficient light inside the duct and dust obscuring the image, resulting in distorted image information and difficulty in accurately determining the true thickness of the ash accumulation, which affects the judgment of dust removal needs. Summary of the Invention

[0007] The purpose of this invention is to improve upon the shortcomings of existing technologies and provide an intelligent dust removal robot that adapts to the pipe diameter in dust removal pipelines. By having a variable diameter walking mechanism, it dynamically plans the pipe diameter adaptation and the execution priority of dust removal operations, effectively solving the problems mentioned in the background technology.

[0008] The objective of this invention can be achieved through the following technical solution: an intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline, comprising: a dust accumulation thickness sensing module: processing in real time the visual image and pulse echo signal sequence of the bottom of the pipeline collected during the robot's movement, analyzing the visual dust accumulation thickness and pulse dust accumulation thickness, and fusing the two for consistency verification, generating a heat map of the dust accumulation thickness distribution at the bottom of the pipeline.

[0009] Pipe diameter status recognition module: While sensing the thickness of dust accumulation, it acquires real-time pressure signals from multiple pressure feedback elements in the robot's walking mechanism to identify the current pipe diameter status of the robot.

[0010] Concurrency condition judgment module: Based on the heat map of ash thickness distribution and pipe diameter status, determine whether the ash cleaning requirement and the pipe diameter adaptation requirement are concurrent.

[0011] Decision execution module: When dust removal and pipe diameter adaptation are required concurrently, the module defines the operation window based on the robot's current position, determines the order of operation execution based on the positional relationship between the dust accumulation area, the operation window, and the current pipe diameter segment, and executes the operation according to the decision instructions.

[0012] Combining all the above technical solutions, the positive effects of this invention are as follows: (1) By integrating a variable diameter walking mechanism on the robot body, this invention enables the robot to sense local pipe diameter changes in real time and dynamically adjust the contact posture of its own support structure during the process of traveling along the dust removal pipeline, thereby achieving stable passage capability throughout the entire pipeline range. This effectively avoids problems such as slippage and jamming that are prone to occur in irregularly shaped pipelines by traditional fixed wheel track or simple elastic structure. At the same time, stable pipe wall contact improves the fitting accuracy of the dust removal structure in subsequent dust removal operations.

[0013] (2) In the process of the robot performing dust removal operation on the dust removal pipeline, the present invention integrates the visual imaging information of the bottom of the pipeline and the pulse echo ranging signal in the dust accumulation perception stage to construct a multimodal dust accumulation thickness analysis, which effectively overcomes the misjudgment problem of single vision method under complex working conditions such as low illumination and dust cover, and improves the accuracy and anti-interference ability of dust accumulation thickness perception.

[0014] (3) When facing the dual tasks of pipe diameter adaptation and dust removal, the present invention defines the operation window based on the current position of the robot, and makes operation sequence decisions based on the positional relationship between the dust accumulation area and the operation window and the current pipe diameter segment. This achieves decoupling and orderly scheduling of multi-task requirements, ensuring that in complex pipeline environments, the robot pipe diameter adaptation is prioritized, and the dust removal action is completed in advance when conditions permit, thereby improving the autonomy, safety and continuity of the operation. Attached Figure Description

[0015] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0016] Figure 1 This is a diagram showing the module connections involved in the dust removal robot's dust removal and pipe diameter adaptation processes in this invention.

[0017] Figure 2 This is a flowchart illustrating the process of identifying the current pipe diameter status of the robot in this invention.

[0018] Figure 3 This is a 3D diagram of the dust removal robot in this invention.

[0019] Figure 4 This is a cross-sectional view of the dust removal robot in this invention.

[0020] Reference numerals: 1. Dust removal structure; 2. Front housing; 3. Rear housing; 4. Adjusting front rod; 5. Elastic compensator; 6. Adjusting rear rod; 7. Anti-slip wheel; 8. Axle; 9. Support rod; 10. Drive motor; 11. Hub frame; 12. Detection device; 13. Power supply; 14. Connecting soft shell; 15. Control processor; 16. Universal connector; 17. Ball bearing slider; 18. Lead screw; 19. Lead screw retainer; 20. Adjusting motor; 21. Dust removal motor. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] This invention provides an intelligent dust removal robot that adapts to the pipe diameter inside a dust removal pipeline.

[0023] Please see Figure 3 and Figure 4 As shown, in one embodiment of the present invention, the main body of the dust cleaning robot is composed of a dust cleaning structure 1, a front shell 2 and a rear shell 3 connected by a connecting soft shell 14.

[0024] A detection device 12 is installed at the front end of the front housing 2. This device can detect the thickness of the dust accumulation inside the pipe and the straightness of the pipe, and transmit this signal to the control processor 15. Three support rods 9 are provided at 120° intervals on the outer side of the wide diameter of the front housing 2 and the rear housing 3. One side of each set of support rods 9 can be connected to the connecting end in a hinged manner. A structure is provided in the middle of the support rod 9 to one side of the adjusting front rod 4, which realizes the movable connection between the support rod 9 and the adjusting front rod 4; the other side of the support rod 9 is movably connected to the hub frame 11 using the same connection structure. The hub frame 11 serves as the robot's leg skeleton, on which two sets of axles 8 are installed, and anti-slip wheels 7 are installed at both ends of each set of axles 8. The anti-slip wheel 7 connected to the front housing 2 is a drive wheel, while the anti-slip wheel 7 connected to the rear housing 3 is a driven wheel. In terms of structure, the drive wheel has a drive motor 10 mounted on its hub, and the output shaft of the drive motor 10 is equipped with a bevel gear. The axle 8 of the drive wheel is equipped with a corresponding bevel gear. The combination of the two bevel gears allows the drive motor 10 to independently and precisely control the drive wheel, thereby achieving the purpose of controlling the robot's movement. The other end of the adjusting front rod 4, which is connected to the support rod 9, is connected to the adjusting rear rod 6. An elastic compensator 5 is installed at the connection point, and the elastic compensator 5 is equipped with a pressure feedback element that can monitor real-time pressure and feed it back to the control processor 15. The other end of the adjusting rear rod 6 is connected to the ball bearing slider 17 via a hinged connection.

[0025] At the front of the robot's interior is a cleaning motor 21, which drives the cleaning structure 1. A power supply 13, providing power to the robot, is installed around the cleaning motor 21. This power supply powers the drive motor 10, the adjusting motor 20, and the cleaning motor 21 via internal wiring. In the middle are two sets of ball screw adjusting devices, fixedly installed inside the front housing 2 and the rear housing 3 respectively, connected by a universal connector 16. Each device consists of a ball slider 17, a screw 18, and a screw retainer 19. Each screw 18 has a screw retainer 19 at both ends, which fixes both ends of the screw to the housing, allowing only rotation. When the adjusting motor 20 connected to the end of the screw 18 rotates, it drives the ball slider 17 to shift, thereby changing the position of the adjusting rod and adjusting the support rod 9.

[0026] After the cleaning robot enters the pipeline, its six sets of support wheels provide stable support, ensuring that it can run smoothly inside the pipeline. When the detection device 12 at the front end of the robot detects the presence of dust accumulation at the bottom of the pipeline, the signal is fed back to the control processor 15. The control processor 15 controls the cleaning motor 21, and the front cleaning structure 1 lifts up the dust accumulation at the bottom of the pipe. The lifted dust is carried away by the airflow to achieve the purpose of cleaning.

[0027] When the pipe diameter changes, the pressure feedback element on each set of elastic compensators 5 transmits the pressure to the control processor 15 in real time. The control processor judges the signal and drives the action of the adjustment motor 20, so that the robot can make timely adjustments to adapt to the pipe diameter change through the lead screw adjustment device.

[0028] See Figure 1 As shown, the robot involves a dust thickness sensing module, a pipe diameter status recognition module, a concurrent condition judgment module, and a decision execution module in the entire dust removal process. The dust thickness sensing module and the pipe diameter status recognition module are both connected to the concurrent condition judgment module, which in turn is connected to the decision execution module.

[0029] The dust accumulation thickness sensing module is used to process in real time the visual image of the bottom of the pipe and the pulse echo signal sequence collected by the detection device 12 during the robot's walking process. By analyzing the visual dust accumulation thickness and the pulse dust accumulation thickness, and fusing the two for consistency verification, a heat map of the dust accumulation thickness distribution at the bottom of the pipe is generated.

[0030] Given the complex internal environment of dust collection ducts, accurate assessment of dust accumulation requires sensing the thickness of the dust at the bottom of the duct. Traditional methods primarily rely on visual sensors to acquire images inside the duct to evaluate dust accumulation. However, due to poor lighting conditions and dust obscuring the images inside the duct, the acquired information is prone to distortion and cannot accurately reflect the actual dust thickness, thus affecting the accuracy of determining dust removal needs.

[0031] To address the aforementioned issues, this invention integrates visual imaging and pulse-echo ranging for dual ash accumulation thickness sensing within the pipeline. Visual imaging estimates the ash accumulation thickness from an appearance perspective by acquiring the surface morphology of the ash accumulation area, while pulse-echo ranging retrieves the ash accumulation thickness from a physical perspective by transmitting short pulse signals and inverting the ash deposition thickness based on the propagation time delay. The two technologies are complementary, improving the accuracy of ash accumulation thickness measurement.

[0032] In a specific embodiment, the visual dust accumulation thickness analysis process is as follows: First, the visual image of the bottom of the pipe is subjected to image enhancement processing, specifically including contrast adjustment and noise suppression, in order to improve image quality.

[0033] Secondly, since the bottom of the pipe is usually symmetrically U-shaped or arc-shaped, and the accumulated ash covers it to form a relatively flat or gently sloping upper surface, the two appear as two approximately parallel curves in the image. Therefore, in the enhanced image, the uppermost edge is extracted through edge detection as the contour line of the ash accumulation surface. In the area below the ash accumulation contour, edge detection is used to search downwards for edge lines as the contour line of the inner wall of the pipe, providing the edge feature basis for subsequent calculation of ash accumulation thickness.

[0034] Next, given that visual imaging inside the pipe is limited by factors such as the installation angle, there is no fixed mapping relationship between the pixel size in the acquired image and the actual physical size. Directly estimating the dust accumulation thickness based on pixel distance will introduce errors. Therefore, by setting a visual reference mark of known size at a fixed position inside the pipe, it is ensured that the reference mark can be synchronously imaged and completely included in the image each time the bottom image of the pipe is acquired.

[0035] By detecting the pixel size of the marker in the visual image and combining it with its known actual physical size, the local scale factor under the current imaging conditions is calculated by comparing the actual physical size of the visual reference marker with the pixel size. This scale factor represents the real physical length corresponding to a unit pixel under the current imaging conditions, thereby realizing the conversion from pixel space to real physical space.

[0036] Subsequently, for each pixel on the surface contour line of the ash accumulation, its corresponding inner wall contour point of the pipe is matched in the direction perpendicular to the pipe wall, and the vertical distance between the two is calculated. This vertical distance represents the deposition depth of the ash accumulation layer in the direction perpendicular to the pipe wall at that location.

[0037] Finally, the vertical distance between each pixel is multiplied by the local scale factor to obtain a series of local gray thickness values.

[0038] Given that the calculated dust accumulation thickness may be affected by edge detection errors, statistical analysis is performed on the obtained local dust accumulation thickness data to extract the median thickness data as the visual dust accumulation thickness, thereby suppressing the interference of abnormal measurements on the overall assessment.

[0039] In a further specific embodiment, the pulse ash thickness analysis process is as follows: When a pulse signal is emitted to the bottom of the pipe, the signal will encounter the surface of the ash layer in the propagation path and generate a reflected echo. At this time, the time interval between the signal emission time and the time of receiving the reflected echo from the ash surface is extracted from the collected pulse echo signal sequence at the bottom of the pipe.

[0040] Based on the preset propagation reference velocity of the pulse signal in the pipeline medium, the time interval is converted into the unidirectional propagation distance from the pulse probe to the ash accumulation surface, which is used as the ash accumulation measurement distance. The specific calculation expression is as follows: ,in This represents the reference speed at which the pulse signal propagates in the pipeline medium. This represents the time interval between the pulse signal transmission time and the time of the echo reflected from the dust-covered surface. This is because the round-trip distance between the pulse signal transmission time and the echo reflected from the dust-covered surface needs to be converted into a one-way distance.

[0041] Obtain the reference distance from the pulse signal to the bottom of the pipe at the same location under the condition of no dust accumulation.

[0042] Since the ash layer is deposited on the bottom surface of the pipe, the reflection interface of the pulse signal moves from the bottom of the pipe to the ash surface. Therefore, in the presence of ash, the measured one-way distance from the probe to the reflecting surface is less than the reference distance in the absence of ash. At this time, the difference between the reference distance and the ash measurement distance is taken as the pulse ash thickness.

[0043] After obtaining the dust accumulation thickness based on visual imaging and the dust accumulation thickness based on pulse echo ranging respectively, the present invention further performs a consistency check of the multi-source thickness. The specific implementation process is as follows: (1) Calculate the absolute deviation between the visual dust accumulation thickness and the pulse dust accumulation thickness.

[0044] (2) If the absolute deviation is less than or equal to the thickness tolerance threshold (the thickness tolerance threshold reflects the acceptable reasonable difference between the measurement results of the two types of sensors, usually 3 to 5 mm), it indicates that the two source data are mutually verified. At this time, the arithmetic mean of the visual dust accumulation thickness and the pulse dust accumulation thickness is taken as the comprehensive dust accumulation thickness. Otherwise, it indicates that at least one sensor is interfered with, resulting in unreliable data. At this time, the sliding average of the most recent consecutive valid historical dust accumulation thickness values ​​is taken as the current comprehensive dust accumulation thickness. The valid historical dust accumulation thickness value refers to the dust accumulation thickness that has passed the consistency check before. This operation can maintain the temporal continuity of dust accumulation thickness perception.

[0045] Given that the robot's perception of the bottom of the pipe through visual imaging and pulse echo during its movement only covers a local area within the sensor's field of view, and that the actual ash distribution inside the pipe is usually non-uniform, it is difficult to accurately characterize the overall ash morphology by relying solely on these sparse sampling points. Therefore, after multi-source consistency verification, a thermal map of the ash thickness distribution at the bottom of the pipe is generated by spatial interpolation based on the comprehensive ash thickness sequence continuously collected by the robot in the direction of movement.

[0046] In a preferred embodiment, the process of generating a heat map of the ash thickness distribution at the bottom of the pipe by spatial interpolation is as follows: First, a three-point moving average is performed on the comprehensive ash thickness sequence after multi-source consistency verification to suppress residual noise.

[0047] Subsequently, using the robot's actual axial position as a node, natural cubic spline interpolation is constructed to continuously reconstruct the entire detection interval.

[0048] Finally, a heat map of ash thickness distribution is generated by sampling and interpolating the results at a fixed step size, such as every 10 mm.

[0049] The pipe diameter status recognition module is used to acquire real-time pressure signals from multiple pressure feedback elements in the robot's walking mechanism while sensing the thickness of the ash accumulation, so as to identify the current pipe diameter status of the robot.

[0050] After sensing the thickness of the accumulated dust, the pipe diameter status needs to be identified simultaneously. If the robot performs the dust cleaning action without adapting to the pipe diameter, it is very easy to cause the support to become unstable and the mechanism to jam.

[0051] Based on this, the present invention uses multi-point pressure feedback elements distributed around the circumference of the robot housing to identify the pipe diameter status. The reason is that when the robot moves in the pipe and actively supports the pipe wall, the real-time contact pressure distribution measured by each pressure feedback element directly reflects the geometric matching relationship between the robot and the pipe wall.

[0052] See Figure 2 As shown, one possible approach for the above module is to identify the pipe diameter status as follows: For the real-time pressure values ​​of multiple pressure feedback elements within the same sampling period, calculate the correlation coefficient between any two different element signals. For example, the correlation coefficient can be the Pearson correlation coefficient.

[0053] The purpose of introducing the correlation coefficient is to quantify the synchronicity and coupling degree of the force changes at each support point. In a uniform circular pipe, all support points should respond synchronously to the pipe diameter change, showing a highly consistent correlation. However, in asymmetric or non-uniform structures, the correlation will differentiate or be partially decoupled. All the correlation coefficients between pairs are used to form a real-time pressure correlation coefficient matrix, which reflects the linear dependence between each pressure channel at the current moment.

[0054] Calculate the mean and standard deviation of all off-diagonal elements in the real-time pressure correlation coefficient matrix. The mean of the correlation coefficient reflects the strength of the overall correlation between the pressure signals and the synchronicity level of the robot's circumferential force response. The standard deviation of the correlation coefficient reflects the dispersion of the correlation between the pressure signals and the uniformity of the force distribution.

[0055] Since the correlation coefficient ranges from -1 to 1, the closer its value is to 1, the stronger the positive correlation. Based on this, if the average value of the correlation coefficient approaches 1 and the standard deviation approaches 0, it indicates that all pressure signals are strongly positively correlated with each other and the correlation strength is uniform. This corresponds to the robot being subjected to uniform force, and at this time it is determined that the robot is currently in the standard pipe diameter section.

[0056] Conversely, it is determined that the robot is currently in a non-standard pipe diameter section.

[0057] Furthermore, the non-standard pipe diameter sections of the present invention include, but are not limited to, pipe diameter variation sections and pipe cross-section deformation states. The pipe diameter variation section refers to the continuous and gradual expansion or contraction of the pipe diameter along the axial direction, such as the transition section. The pipe cross-section deformation state refers to the loss of circular symmetry of the pipe cross-section due to external force or manufacturing defects, resulting in non-circular geometric shapes such as ellipticization and flattening. Both of these states require adjustment of the robot's walking state to adapt to the pipe diameter.

[0058] The concurrent condition judgment module is used to determine whether the dust removal requirement and the pipe diameter adaptation requirement are concurrent based on the dust accumulation thickness distribution state and the pipe diameter state. The specific judgment process is as follows: the dust accumulation thickness distribution heat map is discretized into a thickness sequence along the pipe axis.

[0059] The dust accumulation thickness at each point in the thickness sequence is compared with the dust removal trigger value, where the dust removal trigger value refers to the minimum dust accumulation thickness required to trigger the dust removal action. It is usually set to 5 mm. In large-diameter pipes, the airflow carrying capacity is strong, and small dust accumulations are easily removed naturally. To avoid frequent and inefficient dust removal, the value can be appropriately increased. In small-diameter pipes, dust accumulation is more likely to cause blockages, so the value can be appropriately decreased. In this way, the locations that exceed the dust removal trigger value are selected as candidate points.

[0060] Based on spatial adjacency, consecutive candidate points are aggregated into connected regions.

[0061] For each connected region, calculate its physical extension length along the pipe axis. If the physical extension length of a certain connected region reaches the effective length, it is determined that there is a need for dust removal, and the region is marked as a dust accumulation region.

[0062] The effective length mentioned above refers to the smallest dust accumulation area that the dust removal structure can effectively operate on. For example, the physical coverage width of the dust removal brush can be used as the effective length. By introducing the effective length, dust accumulation areas that are too small can be filtered out, thereby avoiding ineffective operations and saving energy.

[0063] If the pipe diameter status recognition outputs that the robot is currently in a non-standard pipe diameter section, it is determined that there is a need for pipe diameter adaptation.

[0064] When the above two requirements are met simultaneously within the same time window, it is determined that the dust removal requirement and the pipe diameter adaptation requirement are concurrent.

[0065] The collaborative decision-making module is used to define the operation window based on the robot's current position when dust removal and pipe diameter adaptation requirements occur simultaneously. It then determines the order of operation execution based on the positional relationship between the dust accumulation area, the operation window, and the current pipe diameter segment, and executes the operation according to the decision instructions.

[0066] When both dust removal and pipe diameter adaptation requirements are triggered, the robot faces resource competition under multiple concurrent tasks. If it follows a fixed program, it may be easy to force dust removal in a bumpy section with changing diameter, resulting in poor fit of the dust removal structure and poor dust removal effect.

[0067] To address the aforementioned issues, this invention uses the robot's current position as a reference to define a work window. Under this spatial constraint, based on the spatial relationship between the dust accumulation area and the current pipe diameter segment and the work window, collaborative task scheduling with priority and execution sequence is performed.

[0068] The following description is based on a preferred embodiment: S1. Due to the limited physical size and range of motion of the cleaning structure, it can only act on the pipe wall within a certain axial range near the robot body. Based on this, taking the current position of the robot as a reference point, it extends forward and backward along the pipe axis to form a working window. This working window represents the spatial range in which the robot can perform cleaning operations under the current pose. It is a spatial constraint for task scheduling. If this constraint is lacking, it is easy to accidentally trigger the cleaning command for the dust accumulation area that exceeds the range of action of the mechanism.

[0069] S2. Compare the axial boundary contour of the ash accumulation area (the start and end positions along the pipe axis) with the working window range to determine whether the ash accumulation area falls at least partially within the working window. Specifically, this can be determined by intersecting the axial boundary contour of the ash accumulation area with the working window range. If there is an intersection, it means that the ash accumulation area falls at least partially within the working window, and then proceed to S21 or S22; otherwise, proceed to S23.

[0070] S21. If the dust accumulation area falls at least partially within the operation window, it is further determined whether the dust accumulation area is located within the current pipe diameter section. Since both dust removal and pipe diameter adaptation requirements are triggered, the current pipe diameter section is a non-standard pipe diameter section. If it is located within the current pipe diameter section, since the execution of the dust removal structure depends on the robot's support stability within the pipe, if dust removal is forcibly performed in a non-standard pipe diameter section that has not been adapted, it is easy to cause poor fit and incomplete dust removal. In this case, pipe diameter adaptation is performed first, and dust removal is performed after the robot adapts to the pipe diameter.

[0071] S22. If it is not located within the current pipe diameter segment, it indicates that the ash accumulation area may be located in the next pipe diameter segment. At this time, the robot is still within the current pipe diameter segment. The present invention further calculates the axial distance from the robot's current position to the axial boundary of the ash accumulation area. The purpose is to assess whether the ash accumulation area is within the range that the cleaning structure can safely cover in the unadjusted state. The specific implementation is as follows: First, based on the robot's real-time travel position in the pipe, its axial coordinates in the pipe coordinate system are obtained. The pipe coordinate system is a one-dimensional axial coordinate system. The fixed origin set at the pipe inlet is used as the reference, and the direction from the inlet to the outlet along the pipe centerline is defined as the positive axial direction.

[0072] Then, the axial boundary of the ash accumulation area is extracted from the ash thickness distribution heat map, and the starting position coordinates of the axial boundary are located.

[0073] Finally, the distance between the robot's axial coordinate and the starting coordinate of the axial boundary of the dust accumulation area is calculated and used as the axial distance.

[0074] Compare the axial distance with the allowable cleaning distance, which represents the maximum forward distance that the front end of the cleaning mechanism can be safely disturbed, for example, determined by the overhang length of the cleaning brush.

[0075] If the axial distance is less than or equal to the allowable cleaning distance, it indicates that the robot's current cleaning structure can reach the next pipe diameter section and clean the accumulated dust. For example, if there is accumulated dust 40mm ahead, and the cleaning brush can extend 60mm, it means that the cleaning requirements can be met under the current configuration. If pipe diameter adaptation is performed first, the robot needs to retract the support mechanism, travel to the dust accumulation area, and then extend for cleaning, resulting in two deployments / retractions, which is time-consuming and energy-intensive. In this case, the cleaning operation should be performed first, using the current configuration to complete the dust removal in one go, and then pipe diameter adaptation should be performed, thereby improving work efficiency.

[0076] Conversely, if the axial distance is greater than the allowable distance for dust removal, it indicates that the dust accumulation area is beyond the safe forward range of the dust removal mechanism. The robot cannot clean this area without adjusting the pipe diameter. Therefore, the current pipe diameter section should be adapted first, and the dust removal operation should be performed after the robot has traveled to the dust accumulation area.

[0077] S23. If the dust accumulation area falls outside the work window, it means that the dust removal structure cannot reach the area at present. At this time, the pipe diameter is directly adapted, and the dust removal task is re-evaluated when the robot moves to the area and enters the work window.

[0078] When faced with the dual tasks of pipe diameter adaptation and dust removal, this invention defines the operation window based on the robot's current position and makes operation sequence decisions based on the positional relationship between the dust accumulation area, the operation window, and the current pipe diameter segment. This decouples and schedules multiple task requirements in an orderly manner, ensuring that in complex pipeline environments, priority is given to ensuring robot pipe diameter adaptation, and dust removal is completed in advance when conditions permit, thereby improving the autonomy, safety, and continuity of the operation.

[0079] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product.

[0080] Those skilled in the art will recognize that the modules of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0081] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0082] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent dust removal robot with adaptive pipe diameter in dust removal pipelines, characterized in that, include: Dust accumulation thickness sensing module: Processes visual images and pulse echo signal sequences of the bottom of the pipe collected during the robot's movement in real time. By analyzing the visual dust accumulation thickness and pulse dust accumulation thickness, and fusing the two for consistency verification, a heat map of dust accumulation thickness distribution at the bottom of the pipe is generated. Pipe diameter status recognition module: While sensing the thickness of dust accumulation, it acquires real-time pressure signals from multiple pressure feedback elements in the robot's walking mechanism to identify the current pipe diameter status of the robot. Concurrency condition judgment module: Based on the heat map of ash thickness distribution and pipe diameter status, determine whether the ash cleaning requirement and the pipe diameter adaptation requirement are concurrent; Decision execution module: When dust removal and pipe diameter adaptation are required concurrently, the module defines the operation window based on the robot's current position, determines the order of operation execution based on the positional relationship between the dust accumulation area, the operation window, and the current pipe diameter segment, and executes the operation according to the decision instructions. The visual dust accumulation thickness analysis process is as follows: Image enhancement and edge detection processing are performed on the collected visual images of the bottom of the pipe to identify the contour lines of the inner wall of the pipe and the contour lines of the dust-accumulated surface in the image; A visual reference marker of known size is set at a fixed position inside the pipe. The pixel size of the reference marker is detected in the visual image, and the local scale factor under the current imaging conditions is calculated by combining its actual physical size. For each pixel on the contour line of the dust accumulation surface, match its corresponding inner wall contour point in the direction perpendicular to the pipe wall, and calculate the vertical distance between them. Multiply the vertical distance of each pixel by the local scale factor to obtain a series of local dust accumulation thickness values; Statistical analysis was performed on the obtained local dust accumulation thickness data, and the median thickness data was extracted as the visual dust accumulation thickness. The pulse ash accumulation thickness analysis process is as follows: For the collected pulse echo signal sequence at the bottom of the pipe, the time interval between the signal transmission time and the time of receiving the reflected echo from the ash-covered surface is extracted; Based on the preset propagation reference speed of the pulse signal in the pipeline medium, the time interval is converted into the unidirectional propagation distance from the pulse probe to the ash accumulation surface, which is used as the ash accumulation measurement distance. Obtain the reference distance from the pulse signal to the bottom of the pipe at the same location under the condition of no ash accumulation; The difference between the reference distance and the measured distance of ash accumulation is taken as the pulse ash accumulation thickness. The process of fusing the two data points and performing a consistency check generates a thermal map of the ash accumulation thickness distribution at the bottom of the pipe, including the following: Calculate the absolute deviation between the visual dust accumulation thickness and the pulse dust accumulation thickness; If the absolute deviation is less than or equal to the thickness tolerance threshold, the arithmetic mean of the visual dust accumulation thickness and the pulse dust accumulation thickness is taken as the comprehensive dust accumulation thickness; otherwise, the sliding average of the most recent consecutive effective historical dust accumulation thickness values ​​is taken as the comprehensive dust accumulation thickness. Based on the comprehensive ash thickness sequence continuously collected by the robot in the direction of travel, a thermal map of ash thickness distribution at the bottom of the pipe is generated by spatial interpolation.

2. The intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline as described in claim 1, characterized in that: The current pipe diameter status of the identification robot is as follows: For the real-time pressure values ​​of multiple pressure feedback elements in the same sampling period, calculate the correlation coefficient between any two different element signals, and construct a real-time pressure correlation coefficient matrix by combining all pairwise correlation coefficients. Calculate the mean and standard deviation of all off-diagonal elements in the real-time pressure correlation coefficient matrix; If the average value of the correlation coefficient approaches 1 and the standard deviation approaches 0, the robot is determined to be in the standard pipe diameter section; otherwise, the robot is determined to be in the non-standard pipe diameter section.

3. The intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline as described in claim 2, characterized in that: The determination of whether the dust removal requirement and the pipe diameter adaptation requirement occur concurrently includes the following: Assess the heat map of the ash thickness distribution at the bottom of the pipe. If the ash thickness in a certain area exceeds the ash removal trigger value, it is determined that there is a need for ash removal and the area is marked as an ash accumulation area. If the pipe diameter status recognition outputs that the robot is currently in a non-standard pipe diameter section, it is determined that there is a need for pipe diameter adaptation. When the above two requirements are met simultaneously within the same time window, it is determined that the dust removal requirement and the pipe diameter adaptation requirement are concurrent.

4. The intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline as described in claim 3, characterized in that: If the thickness of accumulated dust in a certain area exceeds the dust removal trigger value, the following judgment process is performed: The heat map of ash thickness distribution is discretized into a thickness sequence along the pipe axis; The dust accumulation thickness at each point in the thickness sequence is compared with the dust removal trigger value, thereby filtering out the locations that exceed the dust removal trigger value as candidate points. Based on spatial adjacency, consecutive candidate points are aggregated into connected regions; Calculate the physical extension length along the pipe axis for each connected region; If the physical extension length of a certain connected region reaches the effective length, then the dust accumulation thickness of that region is identified as exceeding the dust removal trigger value.

5. The intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline as described in claim 1, characterized in that: The operation window is defined based on the robot's current position as follows: Using the robot's current position as a reference point, extend a certain distance forward and backward along the pipeline axis to form a working window.

6. The intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline as described in claim 1, characterized in that: The process of determining the order of operations by combining the positional relationship between the ash accumulation area, the work window, and the current pipe diameter section includes the following steps: Compare the axial boundary contour of the dust accumulation area with the range of the working window to determine whether the dust accumulation area falls at least partially within the working window; If the dust accumulation area falls at least partially within the working window, it is further determined whether the dust accumulation area is located within the current pipe diameter section. If it is located within the current pipe diameter section, pipe diameter adaptation is performed first, and dust removal is performed after the robot adapts to the pipe diameter. If the robot is not located within the current pipe diameter section, calculate the axial distance from the robot's current position to the axial boundary of the dust accumulation area, compare the axial distance with the allowable dust removal distance, and if the axial distance is less than or equal to the allowable dust removal distance, prioritize dust removal and adapt the pipe diameter after dust removal is completed; otherwise, first complete the adaptation of the current pipe diameter section, and then perform the dust removal operation after moving to the dust accumulation area. If the dust accumulation area falls outside the work window, then the pipe diameter should be adjusted directly.

7. The intelligent dust removal robot with adaptive pipe diameter in a dust removal pipeline as described in claim 6, characterized in that: The axial distance from the current position of the robot to the axial boundary of the dust accumulation area is calculated as follows: Based on the robot's real-time position within the pipeline, its axial coordinates in the pipeline coordinate system are obtained. The pipeline coordinate system is a one-dimensional axial coordinate system, with a fixed origin set at the pipeline inlet as the reference. The direction from the inlet to the outlet along the pipeline centerline is defined as the positive axial direction. Extract the axial boundary of the ash accumulation area from the ash thickness distribution heat map, and locate the coordinates of the starting position of the axial boundary; Calculate the distance between the axial coordinate of the robot's travel position and the starting position coordinate of the axial boundary of the dust accumulation area, and use this distance as the axial distance.

Images