Pipeline detection method, device and equipment based on software robot and storage medium

Abstract

The application discloses a pipeline detection method, device and equipment based on a soft robot and a medium. The method comprises the following steps: obtaining relative position information of the soft robot in a target pipeline, selecting a target path point from each global path reference point in a pre-generated global reference path according to the relative position information of the soft robot, and generating a current feedforward control instruction according to the relative position information and the target path point; determining a target distance between the soft robot and the pipeline wall of the target pipeline according to the relative position information; if the target distance is less than a preset distance threshold, a current feedback control instruction is generated, and a target control instruction is generated according to the current feedforward control instruction and the current feedback control instruction; and the target control instruction is sent to the soft robot, so that the soft robot performs pipeline detection on the target pipeline based on the target control instruction. The technical scheme of the embodiment of the application can improve the safety of the soft robot in performing the pipeline detection task.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, and in particular to a pipeline detection method, apparatus, equipment and medium based on a soft robot. Background Technology

[0002] In the field of pipeline detection technology, the precise motion control and safe detection of soft robots are the core links to ensure the quality and improve the efficiency of pipeline inspection. They are directly related to the comprehensiveness and safety of pipeline inspection and are of great significance for the investigation, maintenance and repair of various types of pipelines such as industrial pipelines and municipal pipelines.

[0003] Current control methods for soft robot pipeline detection often employ single feedforward control or simple feedback control, lacking collaborative analysis of the robot's real-time relative position and distance to the pipe wall. They fail to integrate with a pre-generated global reference path for precise path guidance and dynamic distance control, making collisions between the soft robot's trajectory and the pipe wall common. This results in poor safety during pipeline detection and fails to provide efficient and reliable technical support for non-destructive testing in various complex pipeline environments. Therefore, there is an urgent need for a technical solution that enables precise motion control and safe pipeline detection for soft robots, addressing the issues of poor safety and collision susceptibility associated with traditional control methods. Summary of the Invention

[0004] This invention provides a pipeline detection method, apparatus, equipment, and medium based on a soft robot, to improve the safety of pipeline detection by the soft robot.

[0005] According to one aspect of the present invention, a pipe detection method based on a soft robot is provided, the method comprising:

[0006] The relative position information of the soft robot in the target pipeline is obtained, and the target path point is selected from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot. The current feedforward control command is generated according to the relative position information and the target path point.

[0007] The target distance between the soft robot and the pipe wall of the target pipe is determined based on the relative position information;

[0008] If the target distance is less than a preset distance threshold, a current feedback control command is generated, and a target control command is generated based on the current feedforward control command and the current feedback control command.

[0009] The target control command is sent to the soft robot so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

[0010] According to another aspect of the present invention, a pipe detection device based on a soft robot is provided, the device comprising:

[0011] The feedforward control command generation module is used to obtain the relative position information of the soft robot in the target pipeline, select the target path point from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot, and generate the current feedforward control command according to the relative position information and the target path point.

[0012] The target distance determination module is used to determine the target distance between the soft robot and the pipe wall of the target pipe based on the relative position information.

[0013] The target control command generation module is used to determine if the target distance is less than a preset distance threshold, generate a current feedback control command, and generate a target control command based on the current feedforward control command and the current feedback control command.

[0014] The pipeline detection module is used to send the target control command to the soft robot, so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory that is communicatively connected to at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the soft robot-based pipe detection method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the pipe detection method based on a soft robot according to any embodiment of the present invention.

[0020] This invention provides a technical solution that acquires the relative position information of a soft robot on a target pipeline, selects a target path point from pre-generated global reference paths based on the robot's relative position information, and generates a current feedforward control command based on the relative position information and the target path point. It then determines the target distance between the soft robot and the pipeline wall based on the relative position information. If the target distance is less than a preset distance threshold, a current feedback control command is generated, and a target control command is generated based on the current feedforward and feedback control commands. The target control command is then sent to the soft robot, allowing it to perform pipeline detection based on the target control command. This method can collaboratively generate feedforward and feedback control commands based on the global reference path and the real-time relative position information of the soft robot, achieving precise control of the soft robot's motion posture and real-time assurance of safe distance from the pipeline wall. This effectively solves the problems of lag in motion control, easy collision with the pipeline wall, and poor passage in narrow areas inherent in traditional pipeline detection robots, improving the stability and safety of the pipeline internal detection process and providing reliable technical support for non-destructive testing and refined inspection in complex pipeline environments.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a pipeline detection method based on a soft robot according to Embodiment 1 of the present invention;

[0024] Figure 2 This is a flowchart of a pipeline detection method based on a soft robot according to Embodiment 2 of the present invention;

[0025] Figure 3 This is a flowchart of a pipeline detection method based on a soft robot according to Embodiment 3 of the present invention;

[0026] Figure 4 This is a schematic diagram of a pipe detection device based on a soft robot according to Embodiment 4 of the present invention;

[0027] Figure 5This is a schematic diagram of the structure of an electronic device that implements a pipeline detection method based on a soft robot according to an embodiment of the present invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Example 1

[0031] Figure 1 This is a flowchart of a pipeline detection method based on a soft robot, provided in Embodiment 1 of the present invention. This embodiment is applicable to pipeline inspection scenarios in the field of pipeline detection technology and can be implemented by an organization performing pipeline inspection. The organization's control terminal acquires the position data of the soft robot and the point cloud data of the pipeline environment in real time, thereby achieving precise motion control of the soft robot and pipeline detection. This method can be executed by a pipeline detection device based on a soft robot, which can be implemented in hardware and / or software. This soft robot-based pipeline detection device can be configured in a server that uses a soft robot for pipeline monitoring. Figure 1 As shown, the method includes:

[0032] S101. Obtain the relative position information of the soft robot in the target pipeline, select the target path point from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot, and generate the current feedforward control command according to the relative position information and the target path point.

[0033] S102. Determine the target distance between the soft robot and the pipe wall of the target pipe based on the relative position information.

[0034] S103. If the target distance is less than the preset distance threshold, generate the current feedback control command and generate the target control command based on the current feedforward control command and the current feedback control command.

[0035] S104. Send the target control command to the soft robot so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

[0036] The soft robot can be a pneumatically driven soft robot used for pipeline inspection, featuring multiple pneumatically driven chambers, autonomous deformation capabilities, and the ability to move and inspect within complex pipelines. The target pipeline is the pipeline to be inspected. The relative position information of the soft robot is its real-time three-dimensional coordinates within the target pipeline. The global reference path is a motion path generated based on the target pipeline for the soft robot's movement. The target path point is the next path point the soft robot will reach within the global reference path. The current feedforward control command is a command calculated based on the global reference path and the soft robot's relative position information to control the soft robot's movement.

[0037] For example, the positioning module acquires the three-dimensional coordinates (x, y, z) of the soft robot inside the target pipe in real time. This position information is then compared with each of the pre-generated global path reference points using Euclidean distance calculations. The global path reference point with the smallest distance and located in the robot's forward direction is selected as the target path point. Based on the soft robot's kinematic model, the position and orientation of the target path point are converted into theoretical pressure values ​​for each drive chamber. A second-order Butterworth low-pass filter is used to smooth the pressure command sequence to obtain the current feedforward control command. For instance, if the soft robot's real-time position is (150.2 mm, 60.5 mm, 40.1 mm) and the nearest next path point in the global reference path is (160.3 mm, 61.2 mm, 40.0 mm), the current feedforward control command can be obtained by solving the robot's kinematic model. The robot kinematic model can be a model that calculates the pressure required for the multiple pneumatic drive chambers of the soft robot, given the known target position the soft robot needs to reach.

[0038] The target pipe wall can be the inner wall of the target pipe, which is the interface that the soft robot needs to avoid colliding with during its movement inside the pipe. The target distance can be the shortest three-dimensional straight-line distance between the relative position information of the soft robot and the pipe wall of the target pipe.

[0039] For example, based on the real-time three-dimensional coordinates (x, y, z) output by the soft robot, the pre-processed pipe wall point cloud data is read, the three-dimensional coordinates (x1, y1, z1) of all points in the pipe wall point cloud are traversed, and the Euclidean distance from the real-time position of the soft robot to each pipe wall point is calculated. The minimum value among all the calculation results is determined as the target distance between the soft robot and the pipe wall of the target pipe.

[0040] Furthermore, to improve the accuracy of target distance calculation, in one optional embodiment, determining the target distance between the soft robot and the pipe wall of the target pipe based on relative position information includes:

[0041] Step a1: Obtain the point cloud data of the pipe wall of the target pipe.

[0042] Step a2: Based on the relative position information and the pipe wall point cloud data, determine the target point cloud position information.

[0043] Step a3: Based on the relative position information and the target point cloud position information, determine the target distance between the soft robot and the pipe wall of the target pipe.

[0044] Among them, the pipe wall point cloud data of the target pipeline can be a pure set of pipe inner wall points obtained by collecting the internal space of the target pipeline through three-dimensional lidar and then filtering and denoising through improved voxel grid.

[0045] For example, the original point cloud data inside the target pipe is obtained by 3D LiDAR, and downsampling and noise reduction are performed using a 5mm×5mm×5mm voxel size. The points on the inner wall of the pipe are segmented using a clustering segmentation algorithm, and the sediment is removed to obtain the pipe wall point cloud data of the target pipe.

[0046] Among them, the target point cloud location information can be the three-dimensional coordinate information of the pipe wall point cloud data that is closest to the real-time relative position of the soft robot in the pipe wall point cloud data.

[0047] For example, the real-time three-dimensional coordinates (x, y, z) of the soft robot are used as the reference point. All pipe wall points in the pipe wall point cloud data are traversed, and the three-dimensional distance from the reference point to each pipe wall point is calculated in turn. The coordinates of the pipe wall point cloud data with the smallest distance are determined as the target point cloud position information.

[0048] For example, the target distance between the soft robot and the pipe wall can be calculated by substituting the real-time coordinates (x, y, z) of the soft robot and the target point cloud position information (x1, y1, z1) into the three-dimensional distance calculation formula. For instance, if the current position of the soft robot is (150.2 mm, 60.5 mm, 40.1 mm), and the coordinates of the nearest point on the pipe wall of the target pipe are (150.2 mm, 85.3 mm, 40.1 mm), and the distance between the two points is 24.8 mm, then 24.8 mm can be used as the target distance between the soft robot and the pipe wall of the target pipe.

[0049] The above technical solution first acquires the pipe wall point cloud data of the target pipe, then searches for the nearest pipe wall point cloud data point and calculates the target distance between the soft robot and the pipe wall of the target pipe. This can accurately and stably determine the actual distance between the soft robot and the target pipe wall, thus improving the accuracy of target distance detection.

[0050] Furthermore, to avoid redundant calculations and improve the smoothness of the soft robot's detection of the target pipe, in an optional embodiment, after determining the target distance between the soft robot and the pipe wall of the target pipe, the following steps are also included:

[0051] If the target distance is not less than the preset distance threshold, the current feedforward control command will be determined as the target control command.

[0052] The preset distance threshold can be a minimum safe distance threshold pre-set by relevant technicians based on the inner diameter of the target pipe and the width of the soft robot structure. The current feedback control command can be an obstacle avoidance correction control command generated when the soft robot approaches the pipe wall, containing the pipe wall repulsion vector and the pipe centerline attraction vector. The target control command can be the final drive command obtained by adaptively fusing the current feedforward control command and the current feedback control command.

[0053] For example, when the target distance is greater than a preset distance threshold, it indicates that the soft robot's current position is within a safe distance range and there is no risk of colliding with the pipe wall. In this case, there is no need to initiate a feedback correction process, and the generated current feedforward control command can be directly used as the final target control command. For instance, if the preset distance threshold is 40mm and the current target distance is 45mm, the current target distance is greater than the preset distance threshold, so the current feedforward command can be used directly to control the movement of the soft robot.

[0054] The above technical solution reduces the computational load of instruction fusion, lowers the system's computational burden, and improves the efficiency of pipeline detection operations by directly using the current feedforward control command as the target control command under safe conditions.

[0055] For example, when the target distance is less than a preset distance threshold, it indicates that the soft robot is approaching the pipe wall, posing a collision risk. In this case, a feedback control strategy based on the artificial potential field method can be used. The repulsive force of the pipe wall is calculated based on the target distance, and the current feedback control command is obtained. The current feedforward control command and the current feedback control command are then fused to obtain the target control command. For instance, if the preset distance threshold is 40mm and the current target distance is 35mm, which is less than the preset distance threshold, a feedback control command needs to be generated to correct the soft robot's position, ensuring the safety of the soft robot's pipe detection. The driving pressure value corresponding to the current feedforward control command is 100kPa, and the driving pressure corresponding to the current feedback control command is 20kPa; therefore, the target control command = 100kPa + 20kPa = 120kPa, and this 120kPa is the final output target control command.

[0056] For example, target control commands are sent to the soft robot in the form of pressure signals. The soft robot adjusts the pressure of each drive chamber according to the commands, moves safely within the target pipeline, and completes the pipeline detection operation.

[0057] This invention provides a technical solution that acquires the relative position information of a soft robot on a target pipeline, selects a target path point from pre-generated global reference paths based on the robot's relative position information, and generates a current feedforward control command based on the relative position information and the target path point. It then determines the target distance between the soft robot and the pipeline wall based on the relative position information. If the target distance is less than a preset distance threshold, a current feedback control command is generated, and a target control command is generated based on the current feedforward and feedback control commands. The target control command is then sent to the soft robot, allowing it to perform pipeline detection based on the target control command. This method can collaboratively generate feedforward and feedback control commands based on the global reference path and the real-time relative position information of the soft robot, achieving precise control of the soft robot's motion posture and real-time assurance of safe distance from the pipeline wall. This effectively solves the problems of lag in motion control, easy collision with the pipeline wall, and poor passage in narrow areas inherent in traditional pipeline detection robots, improving the stability and safety of the pipeline internal detection process and providing reliable technical support for non-destructive testing and refined inspection in complex pipeline environments.

[0058] Example 2

[0059] Figure 2This is a flowchart of a pipeline detection method based on a soft robot provided in Embodiment 2 of the present invention. This embodiment optimizes and improves upon the above-mentioned technical solutions. The step "generating a target control command based on the current feedforward control command and the current feedback control command" is refined to "determining the current pipeline cross-section of the target pipeline based on the relative position information, generating feedforward control command weights and feedback control command weights based on the current pipeline cross-section, and generating a target control command based on the feedforward control command weights and feedback control command weights based on the current feedforward control command and the current feedback control command." This improves the target control command generation method.

[0060] It should be noted that for parts not described in detail in the embodiments of the present invention, please refer to the descriptions in other embodiments. For example... Figure 2 As shown, the method includes the following specific steps:

[0061] S201. Obtain the relative position information of the soft robot in the target pipeline, select the target path point from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot, and generate the current feedforward control command according to the relative position information and the target path point.

[0062] S202. Determine the target distance between the soft robot and the pipe wall of the target pipe based on the relative position information.

[0063] S203. If the target distance is less than the preset distance threshold, generate the current feedback control command and determine the current pipe cross-section of the target pipe based on the relative position information.

[0064] S204. Generate feedforward control command weights and feedback control command weights based on the current pipeline cross-section.

[0065] S205. Based on the current feedforward control command and the current feedback control command, and based on the weights of the feedforward control command and the feedback control command, generate the target control command.

[0066] S206. Send the target control command to the soft robot so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

[0067] Furthermore, to improve the smoothness of the global reference path and thus enhance the detection stability of the soft robot within complex pipes, in one optional embodiment, before acquiring the relative position information of the soft robot in the target pipe, the method further includes:

[0068] Step b1: Obtain the raw pipe point cloud data of the target pipe.

[0069] Step b2: Determine the central axis of the pipeline based on the original pipeline point cloud data.

[0070] Step b3: Determine the pipe cross-section in the narrow area based on the pipe's central axis, and determine the target narrow area of ​​the pipe based on the pipe cross-section in the narrow area.

[0071] Step b4: Construct a target optimization model based on the pipeline's central axis and the narrow target area of ​​the pipeline.

[0072] Step b5: Determine the global reference path based on the target optimization model.

[0073] The original pipeline point cloud data can be the original three-dimensional point set obtained by scanning the inside of the target pipeline with a three-dimensional LiDAR, which includes all spatial information such as the inner wall of the pipeline, the internal space, the deformable area, and obstacles.

[0074] For example, a 3D LiDAR is used to scan the target pipeline, collect complete spatial point cloud information inside the pipeline, and obtain the original pipeline point cloud data.

[0075] The pipeline centerline can be the pipeline centerline curve extracted from the original pipeline point cloud data using a cylindrical fitting algorithm based on random sampling consistency.

[0076] For example, after filtering and denoising the original pipeline point cloud data with a 5mm×5mm×5mm voxel grid, a random sampling consistency cylinder fitting algorithm is used to fit the pipeline cylinder model, and the center line of the cylinder can be extracted as the central axis of the target pipeline based on the pipeline cylinder model.

[0077] The narrow section of the pipe can be multiple pipe sections taken at fixed intervals along the central axis of the pipe. The target narrow area of ​​the pipe can be the section with the smallest minimum inscribed circle radius among the multiple pipe sections taken at fixed intervals along the central axis of the pipe, i.e., the narrowest pipe section.

[0078] For example, narrow pipe sections can be generated at 20mm intervals along the central axis of the pipe, the minimum inscribed circle radius of each narrow pipe section can be calculated, and the section with the smallest radius can be determined as the target narrow pipe area.

[0079] The objective optimization model can be a model used to solve for the optimal global reference path.

[0080] Furthermore, to improve the rationality and applicability of the target optimization model, in an optional embodiment, a target optimization model is constructed based on the pipeline's central axis and the narrow target area of ​​the pipeline, including:

[0081] Step c1: Construct the path length cost function based on the original pipeline point cloud data and the pipeline centerline.

[0082] Step c2: Construct the path smoothness cost function based on the pipeline's central axis.

[0083] Step c3: Construct a path safety cost function based on the narrow target area of ​​the pipeline and the point cloud data of the pipeline wall.

[0084] Step c4: Construct the target optimization model based on the path length cost function, path smoothness cost function, and path safety cost function.

[0085] The path length cost function can be derived by discretizing the global reference path into a series of path points and calculating the sum of the Euclidean distances between adjacent path points. This cost function represents the total path length. The path length cost function can be expressed as:

[0086]

[0087] Where i represents the i-th path point, and n represents the number of path points in the global reference path. Represents the three-dimensional spatial coordinates of the i-th path point. Represents the three-dimensional spatial coordinates of the (i+1)th path point. for The next path point in the global reference path.

[0088] The path smoothness cost function can be obtained by calculating the sum of squares of vector changes between consecutive path segments, representing the degree of path curvature and the magnitude of abrupt turning changes. The path smoothness cost function can be expressed as:

[0089]

[0090] in, Represents the three-dimensional spatial coordinates of the (i-1)th path point. for The previous path point in the global reference path.

[0091] The path safety cost function can be a cost function constructed based on the minimum distance from a path point to the pipe wall or obstacle; the closer the distance, the higher the cost. The path safety cost function can be expressed as:

[0092]

[0093] in, It is a very small positive number, used to prevent the denominator from being zero. This represents the minimum Euclidean distance from the i-th path point pi to the pipe wall.

[0094] For example, the objective optimization model can be represented as:

[0095]

[0096] Among them, ω1, ω2, and ω3 are weighting coefficients used to adjust the priority between different optimization objectives. ω1, ω2, and ω3 can be preset by relevant technical personnel. For example, in narrow, high-risk pipeline environments, the safety weight ω3 can be appropriately increased, and in long-distance detection tasks, the path length weight ω1 can be moderately increased.

[0097] The above technical solution, by constructing three types of cost functions—length, smoothness, and safety—and then weighting and fusing them, can take into account path efficiency, smoothness of movement, and safety of passage, and generate an optimal global reference path that is suitable for complex pipeline environments.

[0098] For example, any path that satisfies the target optimization model can be used as a global reference path.

[0099] Furthermore, to improve the quality of the global reference path, an adaptive particle swarm optimization algorithm can be used to generate the global reference path. In one optional embodiment, the global reference path is determined based on the target optimization model, including:

[0100] Step d1: Obtain the current iteration number and the target iteration number.

[0101] Step d2: Determine the iterative search weights based on the current iteration number and the target iteration number.

[0102] Step d3: Determine the current global reference path based on the iterative search weights and the target optimization model.

[0103] Step d4: If the current iteration count meets the preset termination condition, then the current global reference path is determined as the global reference path.

[0104] Here, the current iteration number can be the number of iterations that the adaptive particle swarm optimization algorithm has completed so far. The target iteration number can be the maximum allowed number of iterations preset by the adaptive particle swarm optimization algorithm.

[0105] For example, the adaptive particle swarm optimization algorithm starts from the first iteration. After each round of particle update and fitness calculation, the current iteration number is incremented by 1 until the target iteration number of 200 is reached.

[0106] The iterative search weights can be the inertia weights used in the adaptive particle swarm optimization algorithm. The iterative search weights are represented as follows:

[0107]

[0108] Where t is the current iteration number, The target number of iterations, and These are weight values ​​pre-set for the relevant technical personnel.

[0109] Here, the current global reference path can be the path corresponding to the particle with the best fitness value in each iteration. The fitness value can be a numerical score calculated based on the objective optimization model to evaluate the quality of the current global reference path.

[0110] For example, in each iteration, the velocity and position of all particles are updated using the current iteration's search weights. Then, the fitness value of each particle's corresponding path is calculated using the target optimization model. The path represented by the particle with the optimal fitness value is determined as the current global reference path. The particle's position can be a complete set of global reference paths represented by the particle, consisting of a series of three-dimensional path point coordinates. The particle's velocity can be the magnitude and direction of path adjustment during the search process, used to control how the particle's position is updated. The preset termination condition can be an iteration termination condition pre-set by relevant technical personnel, such as the current iteration count reaching the target iteration count.

[0111] For example, when the current iteration count reaches 200, the algorithm stops iterating and determines the current global reference path as the final global reference path.

[0112] The above technical solution, through adaptive particle swarm optimization, linearly decreasing inertia weights, and iterative optimization strategy, can quickly and stably converge to the optimal global reference path, thereby improving the accuracy and efficiency of global reference path planning.

[0113] The above technical solution, through point cloud acquisition, preprocessing, extraction of the pipeline's central axis, identification of narrow areas, construction of a multi-objective optimization model, and solution using an adaptive particle swarm optimization algorithm, can generate a safe and smooth global reference path that adapts to narrow pipelines, providing a reliable foundation for soft robots to accurately, stably, and safely complete target pipeline detection tasks.

[0114] The current cross-section of the target pipe can be the cross-section of the pipe that the soft robot cuts perpendicular to the central axis of the pipe based on the relative position information.

[0115] For example, if the relevant technicians preset the distance threshold to 40mm and the current target distance is 35mm, since the current target distance is less than the preset distance threshold, a feedback control strategy based on the artificial potential field method is needed to generate the current feedback control command. At the same time, based on the real-time relative position information of the soft robot and the central axis of the pipeline, a cross section perpendicular to the central axis of the pipeline and passing through the current position of the robot is intercepted, and this cross section is determined as the current pipeline cross section of the target pipeline.

[0116] The feedforward control command weight can be the proportion of the current feedforward control command in the target control command. The feedback control command weight can be the proportion of the current feedback control command in the target control command.

[0117] For example, the weights of the feedforward control command and the feedback control command can be dynamically adjusted based on the current pipe cross-section. For instance, if the current pipe cross-section is less than 40mm, the weight of the feedback control command is set to 0.7 and the weight of the feedforward control command is set to 0.3; if the current pipe cross-section is greater than 40mm, the weight of the feedforward control command is set to 0.6 and the weight of the feedback control command is set to 0.4.

[0118] Among them, the target control command can be obtained by weighting and summing the current feedforward control command and the current feedback control command according to the weight of the feedforward control command and the weight of the feedback control command.

[0119] For example, the weight of the feedforward control command is set to 0.3, the weight of the feedback control command is set to 0.7, the pressure value corresponding to the current feedforward control command is 100 kPa, and the pressure value corresponding to the current feedback control command is 20 kPa. First, a weighted summation calculation is performed: target control command = 100 kPa × 0.3 + 20 kPa × 0.7 = 44 kPa. Therefore, 44 kPa is directly determined as the target control command and output to control the movement of the soft robot.

[0120] The technical solution of this embodiment obtains the real-time relative position information of the soft robot in the target pipeline, matches the global reference path to generate feedforward control commands, calculates the target distance between the robot and the pipe wall in real time, dynamically allocates the weight of feedforward control commands and the weight of feedback control commands according to the current pipeline cross-section, and fuses them to generate target control commands, so as to realize the safe, accurate and stable movement of the soft robot in the pipeline. It can effectively solve the problems of control lag, easy collision and poor adaptability of traditional pipeline detection robots.

[0121] Example 3

[0122] Figure 3 This is a flowchart of a pipe detection method based on a soft robot, provided in Embodiment 3 of the present invention. This embodiment provides a preferred example based on the above embodiments. The method includes:

[0123] S301. Obtain the original pipe point cloud data inside the target pipe using a 3D LiDAR, and preprocess the original pipe point cloud data using an improved voxel grid filtering algorithm.

[0124] S302. A cylindrical fitting algorithm based on random sampling consistency is used to extract the central axis of the pipeline from the preprocessed pipeline point cloud data.

[0125] S303. Generate equally spaced analysis sections along the central axis of the pipeline, calculate the minimum inscribed circle radius of each section, and determine the narrow target area of ​​the pipeline.

[0126] S304. Based on the pipeline's central axis, the narrow target area of ​​the pipeline, and the original pipeline point cloud data, a multi-objective optimization model is constructed, which includes path length, smoothness, and safety indicators.

[0127] S305. Discretize the global reference path to be planned into a series of path points, and establish an objective function that is a weighted sum of path length cost, smoothness cost, and security cost.

[0128] S306. An adaptive particle swarm optimization algorithm is used to solve the target optimization model. An initial path population is generated near the central axis of the pipeline and the optimization is iteratively performed.

[0129] S307. Output a smooth curve path based on the iterative optimization results, and determine the global reference path and reference points for each global path.

[0130] S308. Obtain the relative position information of the soft robot in the target pipeline, select the target path point from the global reference path based on the relative position information, and generate the current feedforward control command.

[0131] S309. Determine the target distance between the soft robot and the pipe wall based on the relative position information.

[0132] S310. Determine whether the target distance is not less than the preset distance threshold. If yes, execute S311; otherwise, execute S312.

[0133] S311. Determine the current feedforward control command as the target control command.

[0134] S312. The current feedback control command is obtained by using the artificial potential field method. The current feedforward control command and the current feedback control command are fused to generate the target control command.

[0135] S313. Send the target control command to the soft robot so that the soft robot can complete the pipeline detection operation in the target pipeline based on the target control command.

[0136] The information collected in the above embodiments of the present invention is all information and data authorized by the user or fully authorized by all parties. The collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse.

[0137] Example 4

[0138] Figure 4 This is a schematic diagram of a pipe detection device based on a soft robot, provided in Embodiment 4 of the present invention. The pipe detection device based on a soft robot provided in this embodiment of the invention is applicable to pipe inspection scenarios in the field of pipe detection technology. This soft robot-based pipe detection device can be implemented in hardware and / or software, and can be applied to a pipe detection method based on a soft robot. Specifically, it can be configured in a server that uses a soft robot for pipe monitoring. Figure 4 As shown, the device includes: a feedforward control command generation module 401, a target distance determination module 402, a target control command generation module 403, and a pipeline detection module 404. Wherein:

[0139] The feedforward control command generation module 401 is used to obtain the relative position information of the soft robot in the target pipeline, select the target path point from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot, and generate the current feedforward control command according to the relative position information and the target path point.

[0140] The target distance determination module 402 is used to determine the target distance between the soft robot and the pipe wall of the target pipe based on the relative position information.

[0141] The target control command generation module 403 is used to determine if the target distance is less than a preset distance threshold, generate a current feedback control command, and generate a target control command based on the current feedforward control command and the current feedback control command.

[0142] The pipeline detection module 404 is used to send the target control command to the soft robot, so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

[0143] This invention provides a technical solution that acquires the relative position information of a soft robot on a target pipeline, selects a target path point from pre-generated global reference paths based on the robot's relative position information, and generates a current feedforward control command based on the relative position information and the target path point. It then determines the target distance between the soft robot and the pipeline wall based on the relative position information. If the target distance is less than a preset distance threshold, a current feedback control command is generated, and a target control command is generated based on the current feedforward and feedback control commands. The target control command is then sent to the soft robot, allowing it to perform pipeline detection based on the target control command. This method can collaboratively generate feedforward and feedback control commands based on the global reference path and the real-time relative position information of the soft robot, achieving precise control of the soft robot's motion posture and real-time assurance of safe distance from the pipeline wall. This effectively solves the problems of lag in motion control, easy collision with the pipeline wall, and poor passage in narrow areas inherent in traditional pipeline detection robots, improving the stability and safety of the pipeline internal detection process and providing reliable technical support for non-destructive testing and refined inspection in complex pipeline environments.

[0144] Optionally, the target control instruction generation module 403 is specifically used for:

[0145] Based on the relative position information, determine the current pipe cross-section of the target pipe;

[0146] Based on the current pipe cross-section, generate feedforward control command weights and feedback control command weights;

[0147] Based on the current feedforward control command and the current feedback control command, a target control command is generated according to the weights of the feedforward control command and the feedback control command.

[0148] Optionally, the device further includes:

[0149] The target control command determination module is used to determine the current feedforward control command as the target control command if the target distance is not less than a preset distance threshold after the target distance between the soft robot and the pipe wall of the target pipe is determined.

[0150] Optionally, the target distance determination module 402 is specifically used for:

[0151] Obtain the pipe wall point cloud data of the target pipe;

[0152] Based on the relative position information, the target point cloud position information is determined using the pipe wall point cloud data.

[0153] Based on the relative position information and the target point cloud position information, the target distance between the soft robot and the pipe wall of the target pipe is determined.

[0154] Optionally, the device further includes:

[0155] The point cloud data acquisition module is used to acquire the original point cloud data of the target pipe before acquiring the relative position information of the soft robot in the target pipe;

[0156] The center axis determination module is used to determine the center axis of the pipeline based on the original pipeline point cloud data.

[0157] The target area determination module is used to determine the pipe cross-section of the narrow area based on the pipe's central axis, and to determine the narrow target area of ​​the pipe based on the pipe cross-section of the narrow area.

[0158] An optimization model building module is used to build a target optimization model based on the central axis of the pipeline and the narrow target area of ​​the pipeline;

[0159] The global reference path determination module is used to determine the global reference path based on the target optimization model.

[0160] Optionally, optimize the model building module, specifically for:

[0161] Based on the original pipeline point cloud data and the pipeline centerline, construct the path length cost function;

[0162] Based on the central axis of the pipeline, construct a path smoothness cost function;

[0163] Based on the narrow target area of ​​the pipeline and the point cloud data of the pipeline wall, a path safety cost function is constructed;

[0164] Based on the path length cost function, path smoothness cost function, and path safety cost function, a target optimization model is constructed.

[0165] Optional, a global reference path determination module, specifically used for:

[0166] Get the current iteration number and the target iteration number;

[0167] Determine the iterative search weights based on the current iteration count and the target iteration count;

[0168] Based on the iterative search weights and the target optimization model, the current global reference path is determined;

[0169] If the current iteration count meets the preset termination condition, then the current global reference path is determined as the global reference path.

[0170] The pipe detection device based on a soft robot provided in this embodiment of the invention can execute a pipe detection method based on a soft robot provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0171] Example 5

[0172] Figure 5 A schematic diagram of an electronic device 50 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0173] like Figure 5 As shown, the electronic device 50 includes at least one processor 51 and a memory, such as a read-only memory (ROM) 52 and a random access memory (RAM) 53, communicatively connected to the at least one processor 51. The memory stores computer programs executable by the at least one processor. The processor 51 can perform various appropriate actions and processes based on the computer program stored in the ROM 52 or loaded from storage unit 58 into the RAM 53. The RAM 53 can also store various programs and data required for the operation of the electronic device 50. The processor 51, ROM 52, and RAM 53 are interconnected via a bus 54. An input / output (I / O) interface 55 is also connected to the bus 54.

[0174] Multiple components in electronic device 50 are connected to I / O interface 55, including: input unit 56, such as keyboard, mouse, etc.; output unit 57, such as various types of monitors, speakers, etc.; storage unit 58, such as disk, optical disk, etc.; and communication unit 59, such as network card, modem, wireless transceiver, etc. Communication unit 59 allows electronic device 50 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0175] Processor 51 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 51 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 51 performs the various methods and processes described above, such as a pipe detection method based on soft robotics.

[0176] In some embodiments, the soft robot-based pipe detection method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 58. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 50 via ROM 52 and / or communication unit 59. When the computer program is loaded into RAM 53 and executed by processor 51, one or more steps of the soft robot-based pipe detection method described above can be performed. Alternatively, in other embodiments, processor 51 can be configured for the soft robot-based pipe detection method by any other suitable means (e.g., by means of firmware).

[0177] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0178] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0179] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0180] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0181] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0182] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0183] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0184] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A pipe detection method based on a soft robot, characterized in that, include: The relative position information of the soft robot in the target pipeline is obtained, and the target path point is selected from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot. The current feedforward control command is generated according to the relative position information and the target path point. The target distance between the soft robot and the pipe wall of the target pipe is determined based on the relative position information; If the target distance is less than a preset distance threshold, a current feedback control command is generated, and a target control command is generated based on the current feedforward control command and the current feedback control command. The target control command is sent to the soft robot so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

2. The method according to claim 1, characterized in that, The step of generating a target control command based on the current feedforward control command and the current feedback control command includes: Based on the relative position information, determine the current pipe cross-section of the target pipe; Based on the current pipe cross-section, generate feedforward control command weights and feedback control command weights; Based on the current feedforward control command and the current feedback control command, a target control command is generated according to the weights of the feedforward control command and the feedback control command.

3. The method according to claim 1, characterized in that, After determining the target distance between the soft robot and the pipe wall of the target pipe, the method further includes: If the target distance is not less than a preset distance threshold, then the current feedforward control command is determined as the target control command.

4. The method according to claim 1, characterized in that, Determining the target distance between the soft robot and the pipe wall of the target pipe based on the relative position information includes: Obtain the pipe wall point cloud data of the target pipe; Based on the relative position information, the target point cloud position information is determined using the pipe wall point cloud data. Based on the relative position information and the target point cloud position information, the target distance between the soft robot and the pipe wall of the target pipe is determined.

5. The method according to claim 1, characterized in that, Before acquiring the relative position information of the soft robot in the target pipeline, the method further includes: Obtain the original point cloud data of the target pipeline; Based on the original pipeline point cloud data, the pipeline center axis is determined; Based on the central axis of the pipeline, determine the cross-section of the pipeline in the narrow area, and based on the cross-section of the pipeline in the narrow area, determine the target narrow area of ​​the pipeline; Based on the central axis of the pipeline and the narrow target area of ​​the pipeline, a target optimization model is constructed; The global reference path is determined based on the target optimization model.

6. The method according to claim 5, characterized in that, The step of constructing a target optimization model based on the pipeline's central axis and the narrow target area of ​​the pipeline includes: Based on the original pipeline point cloud data and the pipeline centerline, construct the path length cost function; Based on the central axis of the pipeline, construct a path smoothness cost function; Based on the narrow target area of ​​the pipeline and the point cloud data of the pipeline wall, a path safety cost function is constructed; Based on the path length cost function, path smoothness cost function, and path safety cost function, a target optimization model is constructed.

7. The method according to claim 5, characterized in that, Determining the global reference path based on the target optimization model includes: Get the current iteration number and the target iteration number; Determine the iterative search weights based on the current iteration count and the target iteration count; Based on the iterative search weights and the target optimization model, the current global reference path is determined; If the current iteration count meets the preset termination condition, then the current global reference path is determined as the global reference path.

8. A pipe detection device based on a soft robot, characterized in that, include: The feedforward control command generation module is used to obtain the relative position information of the soft robot in the target pipeline, select the target path point from each global path reference point in the pre-generated global reference path according to the relative position information of the soft robot, and generate the current feedforward control command according to the relative position information and the target path point. The target distance determination module is used to determine the target distance between the soft robot and the pipe wall of the target pipe based on the relative position information. The target control command generation module is used to determine if the target distance is less than a preset distance threshold, generate a current feedback control command, and generate a target control command based on the current feedforward control command and the current feedback control command. The pipeline detection module is used to send the target control command to the soft robot, so that the soft robot can perform pipeline detection on the target pipeline based on the target control command.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the pipe detection method based on a soft robot as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the pipe detection method based on a soft robot as described in any one of claims 1-7.

Images