Robot intelligent control method, system and device and medium

Abstract

The invention discloses a robot intelligent control method, system and device and a storage medium. The method comprises the steps of obtaining underwater environment information of an invisible environment and collecting current motion data of a robot; the underwater environment information comprises a distance metric map between the robot and an obstacle in an invisible environment; detecting corresponding line segments of the walls in the distance metric map to obtain a candidate line segment library of all the walls in the invisible environment; based on the candidate line segment library, static positioning information of the robot is determined; updating dynamic positioning information of the robot based on the current motion data; and based on the static positioning information or the dynamic positioning information, a robot fan-shaped dredging route is constructed, a corresponding robot moving instruction is generated, and the robot moving instruction is used for controlling the robot to execute automatic cruise in a fan-shaped motion mode. The method can be widely applied to the technical field of underwater robot control.

Description

Technical Field

[0001] This application relates to the field of underwater robot control technology, and in particular to a robot intelligent control method, system, device and storage medium. Background Technology

[0002] There is a significant demand for confined space dredging in industries such as municipal sanitation, construction, chemical, and power. While each industry has its own unique characteristics, all face substantial difficulties and challenges. Traditional manual dredging suffers from multiple challenges, including low efficiency, high labor intensity, safety hazards, inconsistent and incomplete dredging quality, and severe secondary environmental pollution. The dredging problem is particularly prominent and urgent in urban sewage pipe network booster pump stations.

[0003] Traditional dredging involves shutting down pumping stations and having dredging personnel descend into sewage pumping pits tens of meters deep to remove silt using sand pumps. This method is not only time-consuming and inefficient, but also poses significant safety risks. Furthermore, the shutdown of pumping stations can easily lead to sewage overflows into rivers and streams, polluting the aquatic environment. If dredging is not carried out for a long period, the accumulated silt will not only severely wear down the booster pumps but also reduce the pumping station's operational efficiency. Ensuring the safety of dredging personnel working in confined spaces, and determining whether dredging can be conducted without shutting down operations to prevent sewage overflows and river pollution, have become key challenges and pain points for pipeline operators.

[0004] Currently, underwater dredging robots, as an efficient dredging method, can quickly remove accumulated silt from sewage pumping stations while ensuring the safety of maintenance personnel. However, the underwater environment of sewage pumping stations is extremely complex and lacks visibility, as well as effective environmental perception and positioning methods. This poses a significant challenge to the automation of dredging robots, and the lack of effective environmental perception and positioning technology also creates potential hazards for the safe operation of the robots. Therefore, there are still technical problems that need to be solved in this field. Summary of the Invention

[0005] The purpose of this application is to at least partially solve one of the technical problems existing in the prior art.

[0006] Therefore, one objective of this application is to provide a robot intelligent control method, system, device, and storage medium, which can improve the stability and safety of the robot.

[0007] To achieve the above-mentioned technical objectives, the technical solution adopted in the embodiments of this application includes: a robot intelligent control method, comprising the following steps: Acquire underwater environmental information of the unseen environment and collect the robot's current motion data; the underwater environmental information includes a distance measurement map between the robot and obstacles in the unseen environment; The corresponding line segments of the walls in the distance metric map are detected to obtain a candidate line segment library for all walls in the invisible environment; Based on the candidate line segment library, the robot's static positioning information is determined; Based on the current motion data, update the robot's dynamic positioning information; Based on the static or dynamic positioning information, a fan-shaped dredging route for the robot is constructed and corresponding robot movement commands are generated. The robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion.

[0008] Furthermore, the candidate line segment library includes candidate sub-line segments of walls in different directions, and the determination of the robot's static positioning information based on the candidate line segment library includes: Based on the preset drawings of the aforementioned invisible environment, construct the robot's positioning frame; The positioning frame is matched with all the candidate sub-segments, and the matching degree between the positioning frame and each candidate sub-segment is calculated. Based on the matching degree, the static positioning information of the robot in the coordinate system of the invisible environment is determined.

[0009] Furthermore, the current motion data includes the robot's real-time turning angle, the robot's real-time displacement, the robot's initial scanning angle relative to the distance metric map at each motion time point, the robot's motion speed, and the robot's real-time displacement direction. Updating the robot's dynamic positioning information based on the current motion data specifically includes: Based on the real-time displacement direction and the movement speed, the extension direction of the positioning frame in the invisible environment coordinate system is determined; The robot position information is updated based on the real-time turning angle, the real-time displacement, the initial scanning angle, and the extension direction.

[0010] Further, the step of detecting the corresponding line segments of the walls in the distance metric map to obtain a candidate line segment library for all walls in the invisible environment specifically includes: The corresponding line segments of the wall in the distance measurement map are detected to obtain the first coordinate of each corresponding line segment and the second coordinate of the positioning frame; Based on the first coordinate and the second coordinate, determine each matching degree between each corresponding line segment and the positioning frame; All corresponding line segments with a matching degree greater than a preset threshold are selected as all candidate line segments in the candidate line segment library.

[0011] Further, determining the matching degree between each corresponding line segment and the positioning box based on the first coordinate and the second coordinate includes: By inputting the first coordinate and the second coordinate into the matching degree calculation formula, each matching degree between each corresponding line segment and the positioning box is obtained; The matching degree calculation formula is as follows: , in For the first The coordinates of the candidate lines, These are the linear coordinates of the positioning box.

[0012] Further, the step of constructing a robot fan-shaped dredging route and generating corresponding robot movement commands based on the static positioning information or the dynamic positioning information includes: The static positioning information or the dynamic positioning information is used to determine the vertices of the fan-shaped dredging route; With the robot's current orientation as the center line of the sector, set the angle and radius of the sector; Based on the fan-shaped dredging route constructed from the vertex, the angle, and the radius, robot movement commands containing the direction of movement, the distance of movement, and the timing of turning are generated.

[0013] Furthermore, the acquisition of underwater environmental information in the unseen environment also includes: By performing convolution operations on underwater environmental information using a Gaussian kernel function with a preset standard deviation, environmental noise and random errors from sensors in the underwater environmental information are filtered out.

[0014] On the other hand, embodiments of this application also provide a robot intelligent control system, including: The acquisition unit is used to acquire underwater environmental information of the invisible environment and collect the robot's current motion data; the underwater environmental information includes a distance measurement map between the robot and obstacles in the invisible environment; The first processing unit detects the corresponding line segments of the walls in the distance measurement map to obtain a candidate line segment library for all walls in the invisible environment. The second processing unit determines the robot's static positioning information based on the candidate line segment library; The third processing unit updates the robot's dynamic positioning information based on the current motion data; The fourth processing unit constructs a fan-shaped dredging route for the robot based on the static positioning information or the dynamic positioning information and generates corresponding robot movement commands. The robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion.

[0015] On the other hand, this application also provides a robot intelligent control device, including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a robot intelligent control method as described in any one of the inventions.

[0016] In addition, this application also provides a computer-readable storage medium storing processor-executable instructions, which, when executed by a processor, are used to perform a robot intelligent control method as described in any of the preceding claims.

[0017] The advantages and beneficial effects of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application: This application can improve the safety and stability of robots underwater by acquiring underwater environmental information from an unseen environment and collecting the robot's current motion data. The underwater environmental information includes a distance measurement map between the robot and obstacles in the unseen environment. Corresponding line segments of walls in the distance measurement map are detected to obtain a candidate line segment library for all walls in the unseen environment. Based on the candidate line segment library, the robot's static positioning information is determined. Based on the current motion data, the robot's dynamic positioning information is updated. Based on the static or dynamic positioning information, a fan-shaped dredging route for the robot is constructed, and corresponding robot movement commands are generated. These robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion. This application uses a novel algorithm to determine the robot's dynamic and static positioning information, and then determines the robot's fan-shaped dredging route and generates corresponding robot movement commands based on the dynamic or static positioning information. These robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion. This application can improve the robot's safety and stability underwater. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the steps of a robot intelligent control method according to a specific embodiment of the present invention; Figure 2 This is a schematic diagram of the steps of a robot intelligent control method according to another specific embodiment of the present invention; Figure 3 This is a result diagram of a denoising and data augmentation algorithm according to a specific embodiment of the present invention; Figure 4 This is a schematic diagram of a similarity-based localization algorithm according to a specific embodiment of the present invention; Figure 5 This is a schematic diagram of a fan-shaped dredging process according to a specific embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a robot intelligent control system according to a specific embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of a robot intelligent control device according to a specific embodiment of the present invention. Detailed Implementation

[0019] The following detailed description, in conjunction with the accompanying drawings, illustrates the principles and processes of the robot intelligent control method, system, device, and storage medium in the embodiments of the present invention.

[0020] There is a significant need for confined space dredging across various industries, each with its own unique characteristics, but all facing substantial difficulties and challenges. Traditional manual dredging presents multiple challenges, including low efficiency, high labor intensity, safety hazards, inconsistent and incomplete dredging quality, and severe secondary environmental pollution.

[0021] Traditional dredging in invisible environments requires dredging personnel to descend into sewage environments tens of meters deep and use sand pumps for dredging, which is not only time-consuming and inefficient, but also poses great safety risks.

[0022] Currently, underwater dredging robots, as a highly efficient dredging method, can quickly remove accumulated silt from invisible environments while ensuring the safety of maintenance personnel. However, due to the extreme complexity of the invisible environments in different areas and the lack of visibility, effective environmental perception and positioning methods are lacking. This poses a significant challenge to the automation of dredging robots, and the lack of effective environmental perception and positioning technology also brings potential hidden dangers to the safe operation of the robots.

[0023] Based on this, this invention discloses a map matching and localization method for sewage environments without visual visibility, and an automatic robot control method for underwater environments without visual visibility. This method uses a 360° surround-scanning sonar sensor to collect underwater environmental information of the sewage environment without visual visibility and converts it into a distance metric map in polar coordinates. After Gaussian filtering for noise reduction and contrast enhancement, a candidate line segment library of walls in the invisible environment is constructed using the Canny edge detection algorithm. A positioning bounding box is constructed by combining the map of the invisible environment with the positioning bounding box matching technology that integrates similarity metrics and real-time motion information to update the robot's positioning information under different working conditions. Finally, a fan-shaped dredging route is constructed based on the positioning results, and movement commands are generated to guide the robot to move back and forth along the edge of the fan and radially, completing automatic dredging. This invention can achieve precise robot positioning and automated cleaning navigation in underwater environments without visual visibility, improving dredging efficiency and positioning accuracy, and providing effective technical support for uninterrupted dredging in sewage environments without visual visibility, ensuring operational safety, and preventing sewage overflow. (Reference) Figure 1 , Figure 1 This is a schematic diagram illustrating the steps of a robot intelligent control method provided in an embodiment of this application. Figure 1In this process, the robot intelligent control method may include, but is not limited to, steps S101-S103.

[0024] S101. Acquire underwater environmental information from the unseen environment and collect the robot's current motion data. The underwater environmental information includes a distance measurement map between the robot and obstacles in the unseen environment.

[0025] S102. Detect the corresponding line segments of the walls in the distance metric map to obtain a candidate line segment library for all walls in the invisible environment.

[0026] S103. Based on the candidate line segment library, determine the robot's static positioning information.

[0027] S104. Update the robot's dynamic positioning information based on the current motion data.

[0028] S105. Based on static or dynamic positioning information, construct a fan-shaped dredging route for the robot and generate corresponding robot movement commands. The robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion.

[0029] Furthermore, the candidate line segment library includes candidate sub-line segments of the wall in different directions. Based on the candidate line segment library, the step of determining the static positioning information of the robot includes steps S201-S203.

[0030] S201. Construct the robot's positioning frame based on the preset drawings of the invisible environment.

[0031] S202. Match the positioning box with all candidate sub-segments and calculate the matching degree between the positioning box and each candidate sub-segment.

[0032] S203. Based on the matching degree, determine the static positioning information of the robot in the coordinate system of the invisible environment.

[0033] Furthermore, the current motion data includes the robot's real-time turning angle, the robot's real-time displacement, the robot's initial scanning angle of the distance metric map at each motion time point, the robot's motion speed, and the robot's real-time displacement direction. The step of updating the robot's dynamic positioning information based on the current motion data may include steps S301-S302.

[0034] S301. Based on the real-time displacement direction and motion speed, determine the extension direction of the positioning frame in the coordinate system of the invisible environment.

[0035] S302. Update the robot's position information based on the real-time turning angle, real-time displacement, initial scanning angle, and extension direction.

[0036] Furthermore, the step of detecting the corresponding line segments of the walls in the distance metric map to obtain a candidate line segment library for all walls in the invisible environment includes steps S401-S403.

[0037] S401. Detect the corresponding line segments of the wall in the distance measurement map to obtain the first coordinate of each corresponding line segment and the second coordinate of the positioning box.

[0038] S402. Based on the first coordinate and the second coordinate, determine each matching degree between each corresponding line segment and the positioning box.

[0039] S403. Select all corresponding line segments with a matching degree greater than the preset threshold as all candidate line segments in the candidate line segment library.

[0040] Furthermore, the step of determining each matching degree between each corresponding line segment and the positioning box based on the first coordinate and the second coordinate includes: Input the first and second coordinates into the matching degree calculation formula to obtain the matching degree between each corresponding line segment and the positioning box.

[0041] The formula for calculating the matching degree is: , in For the first The coordinates of the candidate lines, These are the linear coordinates of the positioning box.

[0042] Furthermore, the step of constructing a robot fan-shaped dredging route and generating corresponding robot movement commands based on static or dynamic positioning information includes steps S501-S503.

[0043] S501. Determine the static or dynamic positioning information as the vertices of the fan-shaped dredging route.

[0044] S502. Using the robot's current orientation as the center line of the sector, set the angle and radius of the sector.

[0045] S503. Based on the fan-shaped dredging route constructed from the vertex, angle, and radius, generate robot movement commands that include the direction of movement, the distance of movement, and the timing of turning.

[0046] Furthermore, obtaining underwater environmental information of the invisible environment also includes step S106.

[0047] S106. By setting a Gaussian kernel function with a preset standard deviation, convolution operation is performed on the underwater environmental information to filter out environmental noise and sensor random errors in the underwater environmental information.

[0048] The specific implementation principle of this application is explained below with reference to the accompanying drawings: In some embodiments, the present invention provides a robot intelligent control method that can be applied to underwater, invisible environments in sewage pumping stations. (See reference...) Figure 2 , Figure 3 , Figure 4 as well as Figure 5 The method includes the following steps: Step S1: Environmental information collection and conversion.

[0049] In underwater dredging operations at sewage pumping stations, effective perception of the underwater environment requires equipping the dredging robot with a 360° surround-scanning sonar sensor. The robot is deployed to the water area of ​​the pumping station, ensuring the sensor is at a stable working depth to prevent interference with data collection due to significant water fluctuations. The sonar sensor continuously emits sound waves, utilizing the time difference of echo, and constructs a two-dimensional distance map in polar coordinates, using the robot's current position as the pole, the sonar scanning angle as the polar angle, and the scanning distance as the polar radius. This distance map reflects the relative positions of underwater obstacles and the robot. Then, using image processing algorithms, it is converted into a grayscale image, presenting the distance information through differences in grayscale values, clearly revealing the morphological features of the environment such as walls and silt accumulation areas, and saving this data as the basis for subsequent processing.

[0050] Step S2: Sample data processing.

[0051] For the grayscale sample data obtained in step S1, a noise reduction operation is first performed. A Gaussian filtering algorithm is selected, and considering the noise characteristics of the underwater environment of the sewage pumping station, such as false echoes generated by water flow and random errors of the sensor itself, the standard deviation of the Gaussian kernel function is determined through trial and error. The grayscale image is input into the Gaussian filtering module, and noise is filtered out through convolution operations to obtain relatively clean sonar data. Next, detail enhancement is performed using a contrast enhancement algorithm to analyze the grayscale distribution patterns of the wall edges and silt areas, adjust the dynamic range of grayscale values, increase the grayscale difference between the two, highlight effective features, and provide clearer data support for subsequent line segment detection.

[0052] Step S3: Line segment detection and candidate library construction.

[0053] Based on the characteristics of underwater sonar data from sewage pumping stations, a dual-threshold algorithm for the Canny edge detection is configured: a low threshold is used to capture weak edges, while a high threshold retains strong edges. The data processed in step S2 is imported into the Canny edge detection module. The algorithm, using a dual-threshold filtering mechanism, distinguishes between genuine wall edges and false interference edges caused by silt accumulation. After detecting line segments, false edge segments are removed, and genuine wall edge segments are retained. These are then aggregated to construct a candidate line segment library for the pumping station walls, used for subsequent localization and matching.

[0054] Step S4: Static positioning.

[0055] Obtain accurate drawings of the sewage pumping station, extracting dimensions of key structures such as walls and location parameters of obstacles such as pumps. Construct a positioning frame adapted to the actual wall structure based on these parameters, ensuring the frame encompasses the features of the wall to be matched. Match the constructed positioning frame with the candidate wall segments from step S3, comprehensively considering the length, angle, and spatial relationship of the segments when calculating the matching degree. Specifically, calculate the angular deviation between the edge of the positioning frame and the candidate segment to measure the directional matching degree; calculate the length overlap to reflect the proportion of segment overlap; and calculate the spatial distance error to reflect the spatial position deviation, obtaining the comprehensive matching degree. Select the two segments with the highest matching degree, requiring parallel segments to have a certain interval or be nearly perpendicular. Update the robot's position information in the pumping station coordinate system based on their positional association with the positioning frame.

[0056] Step S5: Dynamic positioning.

[0057] The robot utilizes its onboard gyroscope to collect real-time turning angles and sensors on moving parts such as tracks to acquire real-time displacement, thus understanding the robot's real-time motion state. During the robot's movement, the starting scan angle of the distance metric map is updated synchronously at each movement time point, ensuring that this angle is consistent with the current turning angle. This discards sonar scan data from older postures, ensuring that the distance metric map accurately reflects the underwater environment from the robot's current perspective. Simultaneously, based on real-time displacement direction and velocity, the extension direction of the localization box is predicted, thereby continuously updating the localization information and maintaining the continuity and accuracy of the localization.

[0058] Step S6: Automated cleaning patrol.

[0059] Based on the robot localization results obtained in steps S4 and S5, a fan-shaped dredging route is planned. The robot's current position is taken as the apex of the fan, and its current orientation is taken as the direction of the edge. The radius is set based on the actual size of the pump station and the dredging requirements. Based on the planned fan-shaped route, robot movement commands containing the movement direction, distance, and turning timing are generated. When the robot performs its cruise, it moves back and forth along the edge of the fan and radially. It identifies surrounding obstacles in real time through localization information and the pump station map. Once an obstacle is detected, the movement path is adjusted in time to avoid collisions and rollovers, thus completing the dredging operation in the entire fan-shaped area.

[0060] In this embodiment, in step S1, sonar data and robot operation data need to be collected using corresponding sensors, and after preprocessing such as denoising and enhancement, they are used as input to the localization model. Specifically, the example sonar image after a single denoising and enhancement algorithm is shown in the figure. After data preprocessing, sonar noise caused by mud and sand is filtered out, and the relevant straight-line morphological features of the wall are enhanced.

[0061] In this embodiment, in steps S3 and S4, the localization algorithm based on the Canny edge detection operator and similarity matching extracts each candidate line segment during similarity calculation. The similarity is calculated against the four sides of the localization frame, and all wall lines with high similarity are selected based on a set similarity threshold. These wall lines may belong to the four sides of the pump station; therefore, it is necessary to confirm the wall to which each candidate line belongs. Using the pump station's shape characteristics, such as square or trapezoidal, the morphological features that the curve should possess are constructed, such as adjacent sides being perpendicular to each other or intersecting at a specific angle. Based on the pump station's preset drawing information, similar curves that satisfy the pump station's morphological characteristics are selected to construct a new pump station localization frame. Furthermore, based on the distance between the robot and these selected wall lines, the robot's position within the pump station localization frame is finally determined, updating the robot's localization information.

[0062] In this embodiment, in step S5, the "current position" determined by positioning is used as the vertex of the fan-shaped area, and the fan-shaped edge line is set in combination with the robot's orientation. The robot's angle and position are obtained in real time using a positioning algorithm. During movement, the robot moves along the fan-shaped edge and radially, centered on the vertex, and based on preset angles and radii. This is achieved by adjusting the steering angle and using displacement control. Polar coordinate positioning constraints enable automatic navigation and obstacle avoidance, ultimately completing the dredging and coverage of the fan-shaped area.

[0063] In other embodiments, the present invention provides a robot intelligent control method that can be applied to drainage ditch boxes, comprising the following steps: Step S11: Environmental information collection and conversion.

[0064] In underwater dredging operations in drainage silt boxes, effective perception of the underwater environment requires equipping the dredging robot with a 360° surround-scanning sonar sensor. The robot is deployed to the drainage silt box's working area, ensuring the sensor is at a stable working depth to prevent interference with data collection due to large water fluctuations. The sonar sensor continuously emits sound waves, utilizing the time difference of echo, with the robot's current position as the pole, the sonar scanning angle as the polar angle, and the scanning distance as the polar radius, constructing a two-dimensional distance map in polar coordinates. This distance map reflects the relative positions of underwater obstacles and the robot. Then, using image processing algorithms, it is converted into a grayscale image, presenting the distance information through differences in grayscale values, clearly revealing the morphological features of the environment such as walls and silt accumulation areas, and saving this data as the basis for subsequent processing.

[0065] Step S12: Sample data processing.

[0066] For the grayscale sample data obtained in step 11, noise reduction is performed first. A Gaussian filtering algorithm is selected, taking into account the noise characteristics of the underwater environment of the drainage ditch box, such as false echoes generated by water flow and random errors of the sensor itself. The standard deviation of the Gaussian kernel function is determined through trial and error. The grayscale image is input into the Gaussian filtering module, and noise is filtered out through convolution operations to obtain relatively clean sonar data. Next, detail enhancement is performed. A contrast enhancement algorithm is used to analyze the grayscale distribution patterns of the wall edges and silt areas, adjust the dynamic range of grayscale values, increase the grayscale difference between the two, highlight effective features, and provide clearer data support for subsequent line segment detection.

[0067] Step S13: Line segment detection and candidate library construction Based on the characteristics of the underwater sonar data from the robot in the drainage ditch box, a dual-threshold algorithm for the Canny edge detection is configured. The low threshold is used to capture weak edges, while the high threshold retains strong edges. The data processed in step S12 is imported into the Canny edge detection module. The algorithm, according to the dual-threshold filtering mechanism, distinguishes between real wall edges and false interference edges caused by silt accumulation. After detecting line segments, false edge segments are removed, and real wall edge segments are retained. These are then aggregated to construct a candidate line segment library for the drainage ditch box wall, which is used for subsequent localization and matching.

[0068] Step S4: Static positioning.

[0069] Obtain accurate drawings of the drainage chute and extract the dimensions of key structures such as walls and the location parameters of obstacles such as water pumps. Construct a positioning frame that matches the actual wall structure based on these parameters, ensuring the frame encompasses the features of the wall to be matched. Match the constructed positioning frame with the candidate wall segments from step S3, comprehensively considering the length, angle, and spatial relationship of the segments when calculating the matching degree. Specifically, calculate the angular deviation between the edge of the positioning frame and the candidate segment to measure the directional matching degree; calculate the length overlap to reflect the proportion of segment overlap; and calculate the spatial distance error to reflect the spatial position deviation, obtaining the comprehensive matching degree. Select the two segments with the highest matching degree, requiring parallel segments to have a certain interval or be nearly perpendicular. Based on their positional association with the positioning frame, update the robot's position information in the drainage chute's coordinate system.

[0070] Step S15: Dynamic positioning.

[0071] The robot utilizes its onboard gyroscope to collect real-time turning angles and sensors on moving parts such as tracks to acquire real-time displacement, thus understanding the robot's real-time motion state. During the robot's movement, the starting scan angle of the distance metric map is updated synchronously at each movement time point, ensuring that this angle is consistent with the current turning angle. This discards sonar scan data from older postures, ensuring that the distance metric map accurately reflects the underwater environment from the robot's current perspective. Simultaneously, based on real-time displacement direction and velocity, the extension direction of the localization box is predicted, thereby continuously updating the localization information and maintaining the continuity and accuracy of the localization.

[0072] Step S16: Automated underwater cleaning and patrol. Based on the robot localization results obtained in steps S14 and S15, a fan-shaped dredging route is planned. The robot's current position is taken as the vertex of the fan, and its current orientation is taken as the direction of the edge. The radius is set based on the actual size of the drainage ditch box and the dredging requirements. Based on the planned fan-shaped route, robot movement commands containing the movement direction, distance, and turning timing are generated. When the robot performs its cruise, it moves back and forth along the edge of the fan and radially. It identifies surrounding obstacles in real time through the localization information and the map of the drainage ditch box. Once an obstacle is detected, the movement path is adjusted in time to avoid collisions and rollovers, thus completing the dredging operation in the entire fan-shaped area.

[0073] In this embodiment, in step S11, sonar data and robot operation data need to be collected using corresponding sensors, and after preprocessing such as denoising and enhancement, they are used as input to the localization model. Specifically, the example sonar image after a single denoising and enhancement algorithm is shown in the figure. After data preprocessing, sonar noise caused by mud and sand is filtered out, and the relevant straight-line morphological features of the wall are enhanced.

[0074] In this embodiment, in steps S13 and S14, the localization algorithm based on the Canny edge detection operator and similarity matching extracts each candidate line segment during similarity calculation. The similarity is calculated against the four sides of the localization frame, and all wall lines with high similarity are selected based on a set similarity threshold. These wall lines may belong to the four sides of a drainage box; therefore, it is necessary to confirm the wall to which each candidate line belongs. Using the shape characteristics of the drainage box, such as a square or trapezoid, the morphological features that the curve should possess are constructed, such as adjacent sides being perpendicular to each other or intersecting at a specific angle. Based on the preset drawing information of the drainage box, similar curves that satisfy the shape characteristics of the drainage box are selected to construct a new drainage box localization frame. Furthermore, based on the distance between the robot and these selected wall lines, the robot's position within the drainage box localization frame is finally determined, updating the robot's localization information.

[0075] In this embodiment, in step S15, the "current position" determined by positioning is used as the vertex of the fan-shaped area, and the fan-shaped edge line is set in combination with the robot's orientation. The robot's angle and position are obtained in real time using a positioning algorithm. During movement, the robot moves along the fan-shaped edge and radially, centered on the vertex, and based on preset angles and radii. This is achieved by adjusting the steering angle and using displacement control. Polar coordinate positioning constraints enable automatic navigation and obstacle avoidance, ultimately completing the dredging and coverage of the fan-shaped area.

[0076] It is understood that the robot control method of this application can be applied in environments other than the two types of invisible environments mentioned above, as well as in other invisible environments (such as drainage tunnels, intercepting ponds, etc.). The above examples are only two specific environments in the invisible environment selected for illustration, and the above two examples should not be regarded as limiting the scope of protection of the application.

[0077] In addition, refer to Figure 6 ,and Figure 1Corresponding to the method described above, this application also provides a robot intelligent control system in its embodiments. The system may include: an acquisition unit 1001, a first processing unit 1002, a second processing unit 1003, a third processing unit 1004, and a fifth processing unit 1005. The acquisition unit 1001 is used to acquire underwater environmental information of the invisible environment and collect the robot's current motion data; the underwater environmental information includes a distance measurement map between the robot and obstacles in the invisible environment; the first processing unit 1002 is used to detect corresponding line segments of walls in the distance measurement map to obtain a candidate line segment library for all walls in the invisible environment; the second processing unit 1003 is used to determine the robot's static positioning information based on the candidate line segment library; the third processing unit 1004 is used to update the robot's dynamic positioning information based on the current motion data; the fourth processing unit 1005 is used to construct a fan-shaped dredging route for the robot based on the static or dynamic positioning information and generate corresponding robot movement commands, which are used to control the robot to perform automatic cruising in a fan-shaped motion pattern.

[0078] It should be noted that the content of the above-described robot intelligent control method embodiments is applicable to this robot intelligent control system embodiment. The specific functions implemented by this robot intelligent control system embodiment are the same as those of the above-described robot intelligent control method embodiments, and the beneficial effects achieved are also the same as those achieved by the above-described robot intelligent control method embodiments.

[0079] and Figure 1 Corresponding to the method described in this application, embodiments also provide a robot intelligent control device, the specific structure of which can be referred to... Figure 7 ,include: At least one processor 1011; At least one memory 1012 is used to store at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the robot intelligent control method.

[0080] The content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0081] and Figure 1 Corresponding to the method described above, embodiments of this application also provide a computer-readable storage medium storing processor-executable instructions, which, when executed by a processor, are used to perform the robot intelligent control method.

[0082] The contents of the above-described robot intelligent control method embodiments are all applicable to this storage medium embodiment. The specific functions implemented by this storage medium embodiment are the same as those of the above-described robot intelligent control method embodiments, and the beneficial effects achieved are also the same as those achieved by the above-described robot intelligent control method embodiments.

[0083] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0084] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional technology for an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0085] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0086] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0087] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0088] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0089] In the foregoing description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0090] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

[0091] The above is a detailed description of the preferred embodiments of this application, but this application is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A robot intelligent control method, characterized in that, Includes the following steps: Acquire underwater environmental information of the unseen environment and collect the robot's current motion data; the underwater environmental information includes a distance measurement map between the robot and obstacles in the unseen environment; The corresponding line segments of the walls in the distance metric map are detected to obtain a candidate line segment library for all walls in the invisible environment; Based on the candidate line segment library, the robot's static positioning information is determined; Based on the current motion data, update the robot's dynamic positioning information; Based on the static or dynamic positioning information, a fan-shaped dredging route for the robot is constructed and corresponding robot movement commands are generated. The robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion.

2. The robot intelligent control method according to claim 1, characterized in that, The candidate line segment library includes candidate sub-line segments of walls in different directions. Determining the robot's static positioning information based on the candidate line segment library includes: Based on the preset drawings of the aforementioned invisible environment, construct the robot's positioning frame; The positioning frame is matched with all the candidate sub-segments, and the matching degree between the positioning frame and each candidate sub-segment is calculated. Based on the matching degree, the static positioning information of the robot in the coordinate system of the invisible environment is determined.

3. The robot intelligent control method according to claim 1, characterized in that, The current motion data includes the robot's real-time turning angle, the robot's real-time displacement, the robot's initial scanning angle relative to the distance metric map at each motion time point, the robot's motion speed, and the robot's real-time displacement direction. Updating the robot's dynamic positioning information based on the current motion data specifically includes: Based on the real-time displacement direction and the movement speed, the extension direction of the positioning frame in the invisible environment coordinate system is determined; The robot position information is updated based on the real-time turning angle, the real-time displacement, the initial scanning angle, and the extension direction.

4. The robot intelligent control method according to claim 1, characterized in that, The step of detecting the corresponding line segments of the walls in the distance metric map to obtain a candidate line segment library for all walls in the invisible environment specifically includes: The corresponding line segments of the wall in the distance measurement map are detected to obtain the first coordinate of each corresponding line segment and the second coordinate of the positioning frame; Based on the first coordinate and the second coordinate, determine each matching degree between each corresponding line segment and the positioning frame; All corresponding line segments with a matching degree greater than a preset threshold are selected as all candidate line segments in the candidate line segment library.

5. The robot intelligent control method according to claim 4, characterized in that, The step of determining each matching degree between each corresponding line segment and the positioning box based on the first coordinate and the second coordinate includes: By inputting the first coordinate and the second coordinate into the matching degree calculation formula, each matching degree between each corresponding line segment and the positioning box is obtained; The matching degree calculation formula is as follows: , in For the first The coordinates of the candidate lines, These are the linear coordinates of the positioning box.

6. The robot intelligent control method according to claim 5, characterized in that, The step of constructing a robot fan-shaped dredging route and generating corresponding robot movement commands based on the static positioning information or the dynamic positioning information includes: The static positioning information or the dynamic positioning information is used to determine the vertices of the fan-shaped dredging route; With the robot's current orientation as the center line of the sector, set the angle and radius of the sector; Based on the fan-shaped dredging route constructed from the vertex, the angle, and the radius, robot movement commands containing the direction of movement, the distance of movement, and the timing of turning are generated.

7. The robot intelligent control method according to claim 1, characterized in that, The acquisition of underwater environmental information in the unseen environment also includes: By performing convolution operations on underwater environmental information using a Gaussian kernel function with a preset standard deviation, environmental noise and random errors from sensors in the underwater environmental information are filtered out.

8. A robot intelligent control system, characterized in that, include: The acquisition unit is used to acquire underwater environmental information of the invisible environment and collect the robot's current motion data; the underwater environmental information includes a distance measurement map between the robot and obstacles in the invisible environment; The first processing unit detects the corresponding line segments of the walls in the distance measurement map to obtain a candidate line segment library for all walls in the invisible environment. The second processing unit determines the robot's static positioning information based on the candidate line segment library; The third processing unit updates the robot's dynamic positioning information based on the current motion data; The fourth processing unit constructs a fan-shaped dredging route for the robot based on the static positioning information or the dynamic positioning information and generates corresponding robot movement commands. The robot movement commands are used to control the robot to perform automatic cruising in a fan-shaped motion.

9. A robot intelligent control device, characterized in that... include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a robot intelligent control method as described in any one of claims 1-7.

10. A computer-readable storage medium storing processor-executable instructions, characterized in that, The processor-executable instructions, when executed by the processor, are used to perform a robot intelligent control method as described in any one of claims 1-7.

Images