A self-positioning wall-climbing robot and system for bridge bottom sensor installation

Abstract

The application discloses a self-positioning wall-climbing robot and system for bridge bottom sensor installation, and relates to the technical field of bridge detection and monitoring equipment. The robot comprises a robot main body, a mobile adsorption mechanism for adsorbing on the surface of a bridge bottom structure for omnidirectional movement, and an autonomous positioning unit for identifying a bent cap region in the bridge bottom space by scanning the environmental image and laser point cloud data between the robot and the bridge bottom features at multiple angles. The autonomous positioning unit determines the real-time longitudinal and lateral positions of the robot main body and the target installation point by fitting the length, direction and two side boundaries of the bent cap, and issues an installation instruction. The installation execution mechanism is used for installing the bridge bottom sensor to the target installation point after the pose solution. The application utilizes the geometric spatial characteristics of the multiple piers and bent caps of the bridge to realize the autonomous positioning and accurate arrival of the device, thereby reducing the manual measurement error, improving the installation accuracy, and facilitating the repeated installation and rapid deployment between different bridges.

Description

Technical Field

[0001] This invention relates to the field of bridge inspection and monitoring equipment technology, and in particular to a self-positioning wall-climbing robot and system for installing sensors under bridges. Background Technology

[0002] Among in-service bridges, small and medium-span bridges account for a very high proportion, are widely distributed, and operate in complex environments. These bridges typically lack adequate power supply and long-term monitoring capabilities. However, in scenarios such as defect investigation, load testing, and pre- and post-reinforcement comparative analysis, it is still necessary to temporarily install sensors or detection units at key locations such as the beam bottom to obtain structural response data. Due to limited space at the beam bottom and high operating height, traditional installation methods often rely on aerial work platforms or scaffolding, resulting in complex construction organization, long preparation periods, and significant time consumption for single-point installation. This makes it difficult to meet the phased testing requirements of small and medium-span bridges, which are characterized by "large number, high frequency, and short cycle."

[0003] In practical engineering applications, to avoid drilling or mechanical anchoring damage to the structure, non-destructive adhesive bonding is commonly used to fix sensing devices. This method has minimal impact on the structure, but during the initial setting and curing stages of the adhesive, continuous stable pressure and accurate orientation are required to ensure bonding quality and installation precision. Because the beam bottom is an inverted working environment, construction personnel must hold or support the equipment at height for extended periods until the adhesive has fully cured. This is labor-intensive, inefficient, and poses significant safety risks, severely hindering the rapid testing of small- and medium-span bridges.

[0004] Furthermore, the bottom structures of small- and medium-span bridges typically exhibit geometric characteristics such as being long, straight, symmetrical, and having simple features, especially for bridges with continuous planar bottom structures and those equipped with cap beams. Manual installation lacks clear spatial reference points, and the positioning process relies on experience and simple measuring tools, making it difficult to achieve rapid and accurate alignment and digital recording of the installation position. Existing bridge climbing robots are mainly designed for surface inspection tasks and lack dedicated structures and actuators for automatic positioning and assisted adhesive installation of bridge bottom devices.

[0005] Therefore, it is necessary to propose a self-positioning wall-climbing robot and system suitable for small and medium-span bridges, so as to realize autonomous movement, precise positioning, and continuous pressing and posture maintenance in the bridge bottom environment, thereby reducing labor consumption, shortening installation time, and improving construction safety and installation quality. Summary of the Invention

[0006] Therefore, it is necessary to provide a self-positioning wall-climbing robot and system for installing sensors under bridges to address the aforementioned technical problems.

[0007] In a first aspect, the present invention provides a self-positioning wall-climbing robot for installing sensors under bridges, comprising:

[0008] Robot body;

[0009] The mobile adsorption mechanism is located on both sides of the top of the robot body and is used to adsorb onto the surface of the bridge bottom structure for omnidirectional movement.

[0010] The autonomous positioning unit, located at the bottom of the robot body, is used to identify the cap beam area in the space under the bridge by scanning the environmental images and laser point cloud data between the robot and the features under the bridge from multiple angles; by fitting the length, direction and two side boundaries of the cap beam to realize the pose calculation, determine the real-time longitudinal and transverse position of the robot body and the target installation point, and issue installation instructions.

[0011] The installation actuator is located at the top center of the robot body and is electrically connected to the autonomous positioning unit. It is used to clean and wipe the installation area under the bridge, and after cleaning and wiping, it installs the bridge bottom sensor to the target installation point after the pose calculation.

[0012] Furthermore, the mobile adsorption mechanism includes negative pressure cavities opened on both sides of the top of the robot body to form a negative pressure space, and a sealing ring is provided at the top edge of the negative pressure cavity.

[0013] The robot has air inlets on both sides of its main body, a fan in the middle of the negative pressure chamber, and an air outlet at the top of the negative pressure chamber and below the fan.

[0014] Mecanum wheels are installed at both ends of the negative pressure chamber to enable the robot to move in all directions and adjust its posture within the plane under the bridge while maintaining the robot in an adsorbed state.

[0015] Furthermore, the autonomous positioning unit includes a rotary motor located on one side of the bottom of the robot body. A vision sensor and a lidar are installed at the bottom output end of the rotary motor, and the vision sensor and lidar are electrically connected to an embedded control unit.

[0016] Among them, the visual sensor is used to acquire images of the environment under the bridge in real time and identify features under the bridge, including piers and cap beams.

[0017] LiDAR is used to obtain the relative distance information between the robot body and the features under the bridge, and to obtain laser point cloud data;

[0018] The embedded control unit is used to divide the cap beam area in the space under the bridge according to the laser point cloud data, transform the spatial geometric features of the cap beam, calculate the real-time position of the robot body during the movement process, and issue the installation command after reaching the target installation point.

[0019] Furthermore, based on the laser point cloud data, the area of ​​the cap beam in the space under the bridge is divided, and the spatial geometric features of the cap beam are transformed. The real-time position of the robot body during its movement is calculated, and installation instructions are issued after reaching the target installation point, including:

[0020] S1. After the robot body is attached to the surface of the bridge structure, power on and initialize the vision sensor and lidar to complete sensor calibration and system self-test.

[0021] S2, drives the vision sensor and lidar to perform a 360° synchronous scan and acquires laser point cloud data of the robot body and the surrounding environment.

[0022] S3. Divide the cap beam according to the point cloud coverage area, and filter and optimize the laser point cloud data in the cap beam area; by fitting the centerline and plane information of the cap beam length direction, and combining the spatial geometric relationship of the two sides of the cap beam, perform pose calculation, and output the real-time position of the robot body in the space under the bridge and the target installation break point.

[0023] S4. Using LiDAR to anchor any geometric feature point as a reference point during the movement, after the robot body reaches the target installation point, it performs a 360° synchronous scan again to obtain multiple geometric feature points for positioning correction, selects the final target installation point, and issues the installation command.

[0024] Furthermore, the cap beam is divided according to the point cloud coverage area, and the laser point cloud data within the cap beam area is filtered and optimized; by fitting the centerline and planar information along the length direction of the cap beam, and combining the spatial geometric relationship of the two sides of the cap beam, the pose is calculated, and the real-time position of the robot body in the space under the bridge and the target installation breakpoint are output, including:

[0025] S31. Convert the original polar coordinate laser point cloud data collected by the lidar into a point cloud set in the Cartesian coordinate system; each point in the point cloud set contains transverse coordinates, longitudinal coordinates, vertical coordinates, and echo intensity.

[0026] S32. Based on the point cloud set, the coordinate system direction is self-calibrated, the longitudinal bridge reference is selected, and the cap beam plane is obtained by probability density statistical fitting.

[0027] S33. Calculate the average height coordinates of the cap beam based on the laser point cloud data in the cap beam plane;

[0028] S34. Based on the plane and average height coordinates of the cap beam, select the centerline of the cap beam parallel to the transverse bridge axis, and extract the transverse bridge boundary points of the centerline.

[0029] S35. Based on the centerline of the cap beam, calculate the width and length of a single span beam slab;

[0030] S36. Calculate the coordinates of the target installation point based on the preset span ratio coefficient and the number of horizontal installations.

[0031] Furthermore, coordinate system orientation self-calibration is performed based on the point cloud set, a longitudinal bridge reference is selected, and the cap beam plane is obtained through probability density statistical fitting, including:

[0032] S321. Select all laser points with a vertical scanning angle of 0° from the point cloud set, and compare the laser ranging values ​​of all points in the point cloud coverage area one by one to find the target point with the smallest ranging value; set the horizontal scanning angle corresponding to the target point as the reference direction along the bridge axis.

[0033] S322. For all laser points in the point cloud set, count the frequency of occurrence of the longitudinal coordinate of each point and generate the probability density distribution curve of the longitudinal coordinate. Select the longitudinal coordinate value with the highest probability density in the probability density distribution curve, and take the spatial plane corresponding to the coordinate value that is parallel to the plane formed by the transverse and vertical directions as the cap beam plane.

[0034] Furthermore, based on the centerline of the cap beam, the calculation of the width and length of a single-span beam includes:

[0035] S351. Taking the transverse coordinates of the boundary points at both ends of the central axis of the cap beam as the reference, take the absolute values ​​of the maximum and minimum values ​​of the transverse bridge direction respectively, add the two absolute values ​​together, and calculate the transverse width of the target single span beam.

[0036] S352. Using the longitudinal coordinates of the reference plane of the two cap beams set opposite each other on both sides of the bridge as the reference, take the absolute value of the minimum value of the two longitudinal coordinates respectively, add the two absolute values ​​together, and calculate the longitudinal length of the target single span beam.

[0037] Furthermore, based on the preset span ratio coefficient and the number of horizontal installations, the coordinates of the target installation points are calculated, including:

[0038] S361. Pre-set the longitudinal span ratio coefficient and the transverse installation quantity; wherein, the longitudinal span ratio coefficient is used to determine the installation position of the sensor in the longitudinal direction, and the transverse installation quantity is used to determine the transverse installation number of the sensor under the bridge on the target section of a single span.

[0039] S362. Starting from the longitudinal coordinates of the cap beam plane, subtract the product of the span ratio coefficient and the length of the single span beam to obtain the longitudinal coordinates of the sensor installation point; and starting from the coordinates of the maximum transverse boundary point of the cap beam's central axis, divide the width of the single span beam into several uniform segments according to the number of transverse installations, and subtract the width of the corresponding number of segments in turn to obtain the transverse coordinates of each target installation point.

[0040] S363. Combine the calculated transverse and longitudinal coordinates of the bridge with the average height coordinates of the cap beam to form the three-dimensional spatial coordinates of each target installation point.

[0041] Furthermore, the installation actuator includes a sweeping wheel and a wiping wheel located on one side of the top center of the robot body;

[0042] A lifting motor is located on the other side of the top center of the robot body, and a U-shaped platform is located on the top of the lifting motor;

[0043] A storage compartment is located at the top center of the robot body. Inside the storage compartment, on the side away from the lifting motor, there is a spring feeder. A telescopic motor is located on one side of the top of the storage compartment, and a fixed guide rail is located on the other side of the top of the storage compartment. A glue bottle is placed between the output end of the telescopic motor and the fixed guide rail.

[0044] In a second aspect, the present invention also provides a self-positioning wall-climbing robot system for installing sensors under bridges, including a self-positioning wall-climbing robot for installing sensors under bridges, and an external controller that is wirelessly connected to the self-positioning wall-climbing robot for installing sensors under bridges.

[0045] The external controller is used to send task parameters to the robot body, receive installation point data and operation status returned by the robot, and enable manual emergency control of the robot.

[0046] The beneficial effects of this invention are as follows:

[0047] 1. To address the problem that existing bridge under-sense devices rely on manual positioning for installation, making it difficult to quickly and accurately determine the target cross-section location and ensuring consistency and repeatability of installation locations, this paper utilizes the geometric spatial characteristics of multiple bridge piers and cap beams to achieve autonomous positioning and precise arrival of the device. This reduces manual measurement errors, improves installation accuracy, and facilitates repeated installation and rapid deployment across different bridges.

[0048] 2. Traditional non-destructive adhesive bonding installation methods require manual support or reinforcement during the adhesive curing stage, resulting in low construction efficiency, high labor intensity, and safety risks associated with working at heights. This invention enables convenient installation of the bridge under-span sensing device, reducing manual intervention time, improving installation efficiency and posture stability, while ensuring the reliability of the adhesive bonding process and the consistency of the installation position. Therefore, it is suitable for phased or temporary monitoring scenarios of small and medium-span bridges, improving the safety and controllability of construction. Attached Figure Description

[0049] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0050] Figure 1 This is a schematic diagram of the structure of a self-positioning wall-climbing robot for installing sensors under a bridge, according to an embodiment of the present invention.

[0051] Figure 2 This is a top view of a self-positioning wall-climbing robot for installing sensors under a bridge, according to an embodiment of the present invention.

[0052] Figure 3 This is a side view of a self-positioning wall-climbing robot for installing sensors under a bridge, according to an embodiment of the present invention.

[0053] Figure 4 This is a schematic diagram of the bottom structure of a self-positioning wall-climbing robot for installing sensors under a bridge, according to an embodiment of the present invention.

[0054] Figure 5 This is an execution flowchart of a self-positioning wall-climbing robot for installing sensors under a bridge, according to an embodiment of the present invention.

[0055] Figure 6 This is a positioning schematic diagram of a self-positioning wall-climbing robot for installing sensors under a bridge, according to an embodiment of the present invention.

[0056] Reference numerals: 1. Robot body; 2. Mobile adsorption mechanism; 201. Negative pressure chamber; 202. Sealing ring; 203. Air inlet; 204. Fan; 205. Air outlet; 206. Mecanum wheel; 3. Autonomous positioning unit; 301. Rotary motor; 302. Vision sensor; 303. LiDAR; 304. Embedded control unit; 4. Installation actuator; 401. Sweeping wheel; 402. Wiping wheel; 403. Lifting motor; 404. U-shaped platform; 405. Storage compartment; 406. Feeder; 407. Telescopic motor; 408. Fixed guide rail; 409. Glue bottle. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0058] This invention addresses the problems of high reliance on manual labor, low construction efficiency, and difficulty in ensuring installation accuracy during the installation of sensing devices under bridges of small and medium span bridges. It aims to solve the technical problem of difficulty in achieving rapid, accurate, and stable installation of devices in complex environments under bridges.

[0059] Specifically, in the inverted working environment under bridges, traditional manual installation methods using non-destructive adhesive bonding require prolonged manual support or pressure during the adhesive curing stage, increasing labor intensity and safety risks; the positioning process lacks a reliable spatial reference, making it difficult to achieve rapid alignment of target points; and in scenarios involving repeated disassembly and assembly, the consistency of installation position and posture stability are difficult to control. Therefore, this invention aims to solve the technical problems of difficulty in maintaining continuous pressure and posture during the adhesive bonding process in bridge-under environments, long manual installation times, and difficulties in autonomous positioning of sensing devices.

[0060] To achieve the above objectives, this invention discloses a self-positioning wall-climbing robot and system for installing sensing devices under bridges of small and medium spans. The system mainly comprises three parts: an autonomous positioning unit, a mobile adsorption mechanism, and an installation execution device. The adsorption and mobile mechanism is used to achieve stable attachment and movement on the bridge underside surface; the autonomous positioning unit is used to determine the target installation position; and the installation execution device is used to continuously apply pressure to the device and maintain its stable posture during the adhesive application process.

[0061] Please see Figures 1-4 A self-positioning wall-climbing robot for installing sensors under bridges is provided, comprising: robot body 1.

[0062] The mobile adsorption mechanism 2 is located on both sides of the top of the robot body 1 and is used to adsorb onto the surface of the bridge bottom structure for omnidirectional movement.

[0063] The autonomous positioning unit 3 is located at the bottom of the robot body 1. It is used to identify the cap beam area in the space under the bridge by scanning the environmental images and laser point cloud data between the robot and the features under the bridge from multiple angles. By fitting the length, direction and two side boundaries of the cap beam, the robot body 1 is used to calculate the pose, determine the real-time longitudinal and transverse position of the robot body 1 and the target installation point, and issue installation instructions.

[0064] The actuator 4 is installed at the top center of the robot body 1 and is electrically connected to the autonomous positioning unit 3. It is used to clean and wipe the installation area under the bridge and, after cleaning and wiping, install the bridge bottom sensor to the target installation point after the pose calculation.

[0065] In the description of the present invention, the mobile adsorption mechanism 2 includes negative pressure cavities 201 opened on both sides of the top of the robot body 1 to form a negative pressure space, and a sealing ring 202 is provided on the top edge of the negative pressure cavity 201; air inlets 203 are opened on both sides of the robot body 1, a fan 204 is provided in the middle of the negative pressure cavity 201, and an air outlet 205 is opened on the top of the negative pressure cavity 201 and below the fan 204; Mecanum wheels 206 are provided at both ends of the negative pressure cavity 201 to realize omnidirectional movement and attitude adjustment of the robot in the plane under the bridge while the robot is in an adsorption state.

[0066] In the description of the present invention, the autonomous positioning unit 3 includes a rotary motor 301 disposed on one side of the bottom end of the robot body 1. A vision sensor 302 and a lidar 303 are disposed at the bottom output end of the rotary motor 301, and the vision sensor 302 and the lidar 303 are electrically connected to an embedded control unit 304.

[0067] The visual sensor 302 is used to acquire images of the environment under the bridge in real time and identify features under the bridge, including piers and cap beams.

[0068] The lidar 303 is used to obtain the relative distance information between the robot body 1 and the features under the bridge, and to obtain laser point cloud data.

[0069] The embedded control unit 304 is used to divide the cap beam area in the space under the bridge according to the laser point cloud data, and to transform the spatial geometric features of the cap beam, calculate the real-time position of the robot body 1 during the movement process, and issue the installation command after reaching the target installation point.

[0070] In the description of this invention, as Figures 5-6 As shown, the cap beam area in the space under the bridge is divided according to the laser point cloud data, and the spatial geometric features of the cap beam are transformed. The real-time position of the robot body 1 during its movement is calculated, and the installation command is issued after reaching the target installation point, including:

[0071] S1. After the robot body 1 is adsorbed onto the surface of the bridge bottom structure, the vision sensor 302 and the lidar 303 are powered on and initialized to complete the sensor calibration and system self-test.

[0072] S2, drives the vision sensor 302 and the lidar 303 to perform a 360° synchronous scan and acquire the laser point cloud data of the robot body 1 and the surrounding environment.

[0073] S3. Divide the cap beam according to the point cloud coverage area, and filter and optimize the laser point cloud data within the cap beam area. By fitting the centerline and planar information along the length direction of the cap beam, and combining the spatial geometric relationship of the two sides of the cap beam, perform pose calculation, and output the real-time position of the robot body 1 in the space under the bridge and the target installation breakpoint.

[0074] In the description of this invention, the cap beam is divided according to the point cloud coverage area, and the laser point cloud data within the cap beam area is filtered and optimized; by fitting the centerline and planar information along the length direction of the cap beam, and combining the spatial geometric relationship of the two side boundaries of the cap beam, the pose is calculated, and the real-time position of the robot body 1 in the space under the bridge and the target installation breakpoint are output, including:

[0075] S31. Convert the original polar coordinate laser point cloud data collected by the lidar 303 into a point cloud set in the Cartesian coordinate system; each point in the point cloud set contains transverse bridge coordinates, longitudinal bridge coordinates, vertical coordinates, and echo intensity.

[0076] Specifically, the raw measurement data from the LiDAR 303 is typically expressed in the form of distance and angle (polar coordinates), and each point can be represented as:

[0077] ;

[0078] in, This indicates the laser ranging value (distance). Indicates the horizontal scan angle (azimuth angle); Indicates the vertical scan angle (tilt angle); This indicates the echo intensity.

[0079] Convert to Cartesian coordinate system form:

[0080] ;

[0081] in, Lateral coordinates of the bridge Indicates the coordinates along the bridge direction (longitudinal direction). Represents the vertical (height) coordinates. Indicates the intensity of reflection.

[0082] S32. Based on the point cloud set, the coordinate system direction is self-calibrated, the longitudinal bridge reference is selected, and the cap beam plane is obtained by probability density statistical fitting.

[0083] In the description of this invention, coordinate system orientation self-calibration is performed based on point cloud sets, a longitudinal bridge reference is selected, and the cap beam plane is obtained through probability density statistical fitting, including:

[0084] S321. Select all laser points with a vertical scanning angle of 0° from the point cloud set, and compare the laser ranging values ​​of all points in the point cloud coverage area one by one to find the target point with the smallest ranging value; set the horizontal scanning angle corresponding to the target point as the reference direction of the bridge axis.

[0085] Specifically, the coordinate system orientation is first determined using polar coordinate data. That is, within the collected data, the horizontal angle at which the minimum distance is located within the vertical angle range of 0° is the direction along the bridge (y-axis direction).

[0086] ;

[0087] The distance between each point on the cap beam plane and the lidar is different, but the y-coordinate of the cap beam plane is constant in coordinate representation. Therefore, the cap beam plane can be found by processing the point cloud data through probability density estimation. The expression is:

[0088] ;

[0089] in, It is the probability density function of y.

[0090] S322. For all laser points within the point cloud set, count the frequency of occurrence of the longitudinal coordinate of each point, and generate a probability density distribution curve for the longitudinal coordinate. Select the longitudinal coordinate value with the highest probability density in the probability density distribution curve, and use the spatial plane corresponding to this coordinate value, which is parallel to the plane formed by the transverse and vertical directions, as the cap beam plane.

[0091] Specifically, the point cloud data is constrained to the cap beam plane, and the expression is:

[0092] ;

[0093] in, This indicates the plan view of the cap beam.

[0094] S33. Calculate the average height coordinates of the cap beam based on the laser point cloud data in the cap beam plane.

[0095] Specifically, the calculated cap beam plan Calculate the average height coordinates of the cap beam. The expression is:

[0096] ;

[0097] in, This represents the average coordinate point of the cap beam.

[0098] S34. Based on the coordinates of the cap beam plane and average height, select the cap beam centerline parallel to the transverse bridge axis, and extract the transverse bridge boundary points of the centerline.

[0099] Specifically, known , Define the centerline of the cap beam. Since it is parallel to the x-axis, and the boundary points at both ends are determined by the maximum and minimum values, the expression for the parametric point set of the centerline is as follows:

[0100] ;

[0101] ;

[0102] in, Represents the minimum boundary point of the central axis; This represents the point of maximum boundary of the central axis.

[0103] The same method can be used to calculate the centerline on the other side. The expression is:

[0104] ;

[0105] S35. Based on the centerline of the cap beam, calculate the width and length of a single span beam.

[0106] In the description of this invention, calculating the width and length of a single-span beam based on the central axis of the cap beam includes:

[0107] S351. Using the transverse coordinates of the boundary points at both ends of the central axis of the cap beam as a reference, take the absolute values ​​of the maximum and minimum transverse values ​​respectively, add the two absolute values ​​together, and calculate the transverse width of the target single-span beam.

[0108] S352. Using the longitudinal coordinates of the reference plane of the two cap beams set opposite each other on both sides of the bridge as the reference, take the absolute value of the minimum value of the two longitudinal coordinates respectively, add the two absolute values ​​together, and calculate the longitudinal length of the target single span beam.

[0109] Specifically, the formula for calculating the space dimensions under the bridge is:

[0110] ; ;

[0111] in, Indicates the width of a single-span beam / slab; This indicates the length of a single-span beam or slab.

[0112] S36. Calculate the coordinates of the target installation point based on the preset span ratio coefficient and the number of horizontal installations.

[0113] In the description of this invention, the coordinates of the target installation point are calculated based on a preset span ratio coefficient and the number of horizontal installations, including:

[0114] S361. Pre-set the longitudinal span ratio coefficient and the transverse installation quantity; wherein, the longitudinal span ratio coefficient is used to determine the installation position of the sensor in the longitudinal direction, and the transverse installation quantity is used to determine the transverse installation number of the sensor under the bridge on the target section of a single span.

[0115] Specifically, the longitudinal span ratio coefficient is used to determine the installation position of the sensor in the longitudinal direction of the bridge. Optional values ​​include 1 / 2, 1 / 3, and 1 / 4, corresponding to the installation positions at the mid-span, one-third span, and one-quarter span of the bridge, respectively. The expression for the span ratio coefficient is:

[0116] ;

[0117] S362. Starting from the longitudinal coordinates of the cap beam plane, subtract the product of the span ratio coefficient and the length of a single span beam to obtain the longitudinal coordinates of the sensor installation point. Then, starting from the coordinates of the maximum transverse boundary point of the cap beam's central axis, divide the width of a single span beam into several uniform segments according to the number of transverse installations, and subtract the corresponding number of equal segments to obtain the transverse coordinates of each target installation point.

[0118] Specifically, the installation points along the bridge direction are located by horizontal distance. Subtracting the set span ratio length gives the coordinates of the longitudinal installation point, expressed as:

[0119] ;

[0120] Let the number of horizontal installations be Based on this number, the points are divided at equal intervals and proportionally, and then... Subtract the corresponding distance to calculate the coordinates of the horizontal installation point:

[0121] ;

[0122] S363. Combine the calculated transverse and longitudinal coordinates of the bridge with the average height coordinates of the cap beam to form the three-dimensional spatial coordinates of each target installation point.

[0123] Specifically, the coordinates of the corresponding installation points can be calculated for sensor installation:

[0124] ;

[0125] S4. Using the LiDAR 303 to anchor any geometric feature point as a reference point during the movement, after the robot body 1 reaches the target installation point, it performs a 360° synchronous scan again to obtain multiple geometric feature points for positioning correction, selects the final target installation point, and issues the installation command.

[0126] In the description of the present invention, the installation actuator 5 includes a sweeping wheel 401 and a wiping wheel 402 disposed on one side of the top middle position of the robot body 1.

[0127] A lifting motor 403 is installed on the other side of the top center of the robot body 1, and a U-shaped platform 404 is installed at the top of the lifting motor 403.

[0128] A storage compartment 405 is located at the top center of the robot body 1. A spring feeder 406 is located inside the storage compartment 405 on the side away from the lifting motor 403. A telescopic motor 407 is located on one side of the top of the storage compartment 405. A fixed guide rail 408 is located on the other side of the top of the storage compartment 405. A glue bottle 409 is located between the output end of the telescopic motor 407 and the fixed guide rail 408.

[0129] Specifically, the sweeping wheel 401 is used for preliminary dust removal of the target installation location, removing loose particles from the surface; the wiping wheel 402 is used for fine wiping of the installation area, improving the cleanliness of the adhesive interface; the telescopic motor 407 is used to drive the glue bottle to extend and compress in the front-to-back direction, realizing the movement of the glue bottle and the compression of the adhesive material; the fixed guide rail 408 is used to limit the movement of the glue bottle, ensuring the stability and direction of movement, and a partition is set at the top to limit the maximum extension distance; the glue bottle 409 has a double-bottle structure, used to store two kinds of adhesive materials, which can be mixed and applied to the adhesive application location; the lifting motor 403 is used to drive the lifting platform to move vertically, realizing continuous pressing during the sensor bonding process; the U-shaped platform 404 serves as the sensor carrying structure, with an upper U-shaped baffle defining the sensor position and a lower trapezoidal structure facilitating sensor transfer; the storage compartment 405 is used to place the sensor to be installed, realizing integrated carrying and automatic retrieval; the spring-loaded feeder 406 is a spring structure used for automatic sensor transport.

[0130] Secondly, the present invention also provides a self-positioning wall-climbing robot system for installing sensors under bridges, including a self-positioning wall-climbing robot for installing sensors under bridges, and an external controller that is wirelessly connected to the self-positioning wall-climbing robot for installing sensors under bridges.

[0131] An external controller is used to send task parameters to the robot body (1), receive installation point data and operation status returned by the robot, and realize manual emergency control of the robot.

[0132] The specific workflow is as follows:

[0133] (1) Initial attachment and system startup stage: After transporting the wall-climbing robot to the target area under the bridge through the external handle, the mobile adsorption mechanism is started. The fan works to create a stable negative pressure environment in the negative pressure chamber. With the fit of the sealing ring, reliable adsorption on the surface of the bridge structure is achieved. The autonomous positioning unit is powered on and initialized to complete sensor calibration and system self-test.

[0134] (2) Environmental perception and autonomous localization stage: The rotary motor drives the lidar to perform multi-angle scanning to obtain distance information and point cloud data between the robot and the surrounding environment; the system divides the cap beam according to the point cloud coverage area and filters and optimizes the point cloud data within the cap beam area. By fitting the centerline and planar information of the cap beam along its length direction through the aforementioned algorithm, and combining the spatial geometric relationship of the two sides of the cap beam, the robot's real-time position in the space under the bridge and the position of the target installation section are determined.

[0135] (3) Path planning and movement to the target installation point: After determining the current position and the target installation point, the lidar anchors a certain feature point as a reference during the movement process, and drives the Mecanum wheel to move in any direction on the plane under the bridge. When the target installation point is reached, multiple geometric feature points are acquired again in 360° for positioning correction to reduce the error.

[0136] (4) Installation surface pretreatment stage: After the robot is in place, the installation execution device is started. The sweeping wheel first performs preliminary dust removal on the target area to remove loose particles; then the wiping wheel performs fine cleaning treatment on the installation interface to improve the cleanliness and adhesion conditions of the bonding interface.

[0137] (5) Sensor conveying and gluing stage: When the lifting motor lowers the U-shaped platform, the storage compartment will send the sensor to be installed to the U-shaped platform and limit its position. At this time, the U-shaped platform will rise again and rise to the same level as the glue bottle outlet. The telescopic motor drives the glue bottle to extend to the top along the fixed guide rail, and the glue outlet extends to the sensor contact surface at the same time. At the same time, the glue bottle is squeezed to complete the quantitative gluing. After the gluing is completed, the telescopic motor will retract the glue bottle to the initial position.

[0138] (6) Pressing, curing and posture maintenance stage: After the adhesive is applied, the lifting motor drives the U-shaped platform to rise, so that the sensor contacts the target installation point and maintains a continuous pressing state; during the curing of the adhesive, the platform maintains stable pressure and posture to ensure the bonding reliability and the consistency of the installation position.

[0139] (7) Completion and transfer stage: After the colloid reaches the initial curing strength, the U-shaped platform descends and resets, and the installation execution device is retrieved; the robot continues to move to the next installation position according to the task plan.

[0140] The entire process achieves a continuous closed-loop operation from autonomous positioning, precise placement, interface processing, adhesive application and compaction to automatic reset, without the need for continuous manual intervention under the bridge.

[0141] In summary, by utilizing the above-mentioned technical solution of this invention, the problems of existing bridge under-span sensing devices relying on manual positioning, making it difficult to quickly and accurately determine the target cross-section position, and ensuring the consistency and repeatability of the installation position are addressed. By leveraging the geometric spatial characteristics of multiple bridge piers and cap beams, the device achieves autonomous positioning and precise arrival, thereby reducing manual measurement errors, improving installation accuracy, and facilitating repeated installation and rapid deployment across different bridges. Regarding traditional non-destructive adhesive bonding installation methods, which require prolonged manual support or reinforcement during the adhesive curing stage, resulting in low construction efficiency, high labor intensity, and safety risks associated with working at heights, this invention enables convenient installation of bridge under-span sensing devices, reducing manual intervention time, improving installation efficiency and posture stability, while ensuring the reliability of the adhesive bonding process and the consistency of the installation position. Therefore, it is suitable for phased or temporary monitoring scenarios on small and medium-span bridges, improving construction safety and controllability.

[0142] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

Claims

1. A self-positioning wall-climbing robot for installing sensors under bridges, characterized in that, include: Robot body (1); The mobile adsorption mechanism (2) is set on both sides of the top of the robot body (1) and is used to adsorb onto the surface of the bridge bottom structure for omnidirectional movement; The autonomous positioning unit (3) is set at the bottom of the robot body (1) and is used to identify the cover beam area in the bridge bottom space by scanning the environmental image and laser point cloud data between the robot and the bridge bottom feature from multiple angles; by fitting the length, direction and two side boundaries of the cover beam to realize the pose calculation, determine the real-time longitudinal and transverse position of the robot body (1) and the target installation point, and issue installation instructions. The installation actuator (4) is located at the top center of the robot body (1) and is electrically connected to the autonomous positioning unit (3). It is used to clean and wipe the installation area under the bridge, and after cleaning and wiping, the bridge bottom sensor is installed at the target installation point after the pose calculation.

2. The self-positioning wall-climbing robot for installing sensors under bridges according to claim 1, characterized in that, The mobile adsorption mechanism (2) includes negative pressure cavities (201) opened on both sides of the top of the robot body (1) to form a negative pressure space, and a sealing ring (202) is provided at the top edge of the negative pressure cavity (201). The robot body (1) has air inlets (203) on both sides, a fan (204) is provided in the middle of the negative pressure chamber (201), and an air outlet (205) is provided at the top of the negative pressure chamber (201) and below the fan (204). Both ends of the negative pressure chamber (201) are provided with Mecanum wheels (206) to enable the robot to move in all directions and adjust its posture in the plane under the bridge while the robot is in an adsorption state.

3. The self-positioning wall-climbing robot for installing sensors under bridges according to claim 1, characterized in that, The autonomous positioning unit (3) includes a rotary motor (301) disposed on one side of the bottom end of the robot body (1). A visual sensor (302) and a lidar (303) are disposed at the bottom output end of the rotary motor (301), and an embedded control unit (304) is electrically connected to the visual sensor (302) and the lidar (303). The visual sensor (302) is used to acquire images of the bridge underside environment in real time and identify bridge underside features in the bridge underside environment, including bridge piers and cap beams. The lidar (303) is used to obtain the relative distance information between the robot body (1) and the features under the bridge, and to obtain laser point cloud data; The embedded control unit (304) is used to divide the cap beam area in the bridge bottom space according to the laser point cloud data, and to convert the spatial geometric features of the cap beam, calculate the real-time position of the robot body (1) during the movement process, and issue an installation command after reaching the target installation point.

4. A self-positioning wall-climbing robot for installing sensors under bridges according to claim 3, characterized in that, The process of dividing the bridge underpass area into cap beam regions based on laser point cloud data, transforming the spatial geometric features of the cap beams, calculating the real-time position of the robot body (1) during its movement, and issuing installation instructions after reaching the target installation point includes: After the robot body (1) is adsorbed on the surface of the bridge bottom structure, the vision sensor (302) and the lidar (303) are powered on and initialized to complete sensor calibration and system self-test. The vision sensor (302) and the lidar (303) are driven to perform a 360° synchronous scan and acquire the laser point cloud data of the robot body (1) and the surrounding environment. The cap beam is divided according to the point cloud coverage area, and the laser point cloud data in the cap beam area is filtered and optimized; by fitting the centerline and plane information of the cap beam length direction, and combining the spatial geometric relationship of the two sides of the cap beam, the pose is calculated, and the real-time position of the robot body (1) in the space under the bridge and the target installation break point are output. Using a lidar (303) to anchor any geometric feature point as a reference point during the movement, when the robot body (1) reaches the target installation point, it performs a 360° synchronous scan again to obtain multiple geometric feature points for positioning correction, selects the final target installation point, and issues an installation command.

5. A self-positioning wall-climbing robot for installing sensors under bridges according to claim 3, characterized in that, The process involves dividing the cap beam according to the point cloud coverage area, filtering and optimizing the laser point cloud data within the cap beam area; by fitting the centerline and planar information along the length direction of the cap beam, and combining the spatial geometric relationship of the two sides of the cap beam to perform pose calculation, the real-time position of the robot body (1) in the space under the bridge and the target installation breakpoint position are output, including: The original polar coordinate laser point cloud data collected by the lidar (303) is converted into a point cloud set in the Cartesian coordinate system; each point in the point cloud set contains transverse bridge coordinates, longitudinal bridge coordinates, vertical coordinates and echo intensity; The coordinate system orientation is self-calibrated based on point cloud set, the longitudinal bridge reference is selected, and the cap beam plane is obtained by probability density statistical fitting. Calculate the average height coordinates of the cap beam based on the laser point cloud data within the cap beam plane. Based on the plane and average height coordinates of the cap beam, the centerline of the cap beam parallel to the transverse bridge axis is selected, and the transverse bridge boundary points of the centerline are extracted. Based on the centerline of the cap beam, calculate the width and length of a single span beam slab; Based on the preset span ratio coefficient and the number of horizontal installations, the coordinates of the target installation point are calculated.

6. A self-positioning wall-climbing robot for installing sensors under bridges according to claim 5, characterized in that, The coordinate system orientation self-calibration based on point cloud sets, selecting a longitudinal bridge reference, and obtaining the cap beam plane through probability density statistical fitting includes: Select all laser points with a vertical scanning angle of 0° from the point cloud set, and compare the laser ranging values ​​of all points in the point cloud coverage area one by one to find the target point with the smallest ranging value; set the horizontal scanning angle corresponding to the target point as the reference direction along the bridge axis; For all laser points in the point cloud set, the frequency of occurrence of the longitudinal coordinate of each point is counted to generate the probability density distribution curve of the longitudinal coordinate. The longitudinal coordinate value with the highest probability density in the probability density distribution curve is selected, and the spatial plane corresponding to this coordinate value, which is parallel to the plane formed by the transverse and vertical directions, is used as the cap beam plane.

7. A self-positioning wall-climbing robot for installing sensors under bridges according to claim 6, characterized in that, The calculation of the width and length of a single span beam based on the centerline of the cap beam includes: Using the transverse coordinates of the boundary points at both ends of the central axis of the cap beam as a reference, the absolute values ​​of the maximum and minimum transverse values ​​are taken respectively, and the two absolute values ​​are added together to calculate the transverse width of the target single-span beam. Using the longitudinal coordinates of the reference plane of the two cap beams set opposite each other on both sides of the bridge as the reference, the absolute values ​​of the minimum values ​​of the two longitudinal coordinates are taken respectively, and the two absolute values ​​are added together to calculate the longitudinal length of the target single span beam.

8. A self-positioning wall-climbing robot for installing sensors under bridges according to claim 5, characterized in that, The calculation of the coordinates of the target installation point based on the preset span ratio coefficient and the number of horizontal installations includes: The longitudinal span ratio coefficient and the transverse installation quantity are preset; the longitudinal span ratio coefficient is used to determine the installation position of the sensor in the longitudinal direction, and the transverse installation quantity is used to determine the number of transverse installations of the sensor under the bridge on the target section of a single span. Starting from the longitudinal coordinates of the cap beam plane, subtract the product of the span ratio coefficient and the length of the single span beam to obtain the longitudinal coordinates of the sensor installation point; and starting from the coordinates of the maximum transverse boundary point of the cap beam's central axis, divide the width of the single span beam into several uniform segments according to the number of transverse installations, and subtract the width of the corresponding number of segments to obtain the transverse coordinates of each target installation point. The calculated transverse and longitudinal coordinates of the bridge are combined with the average height coordinates of the cap beam to form the three-dimensional spatial coordinates of each target installation point.

9. A self-positioning wall-climbing robot for installing sensors under bridges according to claim 1, characterized in that, The installation actuator (5) includes a sweeping wheel (401) and a wiping wheel (402) located on one side of the top middle position of the robot body (1). A lifting motor (403) is provided on the other side of the top middle position of the robot body (1), and a U-shaped platform (404) is provided on the top of the lifting motor (403). The robot body (1) has a storage compartment (405) at the top center. Inside the storage compartment (405), on the side away from the lifting motor (403), there is a spring feeder (406). On one side of the top of the storage compartment (405), there is a telescopic motor (407). On the other side of the top of the storage compartment (405), there is a fixed guide rail (408). A glue bottle (409) is placed between the output end of the telescopic motor (407) and the fixed guide rail (408).

10. A self-positioning wall-climbing robot system for installing sensors under bridges, comprising the self-positioning wall-climbing robot for installing sensors under bridges as described in any one of claims 1-9, and an external controller wirelessly connected to the self-positioning wall-climbing robot for installing sensors under bridges. The external controller is used to send operation task parameters to the robot body (1), receive installation point data and operation status returned by the robot, and realize manual emergency control of the robot.

Images