CN117705800A - A robotic arm visual bridge detection system based on guide rail sliding and its control method - Google Patents

Abstract

The invention relates to a mechanical arm vision bridge detection system based on guide rail sliding and a control method thereof, wherein the mechanical arm vision bridge detection system comprises a linear guide rail sub-module, a mechanical arm sub-module, an onboard computer sub-module and a high-speed camera sub-module; the high-speed camera sub-module shoots bridge disease images, the mechanical arm sub-module drives the high-speed camera sub-module to translate on a plane vertical to the longitudinal direction of the bridge, and the linear guide rail sub-module drives the mechanical arm sub-module and the airborne computer sub-module to translate along the longitudinal direction of the bridge; the linear guide rail sub-module is additionally arranged on the bridge structure and is cut into a region outside the driving region; the airborne computer sub-module comprises a guide rail displacement control system, a mechanical arm rotation control system, a bridge disease image recognition system and a network transmission system. The invention has no influence on traffic in the later detection, has good universality and belongs to the technical field of bridge disease detection.

Description

A robotic arm visual bridge detection system based on guide rail sliding and its control method

Technical field

The invention relates to bridge disease detection technology, and in particular to a robotic arm visual bridge detection system based on guide rail sliding and its control method.

Background technique

Bridge structures are subject to the coupling effects of multiple factors such as aging concrete materials, severe vehicle overloading, and harsh operating environments. In-service bridges will inevitably suffer from cracking, peeling of protective layers, water seepage, alkalinization, and corrosion of exposed steel bars, which will seriously affect the safety and safety of the bridges. Durability is a great test. Therefore, it is necessary to detect defects in the bridge structure as early as possible to prevent further reduction of its structural load-bearing capacity and durability. Among them, apparent structural disease is the most obvious sign, indicating that the structure may be deteriorated or damaged. It is also the "Standard for Evaluation of Technical Condition of Highway Bridges" (JTG/TH21-2011) and "Code for Maintenance of Highway Bridge Culverts" (JTG/H11-2004). ), "Technical Specifications for Urban Bridge Inspection and Assessment" (CJJ/T233-2015) and other key evaluation indicators in many bridge technical condition assessment manuals.

CN 116147505 A discloses an intelligent detection terminal for bridge defect detection, including a platform for bridge detection, a robotic arm, a crack identification system and a crack detection system. The crack identification system is installed on the robotic arm, and then the robotic arm is installed on the On the vehicle, the detection of the back side of the bridge is realized through the movement of the vehicle. This technology drives the robotic arm to translate along the length of the bridge through the carrying vehicle. It has the following shortcomings: 1. It occupies the bridge deck lane during inspection, affecting traffic; 2. It is difficult for the carrying vehicle to ensure straight operation. If there is deflection in the left and right directions, Affects the alignment of the crack identification system; 3. The mounted vehicle yaws in the left and right directions during driving and is affected by the road surface, so the mounted vehicle is not easy to achieve automatic control and precise alignment, which will affect the bridge detection efficiency and detection Effect.

Contents of the invention

In view of the technical problems existing in the prior art, the purpose of the present invention is to provide a robotic arm visual bridge inspection system based on guide rail sliding that will not affect traffic during the inspection process.

Another object of the present invention is to provide a control method for a robotic arm visual bridge detection system based on guide rail sliding that can realize intelligent control and accurate detection.

In order to achieve the above objects, the present invention adopts the following technical solutions:

A robotic arm visual bridge inspection system based on guide rail sliding, including a linear guide rail sub-module, a robotic arm sub-module, an airborne computer sub-module and a high-speed camera sub-module; the high-speed camera sub-module captures bridge disease images, and the robotic arm sub-module drives the high-speed The camera sub-module translates on a plane perpendicular to the longitudinal direction of the bridge. The linear guide rail sub-module drives the robotic arm sub-module and the airborne computer sub-module to translate longitudinally along the bridge. The linear guide rail sub-module is installed on the bridge structure and is located outside the driving area. area; the airborne computer sub-module includes a guide rail displacement control system, a robotic arm rotation control system, a bridge disease image recognition system, and a network transmission system; the rail displacement control system controls the linear guide rail sub-module, thereby controlling the longitudinal position of the robotic arm sub-module; The robot arm rotation control system controls the action of the robot arm sub-module to control the target position and shooting angle of the high-speed camera sub-module; the bridge disease image recognition system identifies the captured bridge disease images; the network transmission system transmits the identification results to the terminal .

As an option, the linear guide rail sub-module includes a servo motor, a guide rail slider mechanism, an encoder, a rack and pinion mechanism, and a position feedback loop; in the guide rail slider mechanism, the guide rails are laid longitudinally along the bridge, and the slider carries the robotic arm submodule and airborne computer submodule; in the rack and pinion mechanism, the rack is set parallel to the guide rail, and the servo motor drives the gear to rotate; the servo motor drives the slider to translate along the guide rail through the rack and pinion mechanism; the encoder detects the position of the slider, and Position information is transmitted to the rail displacement control system through a position feedback loop.

Preferably, the slider is provided with a slider base for installing the robotic arm sub-module and the airborne computer sub-module.

As a preference, the guide rails are laid outside the bridge deck fence, on the side of the bridge or at the bottom of the bridge.

As an option, the robotic arm sub-module is a six-degree-of-freedom robotic arm. A mounting device is provided at the end of the robotic arm. The mounting device is used to install the high-speed camera sub-module. The robotic arm rotation control system controls the rotation angle of the robotic arm and will be mounted on the mechanical arm. The high-speed camera submodule at the end of the arm is moved to the bottom of the bridge.

As an option, the high-speed camera sub-module includes a high-speed camera and a lighting system; the high-speed camera is used to capture bridge disease images, and the lighting system is used to improve the quality of picture shooting, thereby improving the accuracy of picture recognition.

As an option, in the airborne computer sub-module, by presetting the working route, the guide rail displacement control system and the robotic arm rotation control system automatically calculate the best working path and the pause time of the high-speed camera sub-module for image recognition, bridge The disease image recognition system receives the pictures transmitted by the high-speed camera sub-module, uses deep learning algorithms to identify bridge diseases, and transmits them to the terminal through the network transmission system.

As a preferred method, the algorithm for bridge disease identification should be selected according to different tasks.

As an option, a robotic arm visual bridge inspection system based on guide rail sliding is detachably installed on the bridge structure.

A control method for a robotic arm visual bridge detection system based on guide rail sliding, including the following steps:

Step 1: Preset the working route of the bridge inspection at the terminal and transmit it to the airborne computer sub-module through the network transmission system. The airborne computer sub-module uses the predefined working route and controls it through the guide rail displacement control system and the mechanical arm rotation. The system automatically calculates the best working path and shooting pause time; Step 2: The onboard computer sub-module controls the linear guide rail sub-module equipped with the robotic arm sub-module and the high-speed camera sub-module to move to the target position, and then controls the robotic arm sub-module to adjust the rotation Angle, so that the high-speed camera sub-module is in a position to clearly capture the bottom of the bridge; Step 3: The high-speed camera sub-module transfers the captured bridge disease pictures to the airborne computer sub-module, and the airborne computer sub-module performs image processing through the bridge disease image recognition system Recognition; Step 4: The airborne computer sub-module transmits the recognition result to the terminal through the network transmission system.

The invention has the following advantages:

1. The system of the present invention is set up in an area outside the driving area. It does not require a full road closure for detection. It only performs short-term enclosure construction when installing the linear guide rail sub-module, which has little impact on traffic. Later detection has no impact on traffic and can be uninterrupted. Perform testing.

2. The system of the present invention does not need to be permanently fixedly connected to the structure, can be disassembled and easily deployed on different detection structures, and has strong structural applicability and good versatility.

3. The system of the present invention does not require manual on-site operation, has low cost and high safety.

4. The system of the present invention is based on image recognition based on deep learning algorithms, with high recognition accuracy and good robustness.

5. Each sub-module of the system of the present invention works together to adjust the longitudinal bridge position of the robotic arm sub-module and the high-speed camera sub-module based on the parameters of each sub-module calculated by the on-board computer, and adjust the rotation angle of the robotic arm sub-module to adapt to high speed. The shooting position of the camera sub-module can adapt to the bridge detection needs of different bridge types; the system is easy to install and disassemble, and has good structural integration, high device utilization, high control efficiency, and a wide range of applications.

6. The system of the present invention does not need to change the bridge structure, can be installed on various types of bridges, and has strong versatility.

Description of the drawings

Figure 1 is a perspective view of the robot arm sub-module.

Figure 2 is a perspective view of the linear guide sub-module.

Figure 3 is a schematic diagram of an application example of the system of the present invention.

Figure 4 is a control flow chart of the system of the present invention.

In the picture, 1-guide rail, 2-slider, 3-robot arm, 4-high-speed camera, 5-bridge.

Detailed ways

The present invention will be described in further detail below in conjunction with specific embodiments.

A robotic arm visual bridge inspection system based on guide rail sliding, including a linear guide rail sub-module, a robotic arm sub-module, an airborne computer sub-module and a high-speed camera sub-module; the high-speed camera sub-module captures bridge disease images, and the robotic arm sub-module drives the high-speed The camera sub-module translates on a plane perpendicular to the longitudinal direction of the bridge. The linear guide rail sub-module drives the robotic arm sub-module and the airborne computer sub-module to translate longitudinally along the bridge. The linear guide rail sub-module is installed on the bridge structure and is located outside the driving area. area; the airborne computer sub-module includes a guide rail displacement control system, a robotic arm rotation control system, a bridge disease image recognition system, and a network transmission system; the rail displacement control system controls the linear guide rail sub-module, thereby controlling the longitudinal position of the robotic arm sub-module; The robot arm rotation control system controls the action of the robot arm sub-module to control the target position and shooting angle of the high-speed camera sub-module; the bridge disease image recognition system identifies the captured bridge disease images; the network transmission system transmits the identification results to the terminal .

The linear guide sub-module includes a servo motor, a guide rail slider mechanism, an encoder, a rack and pinion mechanism, a position feedback loop, and has a certain additional load capacity. In the guide rail slider mechanism, the guide rail is laid longitudinally along the bridge, and the slider carries the robotic arm sub-module and the airborne computer sub-module; in the gear rack and pinion mechanism, the rack is set parallel to the guide rail, and the servo motor drives the gear to rotate; the servo motor passes through the gear The rack mechanism drives the slider to translate along the guide rail; the encoder detects the position of the slider and transmits the position information to the guide rail displacement control system through the position feedback loop.

The slider is provided with a slider base for installing the robotic arm sub-module and the airborne computer sub-module. There are storage compartments inside the slider base for housing the onboard computer submodules.

The guide rails are laid on the outside of the bridge deck fence, the side of the bridge or the bottom of the bridge. In this embodiment, the guide rails are laid on the outer edge of the bridge.

The robotic arm sub-module is a six-degree-of-freedom robotic arm. The end of the robotic arm is equipped with a mounting device. The mounting device is used to install the high-speed camera sub-module. The robotic arm rotation control system controls the rotation angle of the robotic arm and installs the high-speed camera mounted on the end of the robotic arm. The submodule is moved to the base of the bridge.

The high-speed camera sub-module includes a high-speed camera and lighting system; the high-speed camera is used to capture bridge disease images, and the lighting system is used to improve the quality of picture shooting, thereby improving the accuracy of picture recognition.

In the airborne computer sub-module, by pre-setting the working route, the guide rail displacement control system and the robotic arm rotation control system automatically calculate the optimal working path and the pause time of the high-speed camera sub-module for image recognition. The bridge disease image recognition system receives The pictures transmitted by the high-speed camera sub-module use deep learning algorithms to identify bridge diseases and are transmitted to the terminal through the network transmission system.

The algorithm for bridge disease identification depends on the different tasks. For example, bridge crack identification can use the YOLOv8 algorithm.

A robotic arm visual bridge inspection system based on guide rail sliding is detachably installed on the bridge structure. It can be installed on the existing bridge structure when needed and can be easily dismantled when not needed.

A control method for a robotic arm visual bridge detection system based on guide rail sliding, including the following steps:

Step 1: Preset the working route of the bridge detection on the terminal, including the initial spatial position parameters of the robot arm sub-module and high-speed camera sub-module, the spatial position parameters of the raised obstacles along the linear guide sub-module, and the bridge disease shooting points The distance parameters and shooting time parameters are transmitted to the airborne computer sub-module through the network transmission system; the airborne computer sub-module uses the predefined working route to automatically calculate the optimal position through the guide rail displacement control system and the robotic arm rotation control system. Optimal working path and shooting pause time;

Step 2: After the linear guide sub-module receives the spatial position update command, the onboard computer sub-module controls the linear guide sub-module equipped with the manipulator sub-module and the high-speed camera sub-module to move to the target position. After the manipulator sub-module receives the rotation angle update command, Then control the robotic arm sub-module to adjust the rotation angle so that the high-speed camera sub-module is in a position to clearly photograph the bottom of the bridge to photograph bridge diseases;

Step 3: The high-speed camera sub-module transfers the captured bridge disease pictures to the airborne computer sub-module, which performs image recognition through the bridge disease image recognition system;

Step 4: The onboard computer sub-module transmits the recognition result to the terminal through the network transmission system.

The onboard computer then continues to calculate the spatial position parameters of each sub-module at the next point until the entire bridge detection process is completed.

The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, etc. may be made without departing from the spirit and principles of the present invention. All simplifications should be equivalent substitutions, and are all included in the protection scope of the present invention.

Claims (10)

1. A robotic arm visual bridge detection system based on guide rail sliding, characterized by: including a linear guide rail sub-module, a robotic arm sub-module, an airborne computer sub-module and a high-speed camera sub-module; the high-speed camera sub-module captures bridge disease images, The robotic arm sub-module drives the high-speed camera sub-module to translate on a plane perpendicular to the longitudinal direction of the bridge, and the linear guide sub-module drives the robotic arm sub-module and airborne computer sub-module to translate longitudinally along the bridge; The linear guide rail sub-module is installed on the bridge structure and is located outside the driving area; The airborne computer sub-module includes a guide rail displacement control system, a robotic arm rotation control system, a bridge disease image recognition system, and a network transmission system; the rail displacement control system controls the linear guide rail sub-module to control the longitudinal position of the robotic arm sub-module; the robotic arm rotation The control system controls the action of the robotic arm sub-module to control the target position and shooting angle of the high-speed camera sub-module; the bridge disease image recognition system identifies the captured bridge disease images; the network transmission system transmits the identification results to the terminal. 2. A robotic arm visual bridge detection system based on guide rail sliding according to claim 1, characterized in that: the linear guide rail sub-module includes a servo motor, a guide rail slider mechanism, an encoder, a rack and pinion mechanism, and a position feedback loop. ; In the guide rail slider mechanism, the guide rail is laid longitudinally along the bridge, and the slider carries the robotic arm sub-module and the airborne computer sub-module; in the gear rack and pinion mechanism, the rack is set parallel to the guide rail, and the servo motor drives the gear to rotate; the servo motor passes through The rack and pinion mechanism drives the slider to translate along the guide rail; the encoder detects the position of the slider and transmits the position information to the guide rail displacement control system through the position feedback loop. 3. A robotic arm visual bridge detection system based on guide rail sliding according to claim 2, characterized in that: the slider is provided with a slider base for installing the robotic arm sub-module and the airborne computer sub-module. 4. A robotic arm visual bridge detection system based on guide rail sliding according to claim 2, characterized in that the guide rail is laid outside the bridge deck fence, on the side of the bridge or at the bottom of the bridge. 5. A robotic arm visual bridge detection system based on guide rail sliding according to claim 1, characterized in that: the robotic arm sub-module is a six-degree-of-freedom robotic arm, and a mounting device is provided at the end of the robotic arm, and the mounting device is used for high-speed cameras. Installation of sub-modules; the robotic arm rotation control system controls the rotation angle of the robotic arm and moves the high-speed camera sub-module mounted on the end of the robotic arm to the bottom of the bridge. 6. A robotic arm visual bridge detection system based on guide rail sliding according to claim 1, characterized in that: the high-speed camera sub-module includes a high-speed camera and a lighting system; the high-speed camera is used to capture bridge disease images, and the lighting system Used to improve the quality of picture shooting, thereby improving the accuracy of picture recognition. 7. A robotic arm visual bridge detection system based on guide rail sliding according to claim 1, characterized in that: in the airborne computer sub-module, by presetting the working route, the guide rail displacement control system and the robotic arm rotation The control system automatically calculates the best working path and the pause time for image recognition by the high-speed camera sub-module. The bridge disease image recognition system receives the images transmitted by the high-speed camera sub-module, uses deep learning algorithms to identify bridge diseases, and transmits them through the network transmission system to the terminal. 8. A robotic arm visual bridge detection system based on guide rail sliding according to claim 7, characterized in that: the algorithm for bridge disease identification selects a corresponding algorithm according to different tasks. 9. A robotic arm visual bridge detection system based on guide rail sliding according to claim 1, characterized in that: it is detachably installed on the bridge structure. 10. The control method of a robotic arm visual bridge detection system based on guide rail sliding according to any one of claims 1 to 9, characterized in that it includes the following steps: Step 1: Preset the working route of the bridge inspection at the terminal and transmit it to the airborne computer sub-module through the network transmission system. The airborne computer sub-module uses the predefined working route and controls it through the guide rail displacement control system and the mechanical arm rotation. The system automatically calculates the best working path and shooting pause time; Step 2: The onboard computer sub-module controls the linear guide rail sub-module equipped with the robotic arm sub-module and the high-speed camera sub-module to move to the target position, and then controls the robotic arm sub-module to adjust the rotation angle so that the high-speed camera sub-module is in a position to clearly capture the bottom of the bridge. Location; Step 3: The high-speed camera sub-module transmits the captured bridge disease pictures to the airborne computer sub-module, and the airborne computer sub-module performs image recognition through the bridge disease image recognition system; Step 4: The onboard computer sub-module transmits the recognition result to the terminal through the network transmission system.