Single-wheel self-balancing rail inspection device and method based on sound detection

Abstract

The invention discloses a single-wheel self-balancing rail inspection device and method based on sound detection. The single-wheel self-balancing rail inspection device comprises a single-wheel self-balancing rail inspection device body and a server terminal. The single-wheel self-balancing track inspection device is placed on a rail, the posture and the balance state of the inspection device are detected through a gyroscope in the moving process, a rail image is recognized through a tracking camera, and the single-wheel self-balancing track inspection device is controlled to keep a self-balancing state so as to perform tracking movement along the single rail. In the moving process, the rail is knocked at fixed time or fixed distance, sound is collected, sound data is transmitted to the server terminal for real-time sound detection, and the working condition of the rail is distinguished according to sound classification. According to the invention, the inspection device is driven by means of camera tracking, self-balancing control and single-wheel advancing, so that the inspection device can walk on a single rail and can be kept stable; meanwhile, sound generated by knocking the rail is collected, the sound is detected through a machine learning or deep learning method, the sound is recognized in a multi-classification mode, the rail working condition is judged, and sound-based rail abnormal working condition detection is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent rail transportation, and in particular relates to a rail inspection device and method based on sound detection. Background Art

[0002] Periodic rail inspections are crucial for ensuring the smooth operation of rail transit. Currently, there are two primary methods for periodic rail inspections: traditional manual inspections, which suffer from low efficiency, subjectivity, and low reliability. The other method involves the use of inspection devices, which utilize cameras, ultrasonic flaw detectors, and other testing equipment. However, these devices are difficult to transport and dispatch, have high inspection equipment costs, are susceptible to climate and environmental influences, and are complex to use.

[0003] During manual inspections, maintenance workers use a maintenance hammer to hit the rails and identify the rails by sound, which is a common type of rail inspection. However, manual sound identification requires a lot of experience, and manual inspection is inefficient, time-consuming, labor-intensive, and highly subjective, which can easily lead to errors. The invention with publication number CN112575636A discloses a device that can automatically detect railway tracks based on sound. The device includes a main box body, a sliding guide rail cavity, and a maintenance hammer cavity on the left side of the sliding guide rail cavity. By converting the sound of the maintenance hammer hitting the rail from sound waves into electrical signals, amplifying the electrical signals and converting them into current, and then converting the current into mechanical energy, automatic tapping along the rails is achieved. However, the device has the following disadvantages: simply judging whether the sound produced by the maintenance hammer hitting the rails is too loud or too soft is used to determine whether the rails are abnormal. This is easily affected by the environment, and the definition of normal, too loud, or too soft sound is highly subjective, with large judgment errors and poor reliability. Moreover, the device can only provide a binary judgment of whether the rails are normal or abnormal, and cannot distinguish the type of rail abnormality. The invention with authorization number CN115817570B discloses a rail damage inspection device and inspection method based on monorail self-balancing. The left and right shells are clamped on both sides of the track, the device is driven to move by a motor and a transmission rod, and rail detection is implemented by a linear array camera and a vibration sensor. However, the device has the following shortcomings: the device does not use sound as a detection means, and the left and right shells rub against the two sides of the rail during the movement of the device, which can easily cause the shells to wear, thereby affecting the clamping force between the device and the rail and the stability of the movement.

[0004] The shortcomings of the above-mentioned method of detecting railway tracks based on sound make it of great significance to explore a railway track inspection device and method based on sound detection with simple structure, high reliability and multi-classification discrimination for the application of automatic railway track inspection. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and provide a single-wheeled self-balancing rail inspection device and method based on sound detection. The device has a simple structure, high reliability, and can realize multi-classification detection of rail abnormalities based on sound, effectively improving the accuracy of rail abnormality detection, reducing the workload of inspection personnel, and improving inspection efficiency.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] A single-wheel self-balancing rail inspection device based on sound detection comprises a single-wheel self-balancing rail inspection device and a server terminal.

[0008] The unicycle self-balancing track inspection device comprises a main body frame, a track unicycle, a power module, a sound control and acquisition module, a vision and motion control module, a central controller module, and a data communication module.

[0009] The bottom of the main frame is connected to a track wheel, and the track wheel is connected to the frame assembly; the track wheel, power module, sound control and acquisition module, vision and motion control module, central controller module, and data communication module are all installed and connected to the frame assembly.

[0010] The track wheel is a rubber wheel used to carry the main body frame and drive the main body frame to move along the surface of the rail.

[0011] The power module includes a battery and a motor, which are used to drive the inspection device to move. The motor is used to drive the track wheel to move, and the battery is used to provide energy for the motor and other modules.

[0012] The sound control and acquisition module includes a pendulum strike control assembly and a sound acquisition array assembly, which are used to control the inspection device's strikes on the rails to produce sound and to collect the sound. The pendulum strike control assembly, located outside the frame assembly, controls the pendulum's lift and release at a set strike point to produce the sound. The sound acquisition array assembly, comprised of multiple microphones, is activated synchronously when the pendulum strikes the rails to collect the sound.

[0013] The vision and motion control module includes a tracking camera assembly and a self-balancing control assembly, which are used to detect and identify rails through image detection, and control the inspection device to maintain a self-balancing state, so that the inspection device can track and maintain stability along a single rail. The tracking camera assembly is mounted on the frame assembly and faces the track forward, and is used to capture images of the rails from the forward direction. The self-balancing control assembly includes a gyroscope sensor and a flywheel. The gyroscope sensor is used to capture the posture of the inspection device. The flywheel is arranged on the axis of the frame assembly and is used to adjust the rotation direction and speed of the flywheel according to the balance state of the inspection device, adjusting the forward, backward, left, and right tilt of the inspection device to control the inspection device to maintain a self-balancing state and track along a single rail.

[0014] The central controller module is electrically connected to the power module, sound control and acquisition module, and vision and motion control module, and is used to receive and process data from the power module, sound control and acquisition module, and vision and motion control module, and control the various components of the power module, sound control and acquisition module, and vision and motion control module to perform functions, including but not limited to: 1) real-time detection and identification of rail tracks, 2) real-time control and adjustment of the flywheel rotation direction and speed; 3) timing or fixed-distance control of the pendulum to hit the rails; 4) control of sound acquisition; 5) control of communication with the server terminal.

[0015] The data communication module is electrically connected to the central controller module and is used for bidirectional communication with the server terminal, receiving control information for the inspection device sent from the server terminal, sending the sound collected by the sound collection array component to the server terminal, sending the rail image taken forward by the track tracking camera component to the server terminal, and sending status information of the inspection device to the server terminal.

[0016] The server terminal is wirelessly connected to the inspection device and is used to: 1) receive the rail image taken forward by the track-tracking camera assembly and detect and identify the rail in real time; 2) receive status information of the inspection device, send control information to the inspection device, and remotely control the inspection device; 3) receive and store the sound collected by the sound collection array assembly sent by the data communication module in the inspection device, analyze and detect the received sound, and determine the working condition of the rail through sound classification.

[0017] The above-mentioned single-wheeled self-balancing rail inspection device based on sound detection is adopted. When in use, the frame body is placed on the rail, and the central controller controls the operation of the power module to drive the single wheel to drive the frame body to move on the rail; during the movement of the frame body, the central controller controls the sound control and acquisition module to knock on the rail at a fixed time or interval and collect sound, and transmits the data to the server terminal through the data communication module. The server terminal performs real-time sound detection on the rail and determines whether there are safety hazards on the rail based on sound classification; during the movement of the frame body, the central controller controls the vision and motion control module to detect the posture and balance state of the inspection device in real time, and collect and identify rail images in real time, and control the inspection device to maintain a self-balancing state and track along a single rail.

[0018] Preferably, the function of detecting and identifying rail tracks in the central controller module adopts deep learning or machine learning methods to improve the accuracy of track detection and identification.

[0019] A method for inspecting a single-wheeled self-balancing railroad track based on sound detection comprises the following steps:

[0020] Step S1: The unicycle self-balancing track inspection device is placed on the railway track, and the power module drives the main frame to move on the railway track;

[0021] Step S2: The inspection device detects and identifies the rail track through the tracking camera assembly. The motion control module uses the gyroscope sensor and flywheel to control the inspection device to maintain a self-balancing state, so that the unicycle self-balancing track inspection device tracks along the railway track. At the same time, the track image captured by the tracking camera assembly is sent to the server terminal, and the status information of the inspection device is sent to the server terminal.

[0022] Step S3: The inspection device controls the pendulum to strike the rails according to a preset distance or time interval through the sound control and acquisition module;

[0023] Step S4: The inspection device collects and stores the sound of the pendulum hitting the rails through the sound collection array and transmits it to the server terminal through the data communication module;

[0024] Step S5: The server terminal uses machine learning or deep learning methods to detect sound, identify sound classification, and determine the working condition of the rails;

[0025] Step S6: The server terminal receives the rail image sent by the inspection status, and detects and identifies the rail in real time; the server terminal receives the status information of the inspection device, sends control information to the inspection device, and remotely controls the inspection device.

[0026] Preferably, the rail working condition sound classification types in step S5 include but are not limited to: normal rail, loose rail, loose screw, inclined rail, and foreign matter on the rail.

[0027] The present invention has the following advantages and effects compared to the prior art:

[0028] (1) The present invention collects the sound produced by knocking on the rails, detects the sound through machine learning or deep learning methods, identifies the sound through multiple classifications, and determines the working condition of the rails. This effectively solves the problem of simply judging whether the sound produced by the maintenance hammer knocking on the rails is too loud or too soft, which is easily affected by the environment, has a strong subjectivity in defining normal, too loud or too soft sounds, has large judgment errors, and has poor reliability. In addition, the present invention can only give a two-category judgment of whether the rails are normal or abnormal, and cannot determine the type of abnormal rails.

[0029] (2) The present invention drives the inspection device by adopting camera tracking, self-balancing control, and unicycle travel. During the movement of the device, no guide wheels are required to clamp the two sides of the rail, which will not cause friction and wear. It has the advantages of strong travel stability, simple structure, and easy use. It effectively solves the defects of the rail inspection device such as complex structure, unchanging use, and the friction between the device and the rail affecting the clamping force and travel stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The present invention is a schematic diagram of the overall structure of a single-wheeled self-balancing rail inspection device based on sound detection.

[0031] Figure 2 This is a diagram showing the overall module structure of a single-wheeled self-balancing rail inspection device based on sound detection according to the present invention.

[0032] Figure 3 The present invention is a functional connection diagram of a single-wheeled self-balancing rail inspection device based on sound detection.

[0033] Figure 4 The present invention is a flowchart of a method for inspecting a single-wheeled self-balancing railroad based on sound detection. DETAILED DESCRIPTION

[0034] The present invention will be described in further detail below with reference to the embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0035] like Figure 1 and Figure 2 As shown, a single-wheel self-balancing rail inspection device based on sound detection includes a single-wheel self-balancing rail inspection device and a server terminal.

[0036] The unicycle self-balancing track inspection device includes a main frame 1, a power module 2, a track unicycle 3, a sound control and acquisition module 4, a vision and motion control module 5, a central controller module 6, and a data communication module 7.

[0037] The bottom of the main frame 1 is connected to the track wheel 3, and the track wheel 3 is connected to the frame assembly 11; the track wheel 3, the power module 2, the sound control and acquisition module 4, the vision and motion control module 5, the central controller module 6, and the data communication module 7 are all installed and connected to the frame assembly 11.

[0038] The three wheels of the track are rubber wheels, which are used to carry the main frame 1 and drive the main frame 1 to move along the surface of the rail 8.

[0039] The power module 2 includes a battery 21 and a motor 22, which are used to drive the inspection device to move. The motor 22 is used to drive the track wheel 3 to move, and the battery 21 is used to provide energy for the motor and other modules.

[0040] The sound control and acquisition module 4 includes a pendulum strike control assembly 41 and a sound acquisition array assembly 42, which are used to control the inspection device's strikes on the rails to produce sound and to collect the sound. The pendulum strike control assembly 41, located outside the frame assembly 11, controls the pendulum's lift and release at a set strike point, thereby producing the sound. The sound acquisition array assembly 42, consisting of multiple microphones, activates synchronously when the pendulum strikes the rails, collecting the sound produced by the pendulum striking the rails.

[0041] The vision and motion control module 5 includes a tracking camera assembly 51 and a self-balancing control assembly 52, which are used to detect and identify rails through image detection, and control the inspection device to maintain a self-balancing state, so that the inspection device can track and maintain stability along a single rail. The tracking camera assembly 51 is mounted on the axis of the frame assembly 11 and faces the track forward, and is used to capture images of the rails from the forward direction. The self-balancing control assembly 52 includes a gyroscope sensor 522 and a flywheel 521. The gyroscope sensor 522 is used to capture the posture of the inspection device, and the flywheel 521 is mounted on the axis of the frame assembly 11. It is used to adjust the rotation direction and speed of the flywheel according to the balance state of the inspection device, adjusting the forward, backward, left, and right tilt of the inspection device to control the inspection device to maintain a self-balancing state and track along a single rail 8.

[0042] The central controller module 6 is electrically connected to the power module 2, the sound control and acquisition module 4, and the vision and motion control module 5, and is used to receive and process data from the power module 2, the sound control and acquisition module 4, and the vision and motion control module 5, and control the various components of the power module 2, the sound control and acquisition module 4, and the vision and motion control module 5 to perform functions, including but not limited to: 1) real-time detection and identification of rail tracks, 2) real-time control and adjustment of the flywheel rotation direction and speed; 3) timing or fixed-distance control of the pendulum to hit the rails; 4) control of sound acquisition; 5) control of communication with the server terminal.

[0043] The data communication module 7 is electrically connected to the central controller module 6 and is used for bidirectional communication with the server terminal, receiving control information for the inspection device sent from the server terminal, sending the sound collected by the sound collection array component to the server terminal, sending the rail image taken forward by the track tracking camera component to the server terminal, and sending status information of the inspection device to the server terminal.

[0044] The server terminal is wirelessly connected to the inspection device and is used to: 1) receive images of the rails taken forward by the tracking camera assembly and detect and identify the rails in real time; 2) receive status information of the inspection device, send control information to the inspection device, and remotely control the inspection device; 3) receive and store sounds collected by the sound collection array assembly sent by the data communication module in the inspection device, analyze and detect the received sounds, and determine the working condition of the rails through sound classification.

[0045] like Figure 1 and Figure 3 As shown, the above-mentioned one-wheeled self-balancing rail inspection device based on sound detection is used. When in use, the frame body 1 is placed on the rail, and the central controller 6 controls the power module 2 to work, driving the unicycle 3 to drive the frame body 11 to move on the track 8; during the movement of the frame body 11, the central controller 6 controls the sound control and acquisition module 4 to knock on the rail at a fixed time or interval and collect sound, and transmits the data to the server terminal through the data communication module 7. The server terminal performs real-time sound detection on the track and determines whether there are safety hazards on the rail based on sound classification; during the movement of the frame body, the central controller 6 controls the vision and motion control module 5 to detect the posture and balance state of the inspection device in real time, and collect and identify the rail image in real time, and control the inspection device to maintain a self-balancing state and track along a single rail.

[0046] Preferably, the function of detecting and identifying rail tracks in the central controller module 6 adopts deep learning or machine learning methods to improve the accuracy of track detection and identification.

[0047] like Figure 4 As shown, a method for inspecting a single-wheeled self-balancing railroad track based on sound detection includes the following steps:

[0048] Step S1: The unicycle self-balancing track inspection device is placed on the railway track, and the power module drives the main frame to move on the railway track;

[0049] Step S2: The inspection device detects and identifies the rail track through the tracking camera assembly. The motion control module uses the gyroscope sensor and flywheel to control the inspection device to maintain a self-balancing state, so that the unicycle self-balancing track inspection device tracks along the railway track. At the same time, the track image captured by the tracking camera assembly is sent to the server terminal, and the status information of the inspection device is sent to the server terminal.

[0050] Step S3: The inspection device controls the pendulum to strike the rails according to a preset distance or time interval through the sound control and acquisition module;

[0051] Step S4: The inspection device collects and stores the sound of the pendulum hitting the rails through the sound collection array and transmits it to the server terminal through the data communication module;

[0052] Step S5: The server terminal uses machine learning or deep learning methods to detect sound, identify sound classification, and determine the working condition of the rails;

[0053] Step S6: The server terminal receives the rail image sent by the inspection status, and detects and identifies the rail in real time; the server terminal receives the status information of the inspection device, sends control information to the inspection device, and remotely controls the inspection device.

[0054] Preferably, the rail working condition sound classification types in step S5 include but are not limited to: normal rail, loose rail, loose screw, inclined rail, and foreign matter on the rail.

[0055] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be considered as equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. A single-wheeled self-balancing rail inspection device based on sound detection, characterized by: The utility model comprises a single-wheel self-balancing track inspection device and a server terminal.

2. The one-wheeled self-balancing track inspection device according to claim 1, characterized in that: It includes a main frame, a track wheel, a power module, a sound control and acquisition module, a vision and motion control module, a central controller module, and a data communication module.

3. The one-wheeled self-balancing track inspection device according to claims 1 and 2, characterized in that: The bottom of the main frame is connected to a track wheel, which is connected to the frame assembly; the track wheel, power module, sound control and acquisition module, vision and motion control module, central controller module, and data communication module are all installed and connected to the frame assembly; The track wheel is a rubber wheel used to carry the main frame and drive the main frame to move along the surface of the rail; The power module includes a battery and a motor for driving the inspection device. The motor is used to drive the track wheel, and the battery is used to provide energy for the motor and other modules. The sound control and acquisition module includes a pendulum strike control component and a sound acquisition array component, which are used to control the inspection device to strike the rails to produce sound and to collect the sound. The pendulum strike control component is arranged on the outside of the frame component and is used to control the pendulum to be raised and released at a set strike point on the rail to strike the rails to produce sound. The sound acquisition array component is composed of multiple microphones, which are synchronously turned on when the pendulum strikes the rails to collect the sound produced by the pendulum striking the rails. The vision and motion control module includes a tracking camera assembly and a self-balancing control assembly, which are used to detect and identify rails through image detection, and control the inspection device to maintain a self-balancing state, so that the inspection device can track along a single rail and maintain stability. The tracking camera assembly is installed on the frame assembly and faces the track forward, and is used to capture images of the rails in the forward direction; the self-balancing control assembly includes a gyroscope sensor and a flywheel. The gyroscope sensor is used to capture the posture of the inspection device. The flywheel is set on the axis of the frame assembly and is used to adjust the rotation direction and speed of the flywheel according to the balance state of the inspection device, and adjust the forward, backward, left, and right tilt of the inspection device to control the inspection device to maintain a self-balancing state and track along a single rail. The central controller module is electrically connected to the power module, the sound control and acquisition module, and the vision and motion control module, and is used to receive and process data from the power module, the sound control and acquisition module, and the vision and motion control module, and control the various components of the power module, the sound control and acquisition module, and the vision and motion control module to perform functions, including but not limited to: 1) real-time detection and identification of rail tracks; 2) real-time control and adjustment of the rotation direction and speed of the flywheel; 3) timing or fixed-distance control of the pendulum to strike the rails; 4) control of sound acquisition; 5) control of communication with the server terminal; The data communication module is electrically connected to the central controller module and is used for bidirectional communication with the server terminal, receiving control information for the inspection device sent from the server terminal, sending the sound collected by the sound collection array assembly to the server terminal, sending the rail image taken forward by the track tracking camera assembly to the server terminal, and sending status information of the inspection device to the server terminal; 4. The one-wheeled self-balancing track inspection device according to claim 1, characterized in that: The server terminal is wirelessly connected to the unicycle self-balancing track inspection device and is used to: 1) receive the rail image captured by the track tracking camera assembly in the forward direction and detect and identify the rail in real time; 2) receiving status information of the inspection device, sending control information to the inspection device, and remotely controlling the inspection device; 3) receiving and storing sounds collected by the sound collection array component sent by the data communication module in the inspection device, analyzing and detecting the received sounds, and determining the working condition of the rails through sound classification.

5. A method for inspecting a single-wheeled self-balancing railroad track based on sound detection, characterized in that: The steps include: Step S1: The unicycle self-balancing track inspection device is placed on the railway track, and the power module drives the main frame to move on the railway track; Step S2: The inspection device detects and identifies the rail track through the tracking camera assembly. The motion control module uses the gyroscope sensor and flywheel to control the inspection device to maintain a self-balancing state, so that the unicycle self-balancing track inspection device tracks along the railway track. At the same time, the track image captured by the tracking camera assembly is sent to the server terminal, and the status information of the inspection device is sent to the server terminal. Step S3: The inspection device controls the pendulum to strike the rails according to a preset distance or time interval through the sound control and acquisition module; Step S4: The inspection device collects and stores the sound of the pendulum hitting the rails through the sound collection array and transmits it to the server terminal through the data communication module; Step S5: The server terminal uses machine learning or deep learning methods to detect sound, identify sound classification, and determine the working condition of the rails; Step S6: The server terminal receives the rail image sent from the inspection status and detects and identifies the rail in real time; The server terminal receives status information of the inspection device, sends control information to the inspection device, and remotely controls the inspection device.

6. The method for inspecting a single-wheeled self-balancing railroad based on sound detection according to claim 5, characterized in that: The rail working condition sound classification types in step S5 include but are not limited to: normal rail, loose rail, loose screw, inclined rail, and foreign matter on the rail.

Images