Karst landform cave detection method and underwater robot

Abstract

The application discloses a karst landform cave detection method and an underwater robot, and the method comprises the following steps: 1) entering water and initial deployment; 2) non-powered drift navigation; 3) cave observation and data collection; 4) resistance judgment and self-rescue control; 5) continuous drift and path recording; and 6) a three-dimensional cave construction module. The application has the advantages of compact structure, reliable waterproof, high navigation precision, resistance self-rescue capability and three-dimensional modeling post-processing capability, and is suitable for narrow and complex underwater environment detection.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of underwater detection, and particularly relates to a karst landform cave detection method and underwater robot. BACKGROUND

[0002] Karst landform is a landform formed by the dissolution and deposition of soluble rocks, the erosion and deposition of surface water and underground water, and the gravity collapse, collapse and accumulation, and is also called karst landform. Karst pipeline is an important storage and migration channel of underground water, and efficient and accurate detection and description of karst pipeline of underground water have important significance for water finding and underground water pollution prevention and control in karst mountainous areas.

[0003] Due to the complex distribution of karst pipeline network and underground river water network, often accompanied by narrow and closed space, and the three-no characteristics of no light, no prior map and no stable communication signal, the current karst pipeline survey is still in the primary stage, and the acquisition of original data mainly relies on traditional geophysical prospecting and engineering means, mainly including drilling, geological radar, electrical method (such as resistivity method), seismic wave method and ground laser scanning. Although these methods are relatively mature in shallow structure survey, they have problems of low resolution, limited detection range and inability to enter complex underwater environment.

[0004] Therefore, it is an urgent problem for those skilled in the art to provide an efficient, accurate and controllable karst landform cave detection method and underwater robot. SUMMARY

[0005] Therefore, the application provides a karst landform cave detection method and underwater robot to make up for the shortcomings of traditional detection devices and methods and improve the detection efficiency and accuracy.

[0006] In order to achieve the above purpose, the application adopts the following technical solutions:

[0007] A karst landform cave detection method, comprising the following steps:

[0008] 1) Water entry and initial deployment

[0009] The underwater robot is launched into the water body at the entrance of the karst cave;

[0010] 2) Unpowered drift navigation

[0011] The underwater robot drifts freely by relying on the natural water flow of underground water, and forms an unpowered drift working mode, wherein the navigation module adopts a three-source fusion algorithm for relative positioning solution;

[0012] 3) Cave observation and data acquisition

[0013] The binocular camera collects video images in the forward direction, and the searchlight is used to provide active illumination for the forward field of view. After all the images and original data of the navigation module are packaged and encoded, they are saved in the storage module in the control and power supply module, or real-time data is returned through the communication module interface or remote access is performed;

[0014] 4) Obstruction judgment and self-help control

[0015] The drift is stopped by detecting speed abnormalities through the Doppler speedometer, and the abnormal rotation or floating overturn is recognized by detecting the angular velocity through the inertial measurement unit. In combination with the optical flow anomaly and image recognition, it is determined whether it is in a dead angle or blind area. Once it is determined to be in a difficult situation, the control and power supply module controls the start of the walking propeller, so that the underwater robot gets out of trouble and restores the natural drifting state.

[0016] 5) Continuous drifting and path recording

[0017] After getting out of trouble, continue to drift with the water, and repeatedly complete image acquisition, optical flow positioning, and inertial navigation solution. The structure of the solution cave, the flow direction characteristics, and the path form are recorded segment by segment.

[0018] 6) Three-dimensional solution cave construction module

[0019] After completing a complete solution cave drifting task, the recorded binocular images, inertial measurement unit, optical flow sensor, Doppler speedometer speed information, and magnetic field auxiliary information are exported to the upper computer for processing. First, a lightweight visual SLAM framework is used to match features and estimate inter-frame poses for binocular images. In combination with the data detected by the inertial measurement unit and the Doppler speedometer, the trajectory is corrected and time-synchronized. When some visual frames fail, the optical flow sensor displacement estimation is introduced to maintain the continuity of the trajectory. Then, using the synchronized trajectory and image data, a dense three-dimensional point cloud model of the solution cave is generated by structure light restoration, dense matching, or depth map fusion methods.

[0020] Further, the specific process of the three-source fusion algorithm in step 2) is as follows:

[0021] 1) Optical flow visual perception

[0022] The optical flow sensor continuously captures the visual features of the solution cave top, water surface light and shadow, and skylight structure. By calculating the pixel displacement of feature points between adjacent frames and the sampling time interval , the relative velocity vector is obtained:

[0023]

[0024] wherein, is the camera imaging scale factor;

[0025] 2) Inertial Measurement Unit

[0026] Bottom integrated high sampling rate tri-axial gyroscope and accelerometer, output angular velocity and linear acceleration , attitude angle Obtained by angular velocity integration:

[0027]

[0028] Where, linear acceleration , describes the linear acceleration of the object in three-dimensional space, is the acceleration along the axis direction, is the acceleration along the axis direction, is the acceleration along the axis direction,

[0029] represents the attitude angle at the current time , its three components is the roll angle, is the pitch angle, is the yaw angle,

[0030] represents the attitude angle at the initial time , which is the initial condition or reference value of integration,

[0031] represents the lower limit of integration, represents the upper limit of integration,

[0032] : represents the three-axis angular velocity vector measured by the gyroscope in the inertial measurement unit , which is a function of time τ, where, angular velocity around the axis, angular velocity around the axis, angular velocity around the axis;

[0033] 3) Doppler velocity meter

[0034] Downward emission of multiple sound waves, measurement of relative velocity by reflection of echo waves through water bottom or suspended particles, output of three-dimensional velocity vector ;

[0035] Relative displacement can be obtained by velocity integration:

[0036] ; ​

[0037] in, This is relative displacement.

[0038] The three-dimensional velocity vector measured by DVL;

[0039] 4) Multi-source data fusion

[0040] The standard discrete-time Kalman filter algorithm is introduced into the data fusion process to predict and update multi-source sensor data. Its mathematical expression is as follows:

[0041] =

[0042]

[0043]

[0044]

[0045] =(I )

[0046] In the state prediction formula: for Predicted state estimation at time; for -1 hour arrives The state transition matrix at time t; For a moment -1 state estimation;

[0047] In the covariance prediction formula: covariance matrix express The covariance of the estimation error at time -1; express The prediction error covariance at time; obtained after measurement update express Error covariance after update at time step; The process noise input matrix; for -1 time step noise covariance;

[0048] In the Kalman gain formula: for The Kalman gain at time step 1 is a weight matrix; for Time-based observation matrix; for Observe the noise covariance at all times;

[0049] In the state update equation, is the posterior state estimate at time t; is the actual measurement at time t; is the predicted measurement at time t; is the actual measurement at time t; is the measurement residual, the difference between the actual observation and the predicted observation is the measurement residual, the difference between the actual observation and the predicted observation

[0050] In the covariance update equation, is the updated error covariance.

[0051] An underwater robot for performing the method as described above, comprising:

[0052] a butterfly-shaped body, a lower surface of the butterfly-shaped body being provided with buoyancy material;

[0053] a detection part, the detection part comprising a detection module and a navigation module installed inside the butterfly-shaped body;

[0054] a walking propeller, the walking propeller being installed on the butterfly-shaped body;

[0055] a control part, the control part comprising a control and power supply module and a walking driving module installed inside the butterfly-shaped body, the detection module and the navigation module being electrically connected to the control and power supply module respectively, the control and power supply module being electrically connected to the walking driving module, and the walking driving module being electrically connected to the walking propeller.

[0056] Further, the butterfly-shaped body comprises an upper shell, a skeleton and a lower shell, the upper shell, the skeleton and the lower shell being stacked in order from top to bottom and connected together through a plurality of circumferentially distributed screws, so as to form a mounting space between the upper shell, the skeleton and the lower shell; the detection module is installed on the skeleton; the navigation module is installed on the skeleton, the upper shell and the lower shell; the walking propeller is installed on the lower shell; the control and power supply module and the walking driving module are both installed on the skeleton; and the buoyancy material is coated on the outer surface of the lower shell.

[0057] Further, the detection module comprises a binocular camera, a circumferential wall surface of the skeleton has a detection hole, and a waterproof plate is installed on the outer end portion or the inner portion of the detection hole; the binocular camera is installed on an upper platform of the skeleton close to the upper shell, and the photographing direction of the binocular camera corresponds to the position of the detection hole; and the binocular camera is electrically connected to the control and power supply module.

[0058] Further, the detection module further comprises a searchlight, the searchlight is installed on the top of the upper shell, and the irradiation direction of the searchlight is the same as the camera direction of the binocular camera.

[0059] Further, the navigation module comprises an optical flow sensor, an inertial measurement unit and a Doppler velocity log, the optical flow sensor is vertically installed in the middle of the upper shell through a waterproof shell, the inertial measurement unit is installed on the skeleton, and the Doppler velocity log is installed in the middle of the lower shell; the optical flow sensor, the inertial measurement unit and the Doppler velocity log are electrically connected with the control and power supply module.

[0060] Further, the navigation module further comprises an LED light supplementing lamp, the LED light supplementing lamp is installed on the upper shell and located on both sides of the optical flow sensor.

[0061] Further, the upper shell and the skeleton and the lower shell and the skeleton are both paved with sealing rings, and the inertial measurement unit, the control and power supply module and the walking driving module are all peripherally installed with sealing covers.

[0062] Therefore, the karst landform cave detection method and the underwater robot have the following beneficial effects compared with the prior art:

[0063] (1) Water power and structural performance breakthrough

[0064] The disc-shaped body design significantly reduces water flow resistance and improves the passing performance in narrow and curved caves (the minimum passing diameter is less than or equal to 30 cm); the skeleton type layered sealing structure (sealing ring + sealing cover secondary protection) realizes IP68 level waterproofing, and ensures the long-term reliability of the core sensor in high-pressure and turbid water bodies;

[0065] (2) High-precision navigation in GPS-free environment

[0066] The optical flow sensor (upward observation of the cave top / water surface features) + MEMS-IMU + DVL three-source fusion navigation architecture is adopted, the sensor drift is inhibited in real time through Kalman filtering, the position estimation error is less than 5% of the distance, the attitude angle accuracy is ±0.1°, and the positioning and orientation problems in the cave without light, map and communication are solved;

[0067] (3) Intelligent escape

[0068] The unpowered drift mode reduces energy consumption (the endurance is improved by more than 40%), the walking thruster pulse propulsion is triggered only when resistance is encountered, and disturbance of sediments is avoided; based on the multi-sensor state monitoring (DVL stall alarm, IMU attitude anomaly and optical flow feature loss), the self-rescue system can respond within 200ms and realize a 90% or more stuck escape rate.

[0069] (4) three-dimensional space reconstruction capability

[0070] Synchronous recording of binocular vision, IMU, optical flow displacement and DVL velocity data, post-processing stage through vision-inertial tight coupling SLAM and multi-source data fusion, generation of centimeter-precision cave three-dimensional point cloud model, making up the defect that traditional geophysical prospecting method cannot obtain three-dimensional structure;

[0071] (5) engineering application value improvement

[0072] More than 60% reduction in drilling / prospecting cost, single task can detect ≥3km underground river; modular design supports quick battery / sensor replacement, reuse rate >95%; multi-machine cooperative mode can realize parallel exploration of karst cave network, efficiency improved by more than 3 times. BRIEF DESCRIPTION OF DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of the provided drawings.

[0074] Figure 1 An isometric view of a underwater robot suitable for karst landform cave detection provided by the present application;

[0075] Figure 2 A left view of a underwater robot suitable for karst landform cave detection provided by the present application;

[0076] Figure 3 A bottom view of a underwater robot suitable for karst landform cave detection provided by the present application;

[0077] Figure 4 An isometric view of a underwater robot suitable for karst landform cave detection provided by the present application, with the upper shell removed;

[0078] Figure 5 An isometric view of a underwater robot suitable for karst landform cave detection provided by the present application, with the lower shell removed. DETAILED DESCRIPTION

[0079] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0080] like Figures 1-5 As shown in the figure, an embodiment of the present invention discloses a method for detecting karst caves, including the following steps:

[0081] 1) Water entry and initial deployment

[0082] The underwater robot was deployed into the water at the entrance of the karst cave. Because the lower shell 13 was coated with buoyancy material, the underwater robot was able to quickly achieve automatic horizontal attitude fixation and float stably on the water surface or in the middle layer of the water body, ready to start the drifting exploration mission.

[0083] 2) Non-powered drifting navigation

[0084] The underwater robot drifts freely using the natural currents of underground water bodies, forming a non-powered drifting operation mode. Its navigation module employs a three-source fusion algorithm for relative positioning calculation.

[0085] S1 Optical Flow Visual Perception

[0086] The optical flow sensor 6 continuously captures visual features of the cave ceiling, water surface light and shadow, and skylight structure, calculating the pixel displacement of feature points between adjacent frames. and sampling time interval The relative velocity vector is obtained. :

[0087]

[0088] in, The camera imaging ratio is used; in low-light environments, the optical flow sensor 6 can be combined with the built-in LED fill light 10 to improve imaging contrast and feature recognition accuracy.

[0089] S2 Inertial Measurement Unit 7

[0090] The bottom integrates a high-sampling-rate three-axis gyroscope and accelerometer, outputting angular velocity. With linear acceleration attitude angle Obtained by integrating angular velocity:

[0091]

[0092] Among them, linear acceleration , describes the linear acceleration of the object in three-dimensional space, is the acceleration along the axis direction, is the acceleration along the axis direction, is the acceleration along the axis direction,

[0093] represents the attitude angle at the current time , whose three components are the roll angle, is the pitch angle, is the yaw angle,

[0094] represents the attitude angle at the initial time , which is the initial condition or reference value of integration,

[0095] represents the lower limit of integration, represents the upper limit of integration,

[0096] : represents the three-axis angular velocity vector measured by the gyroscope in the inertial measurement unit as a function of time τ, where, is the angular velocity around the axis, is the angular velocity around the axis, is the angular velocity around the axis;

[0097] Modeling and suppressing sensor noise and drift during integration using Kalman filtering can significantly improve the stability and accuracy of attitude calculation;

[0098] S3 Doppler velocity meter 8

[0099] Downwardly emitting multiple sound waves, measuring relative velocity through reflection of echo waves by the water bottom or suspended particles, outputting three-dimensional velocity vector ;

[0100] The relative displacement can be obtained by integrating the velocity:

[0101] ;

[0102] where, is the relative displacement,

[0103] is the three-dimensional velocity vector measured by the DVL;

[0104] ​During long-term operation, DVL measurements exhibit low-frequency drift, which can be corrected by using optical flow velocity and MEMS predicted velocity, thereby reducing integration error.

[0105] S4 Multi-Source Data Fusion

[0106] In the data fusion process, the standard discrete-time Kalman filter algorithm is introduced to predict and update multi-source sensor data. Kalman filtering calculates the covariance between the prior state and the prediction error based on the system state transition model during the prediction phase, and corrects the prior estimate using observation information during the update phase. This allows for the suppression of sensor noise and drift while obtaining a stable and highly accurate state estimate. Its mathematical expression is as follows:

[0107] =

[0108]

[0109]

[0110]

[0111] =(I )

[0112] In the state prediction formula: for Predicted state estimation at time; for -1 hour arrives The state transition matrix at time t; For a moment -1 state estimation;

[0113] In the covariance prediction formula: covariance matrix express The covariance of the estimation error at time -1; express The prediction error covariance at time; obtained after measurement update express Error covariance after update at time step; The process noise input matrix; for -1 time step noise covariance;

[0114] In the Kalman gain formula: for The Kalman gain at time step 1 is a weight matrix; for Time-based observation matrix; is the observation noise covariance at time k;

[0115] in the state update equation, is the posterior state estimate at time k; is the actual measurement at time k; is the measurement residual, the difference between the actual observation and the predicted observation

[0116] in the covariance update equation, is the updated error covariance;

[0117] 3) Observation and data collection in the cave

[0118] The binocular camera 31 collects video images in the forward direction, and the searchlight 32 is used to provide active illumination for the forward field of view. After all the images and the original data of the navigation module are packaged and encoded, they are saved in the storage module in the control and power supply module 4, or they are transmitted in real time or accessed remotely through the communication module interface;

[0119] 4) Obstruction judgment and self-rescue control

[0120] The drift stop is identified by detecting abnormal speed through the Doppler speedometer 8, and the abnormal rotation or floating overturn is identified by detecting angular velocity through the inertial measurement unit 7. In combination with the optical flow anomaly and image recognition, it is determined whether it is in a dead angle or blind area. Once it is determined to be in a difficult situation (stuck, rapid change of water flow, stasis rotation), the control and power supply module 4 controls the start of the walking thruster 2, so that the underwater robot escapes from the predicament and restores the natural drifting state. The water pump nozzle is arranged at a preset angle towards the back, generating a forward directional thrust; the thrust direction is automatically determined and selected according to the current attitude, or a time-sharing trial switching mode is used; the thrust size is output in a predetermined pulse mode, such as a short pulse of the left pump first, then the right pump, and then both pumps; the pulse length is 100-500 ms; the walking thruster 2 provided in the invention ensures that the equipment has the ability to escape from the predicament autonomously under abnormal conditions, and enhances the adaptability of the system in narrow, winding, unknown and other extreme environments;

[0121] 5) Continuous drifting and path recording

[0122] After escaping from the predicament, continue to drift with the water, and repeatedly complete image acquisition, optical flow positioning, and inertial navigation calculation, to record the cave structure, flow direction characteristics, and path form segment by segment; optionally, positioning calibration or trigger mode switching is performed when passing through specific recognition points (such as infrared markers, artificial labels, flow velocity boundaries, etc.);

[0123] 6) Three-dimensional cave construction module

[0124] ​After completing a complete cave drifting task, the binocular images recorded throughout the journey, the inertial measurement unit 7, the optical flow sensor 6, the speed information of the Doppler speedometer 8 and the magnetic field auxiliary information are exported to the upper computer for processing. First, the lightweight visual SLAM framework is used to perform feature matching and inter-frame pose estimation on the binocular images, and the data detected by the inertial measurement unit 7 and the Doppler speedometer 8 are used for trajectory correction and time synchronization; when part of the visual frame is invalid, the optical flow sensor 6 displacement estimation is introduced to maintain the trajectory continuity; subsequently, using the synchronized trajectory and image data, a dense three-dimensional point cloud model of the cave is generated by structure light restoration, dense matching or depth map fusion method. The three-dimensional model can be used for subsequent scientific analysis, path planning, cave structure modeling and other tasks. This mode does not depend on real-time computing power and is suitable for embedded platforms with limited device resources. The present application realizes stable positioning and three-dimensional space structure reconstruction of karst cave (karst pipeline) under the condition of no artificial intervention by fusing the multi-source navigation module composed of optical flow sensor 6, inertial measurement unit 7 and Doppler speedometer 8, and combining with the post-processing modeling method based on three-dimensional point cloud construction.

[0125] The embodiment of the present application also discloses an underwater robot for executing the method as described above, capable of detecting underground karst cave (karst pipeline) with a diameter greater than 50 cm, comprising a hydrodynamically optimized butterfly-shaped body 1, a detection part, a walking propeller 2 and a control part, the ratio of the diameter to the depth of the butterfly-shaped body 1 is 3:2, the lower surface of the butterfly-shaped body 1 is provided with buoyancy material for providing positive buoyancy and building stable moment to realize automatic righting and failure self-floating; the detection part comprises a detection module 3 and a navigation module installed inside the butterfly-shaped body 1; the walking propeller 2 is installed on the butterfly-shaped body 1, in this embodiment, the walking propeller 2 is selected as a water pump for adjusting the forward direction of the butterfly-shaped body 1 and driving the butterfly-shaped body 1 to move forward; the control part comprises a control and power supply module 4 and a walking driving module 5 installed inside the butterfly-shaped body 1, the detection module 3 and the navigation module are electrically connected with the control and power supply module 4, the control and power supply module 4 is electrically connected with the walking driving module 5, and the walking driving module 5 is electrically connected with the walking propeller 2, the control and power supply module 4 is configured to automatically activate the water pump to provide directional thrust for escape or attitude adjustment when the underwater robot is detected to be stuck or in an abnormal motion state, and the water pump can also be remotely activated by a communication module. The present application has a compact structure, can freely shuttle in complex pipelines and caves, reach areas difficult for humans to reach, has the ability to rely on natural drift of water flow and auxiliary low-interference pulse propulsion (self-rescue ability), and has comprehensive information acquisition, can accurately obtain the trend, scale and structure information of karst pipeline and cave by carrying the detection module 3 and the navigation module, expands the adaptability of traditional underwater robots in narrow, complex flow, insufficient light environment, and can be widely applied to underground cave detection, water flow analysis and underwater space modeling scenes, and has significant scientific research and application value.

[0126] Specifically, the butterfly-shaped body 1 comprises an upper shell 11, a framework 12 and a lower shell 13, the upper shell 11, the framework 12 and the lower shell 13 are stacked from top to bottom and connected together through a plurality of circumferentially distributed screws, so as to form a mounting space between the upper shell 11, the framework 12 and the lower shell 13; the detection module 3 is installed on the framework 12; the navigation module is installed on the framework 12, the upper shell 11 and the lower shell 13; the walking propeller 2 is installed on the lower shell 13; the control and power supply module 4 and the walking driving module 5 are both installed on the framework 12, the control and power supply module 4 comprises a control module and a power supply module which are electrically connected, the power supply module is selected as a lithium battery assembly as a power supply, is centrally arranged near the gravity axis of the underwater robot and maintains the balance of the overall weight; a buoyancy material is coated on the outer surface of the lower shell 13, and the buoyancy material is annular or sheet-shaped.

[0127] In the embodiment, the framework 12 is disc-shaped, a plurality of equidistantly arranged through holes are arranged on the periphery of the framework 12, and threaded nests are arranged in the through holes; the upper shell 11 and the lower shell 13 are provided with through holes at corresponding positions, and the through holes are aligned with the through holes on the framework 12; each screw passes through the through holes on the upper shell 11 and the lower shell 13 and is fastened and connected with the threaded nests on the framework 12, so as to press and combine the upper shell 11, the framework 12 and the lower shell 13.

[0128] Specifically, the detection module 3 comprises a binocular camera 31, the circumferential wall surface of the framework 12 has a detection hole 121, and a waterproof plate is installed at the outer end portion or the inner portion of the detection hole 121, in the embodiment, the waterproof plate is transparent glass; the binocular camera 31 is installed on the upper platform of the framework 12 close to the upper shell 11, and the photographing direction of the binocular camera 31 corresponds to the position of the detection hole 121, is used for photographing the internal structure of the cave, the flow direction of the water body or the obstacle and providing visual information acquisition; the binocular camera 31 is electrically connected with the control and power supply module 4.

[0129] In order to further optimize the technical scheme of the application, the detection module 3 further comprises a searchlight 32, the searchlight 32 is installed on the top of the upper shell 11, and the irradiation direction of the searchlight 32 is the same as the photographing direction of the binocular camera 31, which improves the image quality of the photographing, and is particularly suitable for dark, muddy or irregular structure areas.

[0130] Specifically, the navigation module comprises an optical flow sensor 6, an inertial measurement unit 7 (IMU) and a Doppler velocity log 8 (DVL), the optical flow sensor 6 is vertically installed in the middle of the upper shell 11 through a waterproof shell 9, and is used for detecting the change of the water surface feature observed in the drift of the cave roof structure or through the water body; the inertial measurement unit 7 is installed on the skeleton 12, and is used for real-time attitude angle measurement; the Doppler velocity log 8 is installed in the middle of the lower shell 13, and is used for measuring the relative ground speed vector, thereby improving the underwater autonomous positioning capability; the optical flow sensor 6, the inertial measurement unit 7 and the Doppler velocity log 8 are electrically connected with the control and power supply module 4.

[0131] In order to further optimize the technical scheme of the present application, the navigation module further comprises an LED light supplement lamp 10, which is installed on the upper shell 11 and located on both sides of the optical flow sensor 6.

[0132] Specifically, the upper shell 11 and the skeleton 12, and the lower shell 13 and the skeleton 12 are both paved with sealing rings, in the embodiment, the outer edge top surface and the bottom surface of the skeleton 12 are both provided with annular grooves, and O-shaped sealing rings are embedded in the annular grooves, so that the O-shaped sealing rings are located between the skeleton 12 and the upper shell 11, and between the contact surfaces of the skeleton 12 and the lower shell 13, and are compressed by the pre-tightening force of the screws; the inertial measurement unit 7, the control and power supply module 4 and the walking driving module 5 are all installed with sealing covers.

[0133] The present application adopts the technologies of disc-shaped hydrodynamic force + skeleton sealing integration, three-source fusion navigation, unpowered drift + water pump self-rescue, post-processing modeling, etc., solves the inherent defects of the traditional karst pipeline detection mode in task flexibility, efficiency and safety, and is suitable for karst pipeline detection and other complex underwater operation scenes.

[0134] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0135] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of detecting a karst landform cave, characterized by, Comprising the following steps: 1) Water entry and initial deployment The underwater robot is launched into the water at the entrance of the karst cave; 2) Unpowered drift navigation The underwater robot relies on the natural water flow of the underground water body to drift freely, forming an unpowered drift working mode, wherein the navigation module uses a three-source fusion algorithm for relative positioning calculation; 3) Cave observation and data collection The binocular camera collects video images in the forward direction, and the searchlight is used to provide active illumination for the forward field of view. All images and navigation module raw data are saved through the storage module in the control and power supply module, or real-time data transmission or remote access through the communication module interface; 4) Obstruction judgment and self-rescue control Abnormal drift stop is detected by the Doppler speedometer, and abnormal rotation or floating overturn is detected by the inertial measurement unit. Combined with light flow anomaly and image recognition, it is judged whether it is in a dead angle or blind area. Once it is judged to be in a difficult situation, the control and power supply module controls the start of the walking propeller to make the underwater robot escape and restore the natural drift state; 5) Continuous drift and path recording After escaping, continue to drift with the water, repeat image collection, optical flow positioning, and inertial navigation calculation, and record the cave structure, flow direction characteristics, and path form; 6) Three-dimensional cave construction module After completing a complete cave drifting task, the recorded binocular images, inertial measurement unit, optical flow sensor, Doppler speedometer speed information, and magnetic field auxiliary information are exported to the host computer for processing. First, use the lightweight visual SLAM framework to match features and estimate inter-frame poses for binocular images. Combined with the data detected by the inertial measurement unit and the Doppler speedometer, the trajectory is corrected and time-synchronized; When some visual frames fail, introduce the optical flow sensor displacement estimation to maintain the continuity of the trajectory; Subsequently, using the synchronized trajectory and image data, a dense three-dimensional point cloud model of the cave is generated by structure light restoration, dense matching, or depth map fusion methods.

2. The method according to claim 1, wherein, The specific process of the three-source fusion algorithm in step 2) is as follows: 1) Optical flow visual perception The optical flow sensor continuously captures the features on the top of the cave, the water surface light and shadow, and the visual features of the skylight structure, calculates the pixel displacement of the feature points between adjacent frames and the sampling time interval to obtain the relative velocity vector : ； wherein, is a camera imaging scale factor; 2) Inertial measurement unit Bottom integrated high sampling rate tri-axial gyroscope and accelerometer, output angular velocity with linear acceleration , attitude angle Obtained by integrating angular velocity: ； wherein the linear acceleration of the object in three-dimensional space is described by the acceleration along the axis, the acceleration along the axis, the acceleration along the axis, represents the attitude angle at the current time whose three components are the roll angle, the pitch angle, the yaw angle,​ denotes the attitude angle at the initial time is the initial condition or reference value for the integration, denotes the lower limit of the integral, denotes the upper limit of the integral, denotes a three-axis angular velocity vector measured by a gyroscope in the inertial measurement unit a function of time τ, where, angular velocity about the z-axis, angular velocity about the z-axis, angular velocity about the z-axis; 3) Doppler speedometer Downward firing multi-beam sound waves, measure relative velocity by reflection off the water bottom or suspended particles, output three-dimensional velocity vector ; Relative displacement can be obtained by velocity integration: ； wherein is the relative displacement, Three-dimensional velocity vector measured for the DVL; 4) Multi-source data fusion In the data fusion process, introduce the standard discrete-time Kalman filter algorithm to predict and update the multi-source sensor data, and the mathematical expression is as follows: = ； ； ； ； =(I ) ； State prediction formulae: is the predicted state estimate at time is the state estimate at time is the state transition matrix from time is the state estimate at time -1.​​ Covariance prediction formula: Covariance matrix denotes the estimation error covariance at time -1; denotes the prediction error covariance at time -1; the updated error covariance at time -1 denotes the updated error covariance at time -1; is the process noise input matrix; is the the process noise covariance at time -1; In the Kalman gain formula: for The Kalman gain at time step 1 is a weight matrix; for Time-based observation matrix; for Observe the noise covariance at all times; In the state update equation, is the actual measurement value at time is the posterior state estimate at time is the actual measurement value at time is the actual measurement value at time is the measurement residual, which is the difference between the actual observation and the predicted observation is the measurement residual, which is the difference between the actual observation and the predicted observation In the covariance update formula, is the updated error covariance.

3. An underwater robot for performing the method of claim 1 or 2, characterized in that, Comprising: A butterfly-shaped body, the lower surface of the butterfly-shaped body is provided with a buoyancy material; A detection part, the detection part includes a detection module and a navigation module installed inside the butterfly-shaped body; A walking propeller, the walking propeller is installed on the butterfly-shaped body; A control part, the control part includes a control and power supply module and a walking drive module installed inside the butterfly-shaped body, the detection module and the navigation module are electrically connected with the control and power supply module, the control and power supply module is electrically connected with the walking drive module, and the walking drive module is electrically connected with the walking propeller.

4. An underwater robot according to claim 3, wherein, The delta-shaped body comprises an upper shell, a skeleton and a lower shell, which are stacked from top to bottom and connected together by a plurality of circumferentially distributed screws, thereby forming a mounting space between the upper shell, the skeleton and the lower shell; the detection module is mounted on the skeleton; the navigation module is mounted on the skeleton, the upper shell and the lower shell; the walking propeller is mounted on the lower shell; the control and power supply module and the walking driving module are both mounted on the skeleton; the buoyancy material is coated on the outer surface of the lower shell.

5. An underwater robot according to claim 4, wherein, The detection module comprises a binocular camera, the circumferential wall surface of the skeleton has a detection hole, and a waterproof plate is mounted on the outer end or the inner part of the detection hole; the binocular camera is mounted on the upper platform of the skeleton close to the upper shell, and the photographing direction of the binocular camera corresponds to the position of the detection hole; the binocular camera is electrically connected with the control and power supply module.

6. An underwater robot according to claim 5, wherein, The detection module further comprises a searchlight, which is mounted on the top of the upper shell, and the irradiation direction of the searchlight is the same as the photographing direction of the binocular camera.

7. The underwater robot of claim 4, wherein, The navigation module comprises an optical flow sensor, an inertial measurement unit and a Doppler velocimeter, the optical flow sensor is vertically mounted in the middle part of the upper shell through a waterproof shell; the inertial measurement unit is mounted on the skeleton; the Doppler velocimeter is mounted in the middle part of the lower shell; the optical flow sensor, the inertial measurement unit and the Doppler velocimeter are electrically connected with the control and power supply module respectively.

8. The underwater robot of claim 7, wherein, The navigation module further comprises an LED light supplementing lamp, which is mounted on the upper shell and located on both sides of the optical flow sensor.

9. The underwater robot of claim 7, wherein, Sealing rings are arranged between the upper shell and the skeleton, and between the lower shell and the skeleton; sealing covers are mounted on the periphery of the inertial measurement unit, the control and power supply module and the walking driving module.

Images