An underwater detection video robot system and method based on groundwater

Abstract

The application is suitable for the field of groundwater underwater detection, and provides a groundwater underwater detection video robot system and method, the system comprises: an environment monitoring and processing module, a water flow prediction and analysis module, a path planning and modeling module, a dynamic control and communication module, an anomaly detection and path adjustment module. The method realizes the innovative ability of dynamic path adjustment by continuously monitoring the water flow. Even in the face of sudden environmental changes, the robot can quickly adapt and update the path in time to ensure the smooth progress of the task. By using the virtual model of the water channel, the system can simulate multiple path choices and optimize the final navigation path, thereby effectively avoiding potential obstacles and unfavorable water flow areas. Based on the prediction of water flow changes and real-time analysis of environmental conditions, the system has the function of intelligently adjusting the speed and length of cable winding and unwinding. This not only optimizes resource use, but also improves the overall operation efficiency and safety of the robot.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of groundwater underwater detection, and particularly relates to a groundwater underwater detection video robot system and method. BACKGROUND

[0002] The field of groundwater and underwater detection technology involves various methods and tools for assessing and monitoring groundwater resources and underwater environments. Key technologies in this field include sonar imaging, geological radar, automated water quality sensors, and remote sensing technology. Groundwater underwater detection video robots are specially designed devices for monitoring and assessing groundwater environments. They are equipped with high-definition cameras and various sensors, capable of real-time video capture and data collection in complex groundwater environments. These robots usually have high maneuverability and can navigate in narrow or hard-to-reach areas, avoiding the limitations of human exploration. In addition to video monitoring, their sensors can measure water quality parameters such as temperature, pH, turbidity, and pollutant concentration. These information helps scientists and engineers assess groundwater resources, monitor pollution, and manage the environment.

[0003] In the existing groundwater detection process, the prediction accuracy of traditional methods for underwater conditions is limited, usually relying on static data and historical experience, making it difficult to accurately predict dynamic changes in complex environments, especially in complex groundwater environments. This leads to the system being unable to respond quickly when environmental conditions change, and path adjustments often lag behind actual conditions, increasing the risk of task failure. In addition, existing technologies lack effective real-time risk management mechanisms, and cannot timely issue warnings or adjust strategies when potential threats or obstacles appear. In terms of path planning, traditional methods fail to fully utilize virtual modeling technology, and path selection is often not optimized, easily leading to resource waste and low efficiency. In terms of resource management, the use of cables and other equipment usually relies on the experience of operators, lacking intelligent adjustment mechanisms, and unable to achieve optimal resource allocation in complex environments. SUMMARY

[0004] The purpose of the present application is to provide a groundwater underwater detection video robot system, aiming to solve the technical problems existing in the prior art identified in the background.

[0005] The present application is implemented as follows: a groundwater underwater detection video robot system, the system comprising:

[0006] An environmental monitoring and processing module for continuously monitoring historical environmental data, obtaining real-time water flow speed, direction and turbulence data, and preprocessing the data;

[0007] a water flow prediction and analysis module for integrating current environmental data with historical environmental data, identifying water flow patterns and trends, building a prediction model, predicting upcoming water flow changes and the location of possible obstacles;

[0008] a path planning and modeling module for identifying water flow changes and obstacles that affect the robot's movement based on the water flow change prediction results, building a routing planning model, combining the current waterway layout environment to build a waterway virtual model, and generating a routing path that avoids obstacles and water flow change areas with effects in the waterway virtual model;

[0009] a dynamic control and communication module for calculating and delivering cable deployment and retraction speed and length to the robot based on predicted water flow changes and potential environmental conditions on the new routing path;

[0010] an anomaly detection and path adjustment module for continuously monitoring water flow and issuing water flow change information when unexpected water flow changes and obstacles are detected

[0011] As a further scheme of the present application, the path planning and modeling module comprises:

[0012] a threat identification and classification unit for classifying water flow change and obstacle information generated by the prediction model, identifying potential threats and non-threatening features to the robot;

[0013] a threat ranking unit for prioritizing different threats according to the intensity of water flow changes and the size and location of obstacles, and determining the most critical object to avoid;

[0014] a terrain modeling unit for collecting terrain and structure data of the current waterway, including the fixed positions of existing obstacles, and building a three-dimensional virtual model of the waterway;

[0015] a path planning unit for building a routing planning model and setting benchmark parameters for the planning algorithm, and generating a number of initial routing paths in the three-dimensional virtual model;

[0016] a path screening and optimization unit for screening a number of initial obstacle avoidance paths based on energy consumption, path length, and travel time to obtain pre-selected routing paths and a number of alternative routing paths.

[0017] As a further scheme of the present application, the dynamic control and communication module comprises:

[0018] a water flow mechanics analysis unit for calculating the force of water flow on the robot and cable on different path segments based on water flow speed and direction, and evaluating the impact of possible water flow changes and turbulence on the robot's movement at different positions on the path;

[0019] A cable strategy unit is configured to formulate a strategy of cable winding and unwinding speed and length, and simulate the winding and unwinding operation of the cable in the virtual model to verify the bearing capacity of the cable under different water flow conditions.

[0020] An operation instruction generation unit is configured to convert the calculated cable winding and unwinding strategy into specific operation instructions, including speed, length and adjustment node information.

[0021] Another object of the present application is to provide an underwater detection video robot method based on underground water, which comprises:

[0022] The historical environmental data is continuously monitored, and real-time water flow speed, direction and turbulence data are obtained, and the data is preprocessed;

[0023] The implementation environment data is integrated with the historical environment data, the water flow pattern and the water flow change trend are identified, the prediction model is established, the upcoming water flow change and the possible position of the obstacle are predicted;

[0024] Based on the prediction result of the water flow change, the water flow change and the obstacle affecting the action of the robot are identified, the wiring planning model is established, the virtual model of the waterway is built combined with the current layout environment of the waterway, and the wiring path avoiding the obstacle and the water flow change area with influence is generated in the virtual model of the waterway;

[0025] Based on the predicted water flow change and potential environmental conditions on the new wiring path, the cable winding and unwinding speed and length are calculated and transmitted to the robot;

[0026] The water flow is continuously monitored, and when no unexpected water flow change and obstacle are monitored, an alarm notification containing water flow change information and obstacle information is sent, and the wiring planning model is re-substituted to generate an updated wiring path.

[0027] As a further scheme of the present application, the integration of implementation environment data and historical environment data, identification of water flow pattern and water flow change trend, establishment of prediction model, prediction of upcoming water flow change and possible position of obstacle, specifically includes:

[0028] The real-time collected environmental data and the historical data are aligned in time and space, and different water flow patterns and change rules are identified;

[0029]

[0030] Wherein, X and Y are both item sets, representing environmental and physical condition characteristics, and water quality and ecological result characteristics, The proportion of records in the data set containing both item sets X and Y at the same time, used to measure the universality of the combination, the higher the value, the more frequent the combination appears, count(X U Y) represents the number of records containing both X and Y in the data set, and N represents the total number of all records in the data set;

[0031] The proportion of records containing both X and Y in the records containing X, used to measure the strength of the dependency relationship, count(X) represents the number of records containing X in the data set;

[0032] According to the identified water flow pattern, a water flow state prediction model is established to capture the water flow change trend, identify the upcoming water flow change, and predict the position of the possible obstacle in combination with the environmental conditions.

[0033] As a further scheme of the application, the position of the possible obstacle is predicted in combination with the environmental conditions, specifically:

[0034] Water flow change trend prediction:

[0035] h t =σ(W h ·[h t-1 ,x t ]+b h );

[0036] y t =W y ·h t +b y ;

[0037] Wherein, h t represents the prediction of the water flow state at the next moment, sigma() is an activation function, h t-1 represents the water flow state at the previous moment, x t represents the physical feature data of the current event water flow input, y t represents the estimation of the water flow feature at the next moment, W h and W y are weight matrices, and b h and b y are bias terms.

[0038] Predicting the position of the obstacle:

[0039]

[0040] Wherein, represents the weight of obstacle i at time t, the higher the value, the greater the possibility of obstacle i at this position, represents the position assumption corresponding to obstacle i at time t, represents the position hypothesis of the obstacle i at time t-1, represents the possibility of observing the environment observation data z at the position t , is a transition state transition equation for describing the influence of the environment condition on the obstacle position, u t represents external influencing factors at time t, η t is process noise.

[0041] As a further scheme of the present application, the water flow change prediction result is used to identify the water flow change and the obstacle affecting the robot action, establish a wiring planning model, combine the current waterway layout environment to build a waterway virtual model, and generate a new wiring path in the waterway virtual model to avoid the obstacle and the water flow change area with influence. Specifically, the new wiring path includes:

[0042] The water flow change and obstacle information generated by the prediction model are classified to identify potential threats and non-threatening features to the robot;

[0043] According to the intensity of the water flow change and the size and position of the obstacle, different threats are prioritized to determine the object that needs to be avoided most;

[0044] Collect the terrain and structure data of the current waterway, including the fixed position of the existing obstacle, and build a three-dimensional virtual model of the waterway;

[0045] Establish a wiring planning model and set the benchmark parameters of the planning algorithm to generate a plurality of initial wiring paths in the three-dimensional virtual model;

[0046] Based on energy consumption, path length, and travel time as standards, the plurality of initial obstacle avoidance paths are screened to obtain a preselected wiring path and a plurality of alternative wiring paths.

[0047] As a further scheme of the present application, the water flow change and potential environmental conditions predicted on the new wiring path are used to calculate and deliver the cable winding and unwinding speed and length to the robot. Specifically, the cable winding and unwinding speed and length include:

[0048] Based on the water flow speed and direction, the force of the water flow on the robot and the cable on different path sections is calculated to evaluate the influence of the possible water flow change and turbulence on the robot motion at different positions on the path;

[0049] A strategy for the cable winding and unwinding speed and length is formulated, and the winding and unwinding operation of the cable is simulated in the virtual model to verify the carrying capacity of the cable under different water flow conditions;

[0050] The calculated cable winding and unwinding strategy is converted into specific operation instructions, including speed, length, and adjustment node information.

[0051] As a further aspect of the present invention, the calculation of the force exerted by the water flow on the robot and cables on different path segments based on the water flow velocity and direction specifically includes:

[0052] Calculate drag force:

[0053]

[0054] Among them, F d C represents the drag force, indicating the force exerted by the water flow on the robot and cables. d Here, ρ is the drag coefficient, ρ is the density of water, A is the robot's frontal area, and v is the water flow velocity.

[0055] Calculate lift:

[0056]

[0057] Among them, F l Lift, representing the force perpendicular to the flow direction of water as it flows over the robot and cable surfaces, C. l The lift coefficient;

[0058] As a further embodiment of the present invention, when unpredictable water flow changes and obstacles are detected, an alert notification containing water flow change information and obstacle information is issued, and the information is re-introduced into the wiring planning model to generate an updated wiring path, specifically including:

[0059] By comparing real-time monitoring data with the output of the prediction model, abnormal water flow changes and unpredicted obstacles can be identified. If an anomaly is detected, an alert notification is issued, including detailed information on water flow changes and the location of the obstacle.

[0060] Collect data on abnormal events, including water flow change parameters and obstacle characteristics. Input the new water flow and obstacle information into the cabling planning model, and regenerate new cabling paths in the waterway virtual model to avoid newly detected obstacles and areas of water flow change.

[0061] Adjust the robot's motion control parameters based on the new wiring path.

[0062] The beneficial effects of this invention are:

[0063] This method achieves the innovative ability of dynamic path adjustment by continuously monitoring the water flow. Even in the face of sudden environmental changes, the robot can quickly adapt and update the path in a timely manner to ensure the smooth progress of the task. By utilizing the virtual model of the water channel, the system can simulate multiple path options and optimize the final navigation path, effectively avoiding potential obstacles and unfavorable water flow areas. In addition, based on the prediction of water flow changes and real-time analysis of environmental conditions, the system has the function of intelligently adjusting the cable retraction speed and length. This not only optimizes resource use, but also improves the overall work efficiency and safety of the robot. This comprehensive solution integrating real-time monitoring, virtual modeling, and intelligent resource management provides unprecedented flexibility and precision for underwater detection robots. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 A structural block diagram of an underground water underwater detection video robot system is provided for the embodiments of the present application;

[0065] Figure 2 A structural block diagram of a path planning and modeling module is provided for the embodiments of the present application;

[0066] Figure 3 A structural block diagram of a dynamic control and communication module is provided for the embodiments of the present application;

[0067] Figure 4 A flowchart of an underground water underwater detection video robot method is provided for the embodiments of the present application;

[0068] Figure 5 A flowchart of predicting upcoming water flow changes and the positions of possible obstacles is provided for the embodiments of the present application;

[0069] Figure 6 A flowchart of generating a wiring path that avoids obstacles and water flow change areas with influences in the virtual model of the water channel is provided for the embodiments of the present application;

[0070] Figure 7 A flowchart of calculating and delivering cable retraction speed and length to the robot is provided for the embodiments of the present application;

[0071] Figure 8 A flowchart of issuing a warning notification containing water flow change information and obstacle information when unexpected water flow changes and obstacles are monitored is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0073] Figure 1 A structural block diagram of a groundwater underwater detection video robot system according to an embodiment of the present application is shown in FIG. 1. Figure 1 The system comprises:

[0074] An environmental monitoring and processing module 100 is configured to continuously monitor historical environmental data, acquire real-time water flow speed, direction and turbulence data, and pre-process the data. The robot is equipped with a probe video and can automatically deploy and retract a cable to ensure flexible movement and stable data transmission.

[0075] The video robot supports video recording, storage, photographing and maintenance. The video can be viewed on a display terminal, and the terminal is provided with control buttons. The video content can be uploaded to the background for backup and display. The controller has storage, photographing and video recording functions, and can realize remote control. The size of the controller is 415x340x144.

[0076] The robot is connected to a handheld terminal (such as a mobile phone), and the collected water flow speed, direction and turbulence data are synchronized to the handheld terminal. The handheld terminal displays the current position and wiring path of the robot in real time. Underwater communication transmission can be realized, and the handheld terminal can be used for control and display, and real-time display of underwater conditions, including water flow, seepage, seepage pressure, sand burial conditions and depth information.

[0077] A water flow prediction and analysis module 200 is configured to integrate the implementation environmental data with the historical environmental data, identify the water flow pattern and water flow change trend, establish a prediction model, and predict the upcoming water flow change and the position of possible obstacles;

[0078] A path planning and modeling module 300 is configured to identify the water flow change and obstacles affecting the action of the robot based on the water flow change prediction result, establish a wiring planning model, and build a virtual waterway model combined with the current waterway layout environment. The wiring path that avoids the obstacles and the water flow change area with influence is generated in the virtual waterway model.

[0079] A dynamic control and communication module 400 is configured to calculate and deliver the cable deployment and retraction speed and length to the robot based on the predicted water flow change and potential environmental conditions on the new wiring path. The communication cable has waterproof and shielding functions, and is managed through an automatic retractor and a control switch. The cable is retracted by a disc.

[0080] The anomaly detection and path adjustment module 500 is configured to continuously monitor the water flow and send water flow change information when no unexpected water flow change and obstacle is detected

[0081] As shown in Figure 2 The path planning and modeling module 300 includes:

[0082] The threat identification and classification unit 310 is configured to classify the water flow change and obstacle information generated by the prediction model, and identify potential threats and non-threatening features to the robot.

[0083] The threat ranking unit 320 is configured to prioritize different threats according to the intensity of the water flow change and the size and location of the obstacle, and determine the object that needs to be avoided most.

[0084] The terrain modeling unit 330 is configured to collect terrain and structure data of the current waterway, including the fixed position of existing obstacles, and construct a three-dimensional virtual model of the waterway.

[0085] The path planning unit 340 is configured to establish a routing planning model and set benchmark parameters for the planning algorithm, and generate a number of initial routing paths in the three-dimensional virtual model.

[0086] The path screening and optimization unit 350 is configured to screen a number of initial obstacle avoidance paths based on energy consumption, path length, and transit time as criteria, and obtain pre-selected routing paths and a number of alternative routing paths.

[0087] As shown in Figure 3 The dynamic control and communication module 400 includes:

[0088] The water flow mechanics analysis unit 410 is configured to calculate the force of the water flow on the robot and the cable on different path segments based on the water flow speed and direction, and evaluate the impact of possible water flow changes and turbulence at different positions on the path on the robot's movement.

[0089] The cable strategy formulation unit 420 is configured to formulate strategies for cable retraction and extension speed and length, and simulate cable retraction and extension operations in the virtual model to verify the cable's carrying capacity under different water flow conditions.

[0090] The operation instruction generation unit 430 is configured to convert the calculated cable retraction and extension strategy into specific operation instructions, including speed, length, and adjustment node information.

[0091] Figure 4 A flowchart of a method for detecting underground water using a video robot according to an embodiment of the present application is shown in Figure 4 The method includes:

[0092] S100, continuously monitor historical environmental data, and obtain real-time water flow speed, direction and turbulence data, and preprocess the data;

[0093] S200, integrate the current environmental data with the historical environmental data, identify the water flow patterns and trends of water flow changes, establish a prediction model, predict the upcoming changes in water flow and the possible locations of obstacles;

[0094] This step identifies the water flow patterns and their change rules by aligning the time and space dimensions. This process not only includes the analysis of seasonal changes and periodic fluctuations, but also involves the use of machine learning algorithms to model and predict water flow data, in order to capture the trends of water flow changes and identify upcoming changes in water flow.

[0095] At the same time, combined with environmental conditions such as terrain, vegetation, buildings, etc., the possible locations of obstacles are predicted, and methods such as probability model or Bayesian network are used to evaluate and predict the possibility of obstacles appearing.

[0096] In the model verification and optimization phase, through the analysis of actual monitoring data, the prediction model is continuously adjusted and improved to ensure its accuracy and reliability. Finally, the prediction results are displayed in a visual form, and a warning function is provided to timely issue a warning notification when a possible obstacle is predicted.

[0097] This step significantly improves the accuracy of the prediction, making real-time monitoring and early warning possible, thereby optimizing the wiring planning and improving work efficiency and safety. In addition, the data-driven decision-making process provides scientific basis for users, which helps to make more intelligent decisions, and with the accumulation of data and continuous optimization of the model, the performance of the prediction system is continuously improved, better adapting to actual needs, and providing strong support for underwater detection video robots.

[0098] As shown in Figure 5 , the integration of current environmental data with historical environmental data, identification of water flow patterns and trends of water flow changes, establishment of a prediction model, prediction of upcoming changes in water flow and possible locations of obstacles, specifically includes:

[0099] S210, align the real-time collected environmental data with the historical data in time and space, and identify different water flow patterns and change rules;

[0100]

[0101] wherein X and Y are both item sets, representing environmental and physical condition characteristics, and water quality and ecological result characteristics, The proportion of records in the data set containing both item sets X and Y at the same time, used to measure the universality of the combination of X and Y, the higher the value, the more frequent the combination appears, count(X U Y) represents the number of records containing both X and Y in the data set, and N represents the total number of all records in the data set;

[0102] The proportion of records containing both X and Y in the records containing X, used to measure the strength of the dependency relationship, count(X) represents the number of records containing X in the data set;

[0103] S220, according to the identified water flow pattern, a water flow state prediction model is established to capture the water flow change trend, identify the upcoming water flow change, and predict the position of the possible obstacle in combination with the environmental conditions.

[0104] In this step, the position of the possible obstacle is predicted in combination with the environmental conditions, specifically:

[0105] Water flow change trend prediction:

[0106] h t = sigma(W h ·[h t-1 ,x t ]+b h );

[0107] y t =W y ·h t +b y ;

[0108] Where h t represents the prediction of the next moment water flow state, sigma() is the activation function, h t-1 represents the water flow state at the previous moment, x t represents the current event water flow input physical feature data, y t represents the estimation of the next moment water flow feature, W h and W y are weight matrices, b h and b y are bias terms;

[0109] Predicting the position of the obstacle:

[0110]

[0111] Where, represents the weight of obstacle i at time t, the higher the value, the greater the possibility of obstacle i at this position, represents the position hypothesis corresponding to obstacle i at time t, represents the position assumption of obstacle i at time t-1, represents the position of obstacle i at time t, represents the possibility of observing environment observation data z t at position is a transition equation for describing the influence of environmental conditions on obstacle position, u t represents external influencing factors at time t, η t is process noise.

[0112] S300, based on the prediction results of water flow changes, identifies water flow changes and obstacles affecting the robot's action, establishes a wiring planning model, combines the current layout environment of the waterway to build a virtual model of the waterway, and generates a wiring path in the virtual model of the waterway to avoid obstacles and water flow change areas with influence;

[0113] This step classifies the water flow change and obstacle information generated by the prediction model, identifying which features pose potential threats to the robot and which belong to non-threatening features. This classification process is based on a comprehensive analysis of water flow change intensity and obstacle size and position, ensuring effective identification of the nature and extent of various threats. Then, according to the intensity of the water flow change and the specific characteristics of the obstacle, different threats are prioritized to determine the objects that need to be avoided most, providing targeted obstacle avoidance strategies.

[0114] In addition, by integrating the fixed positions of existing obstacles and other related environmental information, a three-dimensional virtual model of the waterway is constructed. This model not only accurately reflects the layout of the waterway, but also provides necessary spatial information for subsequent path planning. On this basis, a wiring planning model is established, and the benchmark parameters of the planning algorithm are set to ensure the effectiveness and flexibility of the algorithm in practical applications. Subsequently, several initial wiring paths are generated in the three-dimensional virtual model, laying the foundation for subsequent path optimization.

[0115] Finally, based on criteria such as energy consumption, path length, and transit time, the initial obstacle avoidance paths are screened to obtain pre-selected wiring paths and several alternative wiring paths. Through this series of steps, it is ensured that the generated paths not only effectively avoid obstacles, but also achieve optimization in terms of resource consumption and time efficiency.

[0116] This step classifies and prioritizes water flow change and obstacle information to ensure that the robot makes quick and accurate decisions in complex waterway environments, thereby improving safety and efficiency. In addition, the construction of a three-dimensional virtual model provides a realistic and intuitive reflection of the environment for path planning, making the path more reasonable and feasible. Considering energy consumption and transit time, the final wiring path not only has safety, but also significantly reduces energy consumption during execution, improving overall operational efficiency.

[0117] As Figure 6 shown, the predicted results based on water flow changes are used to identify water flow changes and obstacles that affect the robot's movement, establish a wiring planning model, build a virtual model of the waterway based on the current layout of the waterway environment, generate a new wiring path in the virtual model of the waterway that avoids obstacles and areas with water flow changes, and specifically includes:

[0118] S310, classify the water flow changes and obstacle information generated by the prediction model, and identify potential threats and non-threatening features to the robot;

[0119] S320, prioritize different threats according to the intensity of the water flow changes and the size and location of the obstacles, and determine the objects that need to be avoided most;

[0120] S330, collect the terrain and structure data of the current waterway, including the fixed positions of existing obstacles, and build a three-dimensional virtual model of the waterway;

[0121] S340, establish a wiring planning model and set the benchmark parameters of the planning algorithm, and generate several initial wiring paths in the three-dimensional virtual model;

[0122] S350, based on energy consumption, path length, and travel time as criteria, filter the several initial obstacle avoidance paths to obtain pre-selected wiring paths and several alternative wiring paths.

[0123] S400, based on the predicted water flow changes and potential environmental conditions on the new wiring path, calculate and transmit the cable retraction speed and length to the robot;

[0124] This step will calculate the force of the water flow on the robot and the cable on different path segments based on the water flow speed and direction. Specifically, the drag force and lift force calculation formulas are used to represent the drag force and lift force of the water flow on the robot, respectively. Through these calculations, the influence of possible water flow changes and turbulence on the robot's movement at different positions on the path can be evaluated, thereby obtaining the force information on the robot and the cable.

[0125] Next, a strategy for cable retraction speed and length will be developed. This strategy aims to ensure that the cable can maintain appropriate tension under different water flow conditions, avoiding excessive slack or tension. The retraction and extension of the cable are simulated in the virtual model, and the cable's load-bearing capacity under different water flow conditions is verified to ensure that the cable can withstand the predetermined load in actual operation.

[0126] Finally, the calculated cable deployment strategy is converted into specific operation instructions, including speed, length and adjustment node information. These instructions will be directly passed to the robot control system, ensuring that the robot can dynamically adjust the cable deployment strategy according to real-time water flow changes and environmental conditions to adapt to the changing underwater environment.

[0127] By calculating the force of water flow on the robot and cable in detail, scientific basis can be provided for subsequent operations, ensuring the safe and stable operation of the robot in complex underwater environments. At the same time, by formulating detailed cable deployment strategies and simulating them in a virtual model, potential risks can be identified in advance and necessary adjustments can be made, thereby improving the carrying capacity and safety of the cable. In addition, the strategy is converted into specific operation instructions to ensure that the robot can respond to water flow changes in a timely manner, maximizing operational efficiency and safety, thereby making the overall operation more intelligent and efficient.

[0128] As shown in Figure 7 , based on the predicted water flow changes and potential environmental conditions on the new wiring path, the cable deployment speed and length are calculated and transmitted to the robot, specifically including:

[0129] S410, based on the water flow speed and direction, the force of water flow on the robot and cable on different path segments is calculated, and the possible water flow changes and turbulence on the robot movement at different positions on the path are evaluated;

[0130] S420, formulate the strategy of cable deployment speed and length, and simulate the deployment operation of the cable in the virtual model to verify the carrying capacity of the cable under different water flow conditions;

[0131] S430, convert the calculated cable deployment strategy into specific operation instructions, including speed, length and adjustment node information.

[0132] In this step, based on the water flow speed and direction, the force of water flow on the robot and cable on different path segments is calculated, specifically:

[0133] Calculate the drag force:

[0134]

[0135] Where F d is the drag force, representing the force exerted by the water flow on the robot and cable, C d is the drag coefficient, ρ is the density of water, A is the water-facing area of the robot, and v is the water flow speed;

[0136] Calculate the lift:

[0137]

[0138] Where Fl L is the lift force, representing the force perpendicular to the flow direction of the water flow over the robot and the cable surface, C l L is the lift force, representing the force perpendicular to the flow direction of the water flow over the robot and the cable surface, C

[0139] S500, continuously monitor the water flow, and when no predicted water flow changes and obstacles are monitored, issue a warning notification containing water flow change information and obstacle information, and re-substitute the wiring planning model to generate an updated wiring path.

[0140] This step requires comparing real-time monitoring data with the output of the prediction model to accurately identify abnormal water flow changes and unanticipated obstacles. This process relies on high-frequency collection and processing of real-time data to ensure timely capture of environmental changes. When the system detects significant differences between the water flow state and the output of the prediction model, it is considered an abnormal situation. At this time, the system will issue a warning notification containing detailed water flow change information and the location of unanticipated obstacles, so that the operator or automated system can take necessary measures.

[0141] The system will collect data on abnormal events, including water flow change parameters and obstacle characteristics. Data collection is not limited to a single event, but covers multiple data points over a certain period of time for more comprehensive analysis. Inputting these new water flow and obstacle information into the wiring planning model will help update and adjust the model to more accurately reflect the actual situation of the current underwater environment. On this basis, a new wiring path is generated in the virtual model of the waterway to avoid newly detected obstacles and water flow change areas. This process involves path optimization algorithms designed to ensure that the new path generated is both safe and efficient.

[0142] Based on the new wiring path, the system will automatically adjust the motion control parameters of the robot. This includes adjusting the speed, direction of the robot and the cable deployment strategy to ensure that the robot can operate smoothly under new environmental conditions.

[0143] This step enables a quick response to environmental changes by comparing monitoring data and prediction model output in real time, thereby improving the safety of operations. In addition, incorporating abnormal data into the wiring planning model enables the system to continuously learn and optimize, improving the overall level of intelligence. The generation of a new path not only takes into account the current obstacles and water flow changes, but also provides the robot with more flexible response strategies to ensure efficient operation in complex underwater environments. This dynamic adjustment mechanism greatly enhances the adaptability of the robot, enabling it to respond flexibly to changing environments, thereby achieving higher detection efficiency and safety.

[0144] As Figure 8As shown, when no unexpected water flow changes and obstacles are monitored, an alert notification containing water flow change information and obstacle information is issued, and the wiring planning model is re-substituted to generate an updated wiring path, specifically including:

[0145] S510, compare the real-time monitoring data with the output of the prediction model to identify abnormal water flow changes and unexpected obstacles, and if an abnormality is detected, issue an alert notification including detailed water flow change information and the location of the obstacle;

[0146] S520, collect data of the abnormal event, including water flow change parameters and obstacle characteristics, input the new water flow and obstacle information into the wiring planning model, and regenerate a new wiring path in the virtual waterway model that avoids the newly detected obstacle and water flow change area

[0147] S530, adjust the motion control parameters of the robot based on the new wiring path.

[0148] It should be understood that although each step in the flowchart of each embodiment of the present application is displayed in sequence according to the direction of the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least a part of other steps or sub-steps or stages of other steps.

[0149] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0150] Any combination of the technical features of the above-mentioned embodiments can be combined. In order to make the description simple, all possible combinations of the technical features in the above-mentioned embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0151] The above-mentioned embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as limiting the scope of the present application. It should be noted that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

[0152] The above-mentioned embodiments are only the preferred embodiments of the present application, and are not used to limit the present application. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. An underwater groundwater detection video robot system based on, characterized by, The system comprises: An environment monitoring and processing module for continuously monitoring historical environment data and obtaining real-time water flow speed, direction and turbulence data, and pre-processing the data; A water flow prediction and analysis module for integrating current environment data with historical environment data, identifying water flow patterns and water flow change trends, establishing a prediction model, predicting upcoming water flow changes and the positions of possible obstacles; A path planning and modeling module for identifying water flow changes and obstacles affecting the robot's movement based on water flow change prediction results, establishing a wiring planning model, combining the current waterway layout environment to build a waterway virtual model, and generating a wiring path in the waterway virtual model that avoids obstacles and water flow change areas with influence; A dynamic control and communication module for calculating and delivering cable retraction speed and length to the robot based on predicted water flow changes and potential environmental conditions on the new wiring path; An anomaly detection and path adjustment module for continuously monitoring water flow and issuing a warning notification containing water flow change information when no unexpected water flow changes and obstacles are detected; The integration of current environment data with historical environment data, the identification of water flow patterns and water flow change trends, the establishment of a prediction model, and the prediction of upcoming water flow changes and the positions of possible obstacles specifically include: Aligning real-time collected environment data with historical data in time and space, and identifying different water flow patterns and change rules; ； ； in, and Each is an itemset, representing environmental and physical conditions, as well as water quality and ecological outcomes. This indicates that the dataset contains a set of items. and The proportion of records used to measure and A higher frequency of occurrence indicates that the combination of elements appears more frequently. This indicates that the dataset contains both and The number of records, This represents the total number of records in the dataset; denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing According to the identified water flow patterns, establish a water flow state prediction model, capture the water flow change trend, identify the upcoming water flow change, and predict the possible position of the obstacle combined with the environmental conditions.

2. The system of claim 1, wherein, The robot is connected to the handheld terminal and synchronizes the collected water flow speed, direction and turbulence data to the handheld terminal, and the handheld terminal displays the current position of the robot and the wiring path in real time.

3. A method for underwater detection of a video robot based on groundwater, characterized by, The method comprises: Continuously monitoring historical environment data and obtaining real-time water flow speed, direction and turbulence data, and pre-processing the data; Integrating current environment data with historical environment data, identifying water flow patterns and water flow change trends, establishing a prediction model, and predicting upcoming water flow changes and the positions of possible obstacles; Based on water flow change prediction results, identify water flow changes and obstacles affecting the robot's movement, establish a wiring planning model, combine the current waterway layout environment to build a waterway virtual model, and generate a wiring path in the waterway virtual model that avoids obstacles and water flow change areas with influence; Based on the predicted water flow changes and potential environmental conditions on the new wiring path, calculate and deliver cable retraction speed and length to the robot; Continuously monitor the water flow and issue a warning notification containing water flow change information and obstacle information when no unexpected water flow changes and obstacles are detected, and re-enter the wiring planning model to generate an updated wiring path; The integration of current environment data with historical environment data, the identification of water flow patterns and water flow change trends, the establishment of a prediction model, and the prediction of upcoming water flow changes and the positions of possible obstacles specifically include: Align the real-time collected environmental data with the historical data in time and space, and identify different water flow patterns and change rules; ； ； wherein, and are itemsets, respectively representing environmental and physical condition features, and water quality and ecological outcome features, represents the proportion of records in the dataset that contain both itemsets and and is used to measure the prevalence of the combination of and the higher the value of the prevalence of the combination, the more frequent the combination appears, represents the number of records in the dataset that contain both and represents the total number of records in the dataset.​ denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing denotes the proportion of records in the data set containing According to the identified water flow pattern, a water flow state prediction model is established to capture the water flow change trend, identify the upcoming water flow change, and predict the possible position of the obstacle in combination with the environmental conditions.

4. The method of claim 3, wherein, The prediction of the position of the obstacle in combination with the environmental conditions is specific: Water flow change trend prediction: ； ； wherein, represents a prediction of the water flow state at the next time instant, is an activation function, represents the water flow state at the previous time instant, represents the physical feature data of the water flow input of the current event, represents an estimate of the water flow feature at the next time instant, and is a weight matrix, and is a bias term; Prediction of the position of the obstacle: ； ； wherein, represents an obstacle at time , the higher the weight, the greater the likelihood of an obstacle at that location, represents an obstacle at time , represents an obstacle at time , represents the likelihood of observing environment observation data at location is a transition state transition equation describing the influence of environmental conditions on the location of the obstacle, represents an external influencing factor at time , is process noise.

5. The method of claim 4, wherein, Based on the water flow change prediction result, the water flow change and the obstacle affecting the robot's action are identified, a wiring planning model is established, a virtual waterway model is built combining the current waterway layout environment, and a new wiring path is generated in the virtual waterway model to avoid the obstacle and the water flow change area with influence, which specifically includes: Classify the water flow change and obstacle information generated by the prediction model to identify the potential threat and non-threat features to the robot; According to the intensity of the water flow change and the size, position of the obstacle, the different threats are prioritized to determine the object that needs to be avoided most; Collect the terrain and structure data of the current waterway, including the fixed position of the existing obstacles, and build a three-dimensional virtual model of the waterway; Establish a wiring planning model and set the benchmark parameters of the planning algorithm to generate a number of initial wiring paths in the three-dimensional virtual model; Based on energy consumption, path length and travel time as the standard, a number of initial obstacle avoidance paths are screened to obtain pre-selected wiring paths and a number of alternative wiring paths.

6. The method of claim 5, wherein, Based on the predicted water flow change and potential environmental conditions on the new wiring path, the cable reeling speed and length are calculated and transmitted to the robot, which specifically includes: Based on the water flow speed and direction, the force of the water flow on the robot and the cable in different path sections is calculated to evaluate the influence of the possible water flow change and turbulence on the robot's movement in different positions on the path; Develop a cable reeling speed and length strategy and simulate the cable reeling operation in the virtual model to verify the cable's carrying capacity under different water flow conditions; Convert the calculated cable reeling strategy into specific operation instructions, including speed, length and adjustment node information.

7. The method of claim 6, wherein, Based on the water flow speed and direction, the force of the water flow on the robot and the cable in different path sections is calculated, which specifically includes: Calculate the drag force: ； wherein, is the drag force, representing the force exerted by the water flow on the robot and the cable, is the drag coefficient, is the density of water, is the wetted surface area of the robot, is the water flow velocity; Calculate the lift: ； wherein, is the lift force, representing the force of the water flow on the surface of the robot and the cable perpendicular to the direction of the flow, is the lift coefficient.

8. The method of claim 7, wherein, When no unexpected water flow change and obstacle are monitored, an alert notification containing water flow change information and obstacle information is issued, and the wiring planning model is re-substituted to generate an updated wiring path, which specifically includes: Compare the real-time monitoring data with the output of the prediction model to identify abnormal water flow change and un-predicted obstacles. If an anomaly is detected, an alert notification is issued, including detailed water flow change information and the position of the obstacle; Collect data of the abnormal event, including water flow change parameters and obstacle features, input new water flow and obstacle information into the wiring planning model, and regenerate a new wiring path in the virtual waterway model to avoid the newly detected obstacle and water flow change area; Adjust the motion control parameters of the robot based on the new wiring path.

Images