A control method and device of an underwater robot, the underwater robot, and a medium

Abstract

The application discloses a kind of underwater robot control method, device, underwater robot, medium, applied to underwater robot field, the method includes: by obtaining multi-source data parameters, and according to multi-source data parameters, establish the coupling model of cable, obstacle and water flow;According to the coupling model, the cable movement trajectory in future time period is predicted, and the cable winding probability in future preset time period is calculated;When cable winding probability exceeds the preset medium risk threshold value, according to multi-source data parameters, establish the local map for current environment;According to cable winding probability and local map, establish winding risk map;In winding risk map, target path search is carried out, and target path is generated;According to target path, control underwater robot movement, complete to cable full section Real-time prediction, reach pre-judgment and active avoidance winding risk, improve the protection efficiency, for winding event Timely response processing.

Description

Technical Field

[0001] This invention relates to the field of underwater robots, and in particular to a control method, device, underwater robot, and media technology for underwater robots. Background Technology

[0002] Remotely operated vehicles (ROVs) are increasingly used in marine engineering, particularly in offshore wind farms, subsea pipelines, and marine exploration for the inspection and maintenance of underwater structures. However, ROVs rely on umbilical cables for power, data transmission, and control signals. These cables are prone to entanglement in complex underwater environments, such as jacket structures and seabed debris, leading to operational interruptions, cable damage, or even complete equipment loss. Cable entanglement has become a major bottleneck restricting the reliability and safety of ROVs, especially in high-velocity (>2 knot) and turbid waters, where entanglement accidents account for 20-30% of ROV failures, causing significant economic losses.

[0003] Existing technologies mainly focus on mechanical management and passive protection, such as adding cable guide wheels, slip rings, or towing floats. However, these methods can only alleviate local entanglement and cannot predict entanglement, resulting in poor adaptability of ROVs in complex environments, low protection efficiency, and inability to respond to emergencies in a timely manner. Summary of the Invention

[0004] In view of the above problems, a control method, apparatus, underwater robot, and medium for an underwater robot are proposed to overcome or at least partially solve the above problems, including: In a first aspect, embodiments of the present invention provide a control method for an underwater robot, the method comprising: Acquire multi-source data parameters, and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface; Based on the coupling model, the cable trajectory in the future time period is predicted, and the probability of cable entanglement in the future preset time period is calculated. When the probability of cable entanglement exceeds a preset medium-risk threshold, a local map for the current environment is established based on the multi-source data parameters. Based on the cable entanglement probability and the local map, an entanglement risk map is established; wherein, the entanglement risk map is a map with an entanglement probability distribution; In the entanglement risk map, a target path is searched and a target path is generated; The underwater robot is controlled to move according to the target path.

[0005] Optionally, based on the multi-source data parameters, a coupling model of cables, obstacles, and water flow is established, including: Based on the multi-source data parameters, the coupling features of the cable, obstacle, and water flow are extracted; wherein, the coupling features are the coupling relationship between the cable, obstacle, and water flow in the spatiotemporal dimension; Based on the aforementioned coupling characteristics, a coupling model of cables, obstacles, and water flow is established.

[0006] Optionally, in the entanglement risk map, a target path search is performed to generate a target path, including: A multi-constraint objective function is constructed with minimizing the probability of cable entanglement as the optimization objective. In the entanglement risk map, multiple candidate nodes are identified, and a target path is generated based on the candidate nodes and the multi-constraint objective function.

[0007] Optionally, in the entanglement risk map, multiple candidate nodes are identified, and a target path is generated based on the candidate nodes and the multi-constraint objective function, including: In the entanglement risk map, low-entanglement risk areas and high-entanglement risk areas are identified, and multiple candidate nodes are generated in the low-entanglement risk areas and the high-entanglement risk areas. Calculate the change in entanglement probability for each candidate node; Based on the change in entanglement probability and the multi-constraint objective function, the target node is determined from multiple candidate nodes; Generate the target path based on the target node.

[0008] Optionally, the method further includes: When the probability of cable entanglement exceeds a preset low-risk threshold, a preset auxiliary thruster is activated. The underwater robot's attitude is adjusted by the auxiliary thruster, and the mechanical state and length of the cable are also adjusted.

[0009] Optionally, the cable includes a core cable core, a load-bearing cable segment, and auxiliary functional cores, and the method further includes: When the probability of cable entanglement exceeds a preset high-risk threshold, or when the probability of entanglement exceeds a preset safety threshold during a preset time period when the underwater robot moves along the target path, a first response is executed; wherein, the first response is to cut off a portion of the load-bearing cable segment through a controllable release device.

[0010] Optionally, the method further includes: When the probability of cable entanglement exceeds a preset extremely high risk threshold, or when the probability of entanglement exceeds a safety threshold within a preset time period after the execution of the first response, a second response is executed; wherein, the second response is to cut off all non-core cable cores through a controllable release device and control the underwater robot to float up through a buoyancy airbag.

[0011] The embodiments of the present invention have the following advantages: In this embodiment of the invention, multi-source data parameters are acquired, and a coupling model of cable, obstacle, and water flow is established based on these parameters. The cable connects the underwater robot to a control console on the water surface. Based on the coupling model, the cable's trajectory is predicted for a future time period, and the probability of cable entanglement within a preset future time period is calculated. When the cable entanglement probability exceeds a preset medium-risk threshold, a local map of the current environment is established based on the multi-source data parameters. An entanglement risk map is then established based on the cable entanglement probability and the local map. This entanglement risk map displays an entanglement probability distribution. A target path is searched within the entanglement risk map to generate a target path. The underwater robot's movement is controlled according to the target path. This achieves real-time prediction of the entire cable segment through the establishment of the coupling model, enabling the prediction and proactive avoidance of entanglement risks. This improves adaptability and protection efficiency in complex environments, allowing for timely response to entanglement events. Attached Figure Description

[0012] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of the steps of a control method for an underwater robot provided in some embodiments of the present invention; Figure 2 This is a flowchart of the steps of a second control method for an underwater robot provided in some embodiments of the present invention; Figure 3 This is a flowchart of the steps of a control method three for an underwater robot provided in some embodiments of the present invention; Figure 4 This is a flowchart of the steps of a control method for an underwater robot provided in some embodiments of the present invention; Figure 5 This is a flowchart illustrating the steps of a control device for an underwater robot provided in some embodiments of the present invention. Detailed Implementation

[0014] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0015] Reference Figure 1 The diagram illustrates a step flowchart of a control method for an underwater robot according to some embodiments of the present invention, which may specifically include the following steps: Step 101: Obtain multi-source data parameters and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface.

[0016] In some examples, the multi-source data parameters include one or more of the following: the mechanical state of the cable, the attitude of the remotely operated underwater vehicle (ROV), the environmental parameters of the water flow in which the ROV is located, and the location of obstacles; the cable is an umbilical cable that provides power, data transmission, and control signals to the ROV.

[0017] In practical applications, multiple sensors are deployed within the underwater robot's range of motion. These pre-installed multi-source sensors monitor and collect data on the robot, cables, and environment. The results from these multi-source sensors are fused and transmitted to a neural network. The neural network extracts parameters from the multi-source data to establish a coupling model between the cables, obstacles, and water flow. For example, distributed tension and torque sensor arrays are deployed at equal intervals along the entire length of the umbilical cable. Combined with the inertial measurement unit, current meter, and depth gauge mounted on the underwater robot, this allows for real-time capture of stress and torque distribution along the entire cable length, the underwater robot's full attitude data, and water flow environmental parameters, providing multi-dimensional, full-space monitoring input. When fusing the multi-source sensor data, an adaptive Kalman filter is used to preprocess noise. The filter parameters can be dynamically adjusted according to the environmental noise level to ensure data accuracy. Compared to fixed-parameter filtering, the signal-to-noise ratio is improved by more than 40%, outliers are removed, and time synchronization is achieved.

[0018] In some embodiments, if the stress / torque at any location on the cable exceeds a preset initial threshold, it is immediately marked as a potential risk event, and the data is preferentially transmitted to the cable prediction system. The sampling frequency can be dynamically adjusted according to the operating conditions; the sampling frequency is 10Hz for normal operating conditions, and up to 50Hz high-frequency sampling is supported for high-risk operating conditions.

[0019] In some embodiments of the present invention, a coupling model of cables, obstacles, and water flow is established based on the multi-source data parameters, including: Sub-step 11: Extract the coupling features of the cable, obstacle and water flow based on the multi-source data parameters; wherein the coupling features are the coupling relationship between the cable, obstacle and water flow in the spatiotemporal dimension.

[0020] In some examples, the coupling relationship is that the cable, obstacle and water flow are geometrically constrained and mutually influenced in space, and their states evolve synchronously and dynamically in time. The three do not exist independently, but form a relationship of mutual constraint and linkage.

[0021] After acquiring multi-source data parameters, these parameters are preprocessed and then transmitted to a neural network. The neural network performs spatial convolution on the mechanical state of the cable, the attitude of the ROV relative to the obstacle, and the water flow distribution to extract the coupling features of these three factors. For example, the aforementioned multi-source data is input into a CNN-LSTM (Convolutional Neural Network-Long Short-Term Memory hybrid neural network). The CNN (Convolutional Neural Network) part performs spatial convolution on the spatial distribution of cable stress / torque, the relative spatial position of the ROV and the obstacle, and the water flow field distribution to extract the spatial coupling features of "cable-obstacle-water flow" (when a cable segment is simultaneously subjected to lateral thrust from the water flow and geometric constraints from the obstacle, its entanglement risk characteristics are significantly different from those under the influence of only a single factor). The LSTM (Long Short-Term Memory) part receives the spatial feature sequence output by the CNN and extracts the evolutionary trends and patterns in the time dimension.

[0022] Sub-step 12: Based on the coupling characteristics, establish a coupling model of the cable, obstacle and water flow.

[0023] In some examples, the coupling model is used to characterize the geometric constraints between the cable, obstacles, and water flow, and to quantify and extrapolate the cable's motion posture, water flow, and entanglement evolution process.

[0024] After extracting the coupling characteristics of the cable, obstacles, and water flow, the coupling relationship between the cable, obstacles, and water flow is solidified into a dynamic simulation model. The model is then trained based on the acquired historical multi-source data parameters to output the final coupling model. For example, in the training of the dynamic coupling model, the training dataset consists of two parts: first, historical entanglement event data, including labeled entanglement / non-entanglement samples from real operations; second, simulation data. By establishing a cable dynamic simulation model (considering the mechanical properties of the cable's flexible body, fluid resistance, and obstacle collision constraints), a large number of simulation samples are generated under different sea conditions covering the entire flow velocity range of 0-4 knots to expand the training set. The model is trained under supervised learning with the entanglement probability as the output target.

[0025] Step 102: Based on the coupling model, predict the cable movement trajectory in the future time period and calculate the probability of cable entanglement in the future preset time period.

[0026] After obtaining the coupled model, real-time multi-source data parameters are received. Based on these parameters, the model simulates the motion trajectory of each cable segment and the interaction evolution between the cable, obstacles, and water flow over a future time period, calculating and outputting the entanglement probability distribution for each part of the cable. For example, after training, the model runs in real-time on an embedded processor, inputting the latest multi-source data parameters in a sliding time window manner to simulate the motion trajectory of each cable segment and its interaction evolution with obstacles and water flow over the next 10 seconds, outputting the entanglement probability distribution for each location along the entire cable segment, serving as the basis for subsequent path optimization and emergency response decisions.

[0027] Step 103: When the probability of cable entanglement exceeds a preset medium-risk threshold, a local map for the current environment is established based on the multi-source data parameters.

[0028] After obtaining the probability of cable entanglement, the risk value of entanglement is calculated based on the probability of cable entanglement. The risk value is compared with a preset threshold. When the risk value is greater than the preset medium risk threshold, it is determined that there is a medium risk of entanglement in the current operation. Based on multi-source data parameters, a model of the local working environment around the current underwater robot is established to obtain a local map including obstacles and water flow distribution.

[0029] Step 104: Based on the cable entanglement probability and the local map, establish an entanglement risk map; wherein the entanglement risk map is a map with an entanglement probability distribution.

[0030] After establishing a local map, the obtained entanglement probabilities of each cable segment are superimposed onto the local map to obtain a map with an entanglement probability distribution. For example, based on obstacle location and geometric information obtained from sonar / visual sensors, combined with real-time water flow parameters measured by a current meter, a local map of the current operating environment is established. The cable is a physical line suspended in the underwater space, and each segment of the cable corresponds to a real spatial location on the underwater map. By assigning the entanglement probability of the cable segment to its corresponding water area coordinates, it can be superimposed onto the map. The output entanglement probabilities of each cable segment are superimposed onto the map to form an "entanglement risk map".

[0031] Step 105: In the entanglement risk map, perform a target path search and generate a target path.

[0032] After generating the entanglement risk map, a path search is performed on the entanglement risk map to obtain an optimal path with low entanglement risk, safe obstacle avoidance, and meeting the operational requirements. This optimal path is then determined as the target path for the underwater robot to execute.

[0033] In some embodiments of the present invention, target path search is performed in the entanglement risk map to generate a target path, including: Sub-step 21: With minimizing the probability of cable entanglement as the optimization objective, construct a multi-constraint objective function.

[0034] In some examples, the multi-constraint objective function includes at least the constraints of the probability of cable entanglement along the entire cable length, the distance cost between the underwater robot and the target point, the obstacle avoidance safety margin, the underwater robot's power consumption, and the remaining available length of the cable.

[0035] In practical applications, minimizing the probability of cable entanglement is taken as the optimization objective. Based on multiple constraints, the impact of different factors on the risk of cable entanglement is quantified by setting the weights of each constraint, and a calculable multi-constraint objective function is constructed. For example, a multi-constraint objective function is constructed with minimizing the probability of entanglement along the entire cable segment as the core optimization objective. This objective function comprehensively considers constraints such as: the probability of entanglement at each location along the entire cable segment (output in real time by the risk assessment module), the distance cost between the ROV and the target work point, obstacle avoidance safety margin, ROV power consumption, and the remaining usable length of the umbilical cable.

[0036] Sub-step 22: In the entanglement risk map, determine multiple candidate nodes, and generate a target path based on the candidate nodes and the multi-constraint objective function.

[0037] In some examples, candidate nodes are generated as discrete spatial route points within the feasible region of the entangled risk map.

[0038] In practical applications, discrete route points that meet the reachability conditions of underwater robot movement are selected as candidate transit points in the entanglement risk map. Based on the multi-constraint objective function, the final target node is selected from these candidate nodes, and the target nodes are linked to form the target path.

[0039] In some embodiments of the present invention, multiple candidate nodes are identified in the entanglement risk map, and a target path is generated based on the candidate nodes and the multi-constraint objective function, including: Sub-step 221: Determine low-winding-risk areas and high-winding-risk areas in the winding-risk map, and generate multiple candidate nodes in the low-winding-risk areas and the high-winding-risk areas.

[0040] After generating the entanglement risk map, it is divided into low-risk and high-risk entanglement regions based on the entanglement probability on the map. On the map after region division, multiple candidate nodes are sampled and generated biased towards the low-entanglement regions. For example, an improved Rapid Expanding Random Tree (RRT*) algorithm is used to perform path search on the entanglement risk cost map, and when randomly expanding tree nodes, the sampling probability is biased towards the low-entanglement risk regions.

[0041] Sub-step 222: Calculate the change in entanglement probability for each candidate node.

[0042] After obtaining candidate nodes, the underwater robot is simulated moving from its current position to each candidate node. The difference between the cable entanglement probability and the original probability is calculated, and the change in entanglement probability for each candidate node is determined based on the difference.

[0043] Sub-step 223: Based on the change in entanglement probability and the multi-constraint objective function, determine the target node from multiple candidate nodes.

[0044] After obtaining the change in entanglement probability, the change in entanglement probability of each node and multiple constraints are substituted into the multi-constraint objective function to calculate the comprehensive cost of each candidate node. The point with the minimum comprehensive cost and the lowest entanglement risk among all candidate nodes is selected as the target node.

[0045] Sub-step 224: Generate the target path based on the target node.

[0046] After obtaining the target node, connect and process all target nodes to generate an optimal path, i.e., the target path.

[0047] Step 106: Control the underwater robot to move according to the target path.

[0048] After obtaining the target path, the target path is processed to generate instructions for controlling the underwater robot, which then moves along the target path. For example, the searched path is smoothed to ensure that the underwater robot's motion trajectory is continuous and executable (meeting the underwater robot's kinematic and dynamic constraints), and thruster control instructions are generated and sent to the underwater robot's motion control system for execution. In some embodiments, after determining the target path, the omnidirectional 360° vector control thruster drives the ROV to move according to the planned trajectory (usually in the opposite direction of the cable's contact with the obstacle), achieving trajectory redirection to release the cable's contact tendency with the obstacle; simultaneously, the tensioning mechanism releases redundant cables, reduces cable tension, and collaboratively eliminates the risk of entanglement.

[0049] In some embodiments, changes in the overall risk value are continuously monitored during path execution. If a sudden change occurs in the environment (such as a sudden increase in water flow velocity or the appearance of new obstacles), the algorithm re-executes steps two through four within 200ms to iteratively update the planned trajectory.

[0050] In this embodiment of the invention, by establishing a coupling model, real-time prediction of the entire cable segment is achieved, thereby predicting and actively avoiding the risk of entanglement, improving adaptability and protection efficiency in complex environments, and enabling timely response to entanglement events.

[0051] Reference Figure 2 The diagram illustrates a flowchart of a second control method for an underwater robot according to some embodiments of the present invention, which may specifically include the following steps: Step 201: Obtain multi-source data parameters and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface.

[0052] Step 202: Based on the coupling model, predict the cable movement trajectory in the future time period and calculate the probability of cable entanglement in the future preset time period.

[0053] Step 203: When the probability of cable entanglement exceeds a preset medium-risk threshold, a local map for the current environment is established based on the multi-source data parameters.

[0054] Step 204: Based on the cable entanglement probability and the local map, establish an entanglement risk map; wherein the entanglement risk map is a map with an entanglement probability distribution.

[0055] Step 205: In the entanglement risk map, perform a target path search and generate a target path.

[0056] Step 206: Control the underwater robot to move according to the target path.

[0057] Step 207: When the probability of cable entanglement exceeds a preset low-risk threshold, activate the preset auxiliary thruster; Step 208: Adjust the attitude of the underwater robot and the mechanical state and length of the cable through the auxiliary thruster.

[0058] In some examples, the auxiliary thruster is a small attitude control propulsion unit equipped on the underwater robot. It does not undertake long-distance travel tasks but is used for small-angle attitude correction, micro-displacement adjustment, and local flow field disturbance. It precisely changes the spatial attitude and local force environment of the underwater robot, thereby controlling the tension state, suspension shape and spatial swing of the cable.

[0059] After obtaining the probability of cable entanglement, the risk value of entanglement is calculated based on the probability of cable entanglement. The risk value is compared with a preset threshold. When the risk value exceeds the preset low-risk threshold, the preset auxiliary thruster is activated to adjust the roll, pitch, and yaw attitude of the underwater robot. At the same time, the cable tensioning mechanism is adjusted to control the cable tension and redundancy.

[0060] In this embodiment of the invention, by establishing a coupling model, real-time prediction of the entire cable segment is achieved, thereby predicting and actively avoiding the risk of entanglement, improving adaptability and protection efficiency in complex environments, and enabling timely response to entanglement events.

[0061] Reference Figure 3 The diagram shows a flowchart of a third control method for an underwater robot according to some embodiments of the present invention. The cable includes a core cable core, a load-bearing cable segment, and auxiliary functional cable cores.

[0062] In some examples, the core cable is the main core of the entire cable, which is composed of high-voltage power lines, fiber optic communication lines and main control signal lines twisted together. It is responsible for power supply and main communication. The load-bearing cable section is wrapped around the core cable to bear the cable's own weight tension, water current drag force and underwater robot's motor traction tension. The auxiliary function cable core consists of sensor lines, spare signal lines and auxiliary control lines to provide backup channels.

[0063] Specifically, it may include the following steps: Step 301: Obtain multi-source data parameters and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface.

[0064] Step 302: Based on the coupling model, predict the cable movement trajectory in the future time period and calculate the probability of cable entanglement in the future preset time period.

[0065] Step 303: When the probability of cable entanglement exceeds a preset medium-risk threshold, a local map for the current environment is established based on the multi-source data parameters.

[0066] Step 304: Based on the cable entanglement probability and the local map, establish an entanglement risk map; wherein the entanglement risk map is a map with an entanglement probability distribution.

[0067] Step 305: In the entanglement risk map, perform a target path search and generate a target path.

[0068] Step 306: Control the underwater robot to move according to the target path.

[0069] Step 307: When the probability of cable entanglement exceeds a preset low-risk threshold, activate the preset auxiliary thruster; Step 308: Adjust the attitude of the underwater robot and the mechanical state and length of the cable through the auxiliary thruster.

[0070] Step 309: When the probability of cable entanglement exceeds a preset high-risk threshold, or when the probability of entanglement exceeds a preset safety threshold during a preset time period when the underwater robot moves along the target path, a first response is executed; wherein, the first response is to cut off part of the load-bearing cable segment through a controllable release device.

[0071] In some examples, the core cable is the main body of the entire cable, composed of high-voltage power lines, fiber optic communication lines, and main control signal lines twisted together. It is responsible for power supply and main communication. The load-bearing cable section is wrapped around the core cable, bearing the cable's own weight, water current drag, and underwater robot's traction force. The auxiliary function cable core consists of sensor lines, backup signal lines, and auxiliary control lines, providing backup channels. The controllable release device is an electrically controlled intelligent cutting mechanism installed on the cable, used to precisely cut designated sections of the cable.

[0072] After obtaining the probability of cable entanglement, the risk value of entanglement is calculated based on the probability. This risk value is then compared with a preset threshold. When the risk value exceeds the preset high-risk threshold, or when the probability of entanglement exceeds the preset safety threshold within a preset time period during the underwater robot's movement along the target path, a controllable release device cuts a portion of the load-bearing cable segment, releasing the overall tension of the cable and loosening it from entanglement with obstacles. For example, if the risk value does not drop below the safety threshold within the set observation time window after path replanning, or if the risk value directly exceeds the high-risk threshold, the first response is triggered. The electromagnetic locking controllable release device selectively cuts off portions of non-core load-bearing and redundant cable cores, retaining the core power supply, communication, and control cable cores, thus releasing cable tension and entanglement and maximizing the ROV's operational and return capabilities.

[0073] In this embodiment of the invention, by establishing a coupling model, real-time prediction of the entire cable segment is achieved, thereby predicting and actively avoiding the risk of entanglement, improving adaptability and protection efficiency in complex environments, and enabling timely response to entanglement events.

[0074] Reference Figure 4 The diagram illustrates a step-by-step flowchart of a control method for an underwater robot according to some embodiments of the present invention, which may specifically include the following steps: Step 401: Obtain multi-source data parameters and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface.

[0075] Step 402: Based on the coupling model, predict the cable movement trajectory in the future time period and calculate the cable entanglement probability in the future preset time period.

[0076] Step 403: When the probability of cable entanglement exceeds a preset medium-risk threshold, a local map for the current environment is established based on the multi-source data parameters.

[0077] Step 404: Based on the cable entanglement probability and the local map, establish an entanglement risk map; wherein the entanglement risk map is a map with an entanglement probability distribution.

[0078] Step 405: In the entanglement risk map, perform a target path search and generate a target path.

[0079] Step 406: Control the underwater robot to move according to the target path.

[0080] Step 407: When the probability of cable entanglement exceeds a preset low-risk threshold, activate the preset auxiliary thruster; Step 408: Adjust the attitude of the underwater robot and the mechanical state and length of the cable through the auxiliary thruster.

[0081] Step 409: When the probability of cable entanglement exceeds a preset high-risk threshold, or when the probability of entanglement exceeds a preset safety threshold during a preset time period when the underwater robot moves along the target path, a first response is executed; wherein, the first response is to cut off a portion of the load-bearing cable segment through a controllable release device.

[0082] Step 410: When the probability of cable entanglement exceeds a preset extremely high risk threshold, or when the probability of entanglement exceeds a safety threshold within a preset time period after the execution of the first response, the second response is executed; wherein, the second response is to cut off all non-core cable cores through a controllable release device and control the underwater robot to float up through a buoyancy airbag.

[0083] After obtaining the probability of cable entanglement, the risk value of entanglement is calculated based on the probability of cable entanglement. The risk value is compared with a preset threshold. When the risk value exceeds the preset extremely high risk threshold, or when the probability of entanglement exceeds the safety threshold within a preset time period after the execution of the first response, all load-bearing cable segments and auxiliary function wires are cut off through a preset controllable release device, and the preset buoyancy airbag is activated, and the underwater robot automatically rises to the surface and evacuates from the dangerous waters.

[0084] In some embodiments, after obtaining the probability of cable entanglement, the risk assessment module calculates the current entanglement risk value based on the probability of cable entanglement, compares the risk value with a preset classification threshold, and executes a preset level of response strategy if the risk value exceeds one of the classification thresholds. For example, the risk level-based intervention action includes three core actuators: an emergency auxiliary power unit (omnidirectional vector thruster), an electromagnetic locking controllable cable segment release device, and an inflatable buoyancy airbag, with a built-in four-level progressive response logic of "attitude fine-tuning → tension release → precise cutting → emergency buoyancy". If the risk is low, the auxiliary thrusters are activated to fine-tune the underwater robot's attitude, and the cable tension is adjusted simultaneously to eliminate potential risks. If the risk is medium, dynamic path replanning is initiated to generate the optimal escape trajectory, and the omnidirectional thrusters are activated to drive the underwater robot to move according to the planned trajectory. The cable tensioning mechanism releases redundant cables to eliminate the risk of entanglement. If the risk is high, the controllable release device selectively cuts some non-core cable cores, retaining the core power supply and communication cable cores to maximize the underwater robot's operational capabilities. If the risk is extremely high, all non-core cable cores are cut, and buoyancy airbags are deployed simultaneously to assist the underwater robot in surfacing, ensuring the safety of high-value sensors such as high-definition cameras and sonar equipment, as well as the robot itself. The module supports manual overlay mode, and action priorities can be customized.

[0085] In some embodiments, the method further includes a control center, which is an embedded industrial computer responsible for module coordination, data storage, and uplink / downlink communication. The center integrates a full-process log system to record comprehensive data, including risk events, intervention actions, intervention results, and prediction deviations, for later analysis and model self-learning, and supports remote diagnostic interfaces. The center also handles abnormal operating conditions, such as prioritizing the underwater robot recovery process when the power supply is low. The control center has a built-in online self-learning engine that can automatically extract entanglement risk feature labels based on the operation logs, update AI model parameters through incremental learning algorithms, and automatically optimize model weights in reverse for false alarms / missed alarms, achieving closed-loop iterative optimization of the model without manual parameter tuning.

[0086] In some embodiments, after a single operation is completed, the control center's online self-learning engine analyzes the entire process log, extracts risk event characteristics, intervention effects, and prediction deviation data, and constructs an entanglement risk feature label library. Through incremental learning algorithms, the coupled model is adaptively updated, and for false alarms and missed alarms, the feature weight parameters of the dynamically coupled model are automatically optimized in reverse. For new operating areas and conditions with limited sample data, model adaptation and optimization can be quickly completed without manual parameter readjustment. This achieves a closed-loop process of "data acquisition - risk prediction - intervention execution - log analysis - model optimization," continuously improving the system's prediction accuracy and environmental adaptability.

[0087] In this embodiment of the invention, by establishing a coupling model, real-time prediction of the entire cable segment is achieved, thereby predicting and actively avoiding the risk of entanglement, improving adaptability and protection efficiency in complex environments, and enabling timely response to entanglement events.

[0088] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0089] Reference Figure 5 The diagram shows a structural schematic of a control device for an underwater robot according to some embodiments of the present invention, which may specifically include the following modules: The coupling model establishment module 501 is used to acquire multi-source data parameters and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface; The cable entanglement probability calculation module 502 is used to predict the cable movement trajectory in a future time period based on the coupling model, and to calculate the cable entanglement probability in a future preset time period. The map building module 503, when the probability of cable entanglement exceeds a preset medium-risk threshold, builds a local map for the current environment based on the multi-source data parameters. The entanglement risk map establishment module 504 is used to establish an entanglement risk map based on the cable entanglement probability and the local map; wherein, the entanglement risk map is a map with an entanglement probability distribution; The target path generation module 505 is used to search for target paths in the entanglement risk map and generate target paths. The underwater robot motion module 506 is used to control the movement of the underwater robot according to the target path.

[0090] In some embodiments of the present invention, the coupling model establishment module 501 includes: The coupling relationship extraction submodule is used to extract the coupling features of the cable, obstacle and water flow based on the multi-source data parameters; wherein, the coupling features are the coupling relationship between the cable, obstacle and water flow in the spatiotemporal dimension; The coupling model establishment submodule is used to establish a coupling model of cables, obstacles and water flow based on the coupling characteristics.

[0091] In some embodiments of the present invention, the target path generation module 505 includes: The multi-constraint objective function submodule is used to construct a multi-constraint objective function with the goal of minimizing the probability of cable entanglement. The target path generation submodule is used to identify multiple candidate nodes in the entanglement risk map and generate a target path based on the candidate nodes and the multi-constraint objective function.

[0092] In some embodiments of the present invention, the target path generation submodule includes: A candidate node generation unit is used to determine low-winding-risk areas and high-winding-risk areas in the winding-risk map, and to generate multiple candidate nodes in the low-winding-risk areas and the high-winding-risk areas. The entanglement probability change calculation unit is used to calculate the entanglement probability change of each candidate node; The target node determination unit is used to determine the target node from multiple candidate nodes based on the change in the entanglement probability and the multi-constraint objective function; The target path generation unit is used to generate a target path based on the target node.

[0093] In some embodiments of the present invention, the apparatus further includes: An auxiliary thruster activation module is used to activate a preset auxiliary thruster when the probability of cable entanglement exceeds a preset low-risk threshold. The cable adjustment module is used to adjust the attitude of the underwater robot via the auxiliary thruster, and to adjust the mechanical state and length of the cable.

[0094] In some embodiments of the present invention, the cable includes a core cable core, a load-bearing cable segment, and auxiliary functional cores, and the device further includes: The first response module is used to execute a first response when the probability of cable entanglement exceeds a preset high-risk threshold, or when the probability of entanglement exceeds a preset safety threshold during a preset time period when the underwater robot moves along the target path; wherein, the first response is to cut off a portion of the load-bearing cable segment through a controllable release device.

[0095] In some embodiments of the present invention, the apparatus further includes: The second response module is used to execute a second response when the probability of cable entanglement exceeds a preset extremely high risk threshold, or when the probability of entanglement exceeds a safety threshold within a preset time period after the execution of the first response; wherein, the second response is to cut off all non-core cable cores through a controllable release device and control the underwater robot to float up through a buoyancy airbag.

[0096] In this embodiment of the invention, a coupled model of cable, obstacle, and water flow is established to calculate the probability of cable entanglement, and an emergency response is initiated based on this probability. This achieves real-time prediction of the entire cable segment through the establishment of the coupled model, enabling the anticipation and proactive avoidance of entanglement risks. This improves adaptability and protection efficiency in complex environments, allowing for timely response to entanglement events.

[0097] Some embodiments of the present invention also provide an underwater robot, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method described above.

[0098] Some embodiments of the present invention also provide a computer-readable storage medium on which a computer program is stored, and which, when executed by a processor, implements the method described above.

[0099] Some embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the method described above.

[0100] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0101] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this invention are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0102] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0103] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0105] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0107] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0108] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the aforementioned element.

[0109] The control method, device, underwater robot, and medium for an underwater robot have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A control method for an underwater robot, characterized in that, The method includes: Acquire multi-source data parameters, and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein the cable is used to connect the underwater robot and the control console on the water surface; Based on the coupling model, the cable trajectory in the future time period is predicted, and the probability of cable entanglement in the future preset time period is calculated. When the probability of cable entanglement exceeds a preset medium-risk threshold, a local map for the current environment is established based on the multi-source data parameters. Based on the cable entanglement probability and the local map, an entanglement risk map is established; wherein, the entanglement risk map is a map with an entanglement probability distribution; In the entanglement risk map, a target path is searched and a target path is generated; The underwater robot is controlled to move according to the target path.

2. The method according to claim 1, characterized in that, Based on the multi-source data parameters, a coupling model of cables, obstacles, and water flow is established, including: Based on the multi-source data parameters, the coupling features of the cable, obstacle, and water flow are extracted; wherein, the coupling features are the coupling relationship between the cable, obstacle, and water flow in the spatiotemporal dimension; Based on the aforementioned coupling characteristics, a coupling model of cables, obstacles, and water flow is established.

3. The method according to claim 1, characterized in that, In the entanglement risk map, a target path search is performed to generate a target path, including: A multi-constraint objective function is constructed with minimizing the probability of cable entanglement as the optimization objective. In the entanglement risk map, multiple candidate nodes are identified, and a target path is generated based on the candidate nodes and the multi-constraint objective function.

4. The method according to claim 3, characterized in that, In the entanglement risk map, multiple candidate nodes are identified, and a target path is generated based on the candidate nodes and the multi-constraint objective function, including: In the entanglement risk map, low-entanglement risk areas and high-entanglement risk areas are identified, and multiple candidate nodes are generated in the low-entanglement risk areas and the high-entanglement risk areas. Calculate the change in entanglement probability for each candidate node; Based on the change in entanglement probability and the multi-constraint objective function, the target node is determined from multiple candidate nodes; Generate the target path based on the target node.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: When the probability of cable entanglement exceeds a preset low-risk threshold, a preset auxiliary thruster is activated. The underwater robot's attitude is adjusted by the auxiliary thruster, and the mechanical state and length of the cable are also adjusted.

6. The method according to any one of claims 1-4, characterized in that, The cable includes a core cable core, a load-bearing cable segment, and auxiliary functional cable cores; the method further includes: When the probability of cable entanglement exceeds a preset high-risk threshold, or when the probability of entanglement exceeds a preset safety threshold during a preset time period when the underwater robot moves along the target path, a first response is executed; wherein, the first response is to cut off a portion of the load-bearing cable segment through a controllable release device.

7. The method according to claim 6, characterized in that, The method further includes: When the probability of cable entanglement exceeds a preset extremely high risk threshold, or when the probability of entanglement exceeds a safety threshold within a preset time period after the execution of the first response, a second response is executed; wherein, the second response is to cut off all non-core cable cores through a controllable release device and control the underwater robot to float up through a buoyancy airbag.

8. A control device for an underwater robot, characterized in that, The device includes: The coupling model establishment module is used to acquire multi-source data parameters and establish a coupling model of cable, obstacle and water flow based on the multi-source data parameters; wherein, the cable is used to connect the underwater robot and the control console on the water surface; The cable entanglement probability calculation module is used to predict the cable movement trajectory in a future time period based on the coupling model, and to calculate the cable entanglement probability in a future preset time period. The map building module, when the probability of cable entanglement exceeds a preset medium-risk threshold, builds a local map for the current environment based on the multi-source data parameters. The entanglement risk map creation module is used to create an entanglement risk map based on the cable entanglement probability and the local map; wherein, the entanglement risk map is a map with an entanglement probability distribution; The target path generation module is used to search for target paths and generate target paths in the entanglement risk map. An underwater robot motion module is used to control the movement of the underwater robot according to the target path.

9. An underwater robot, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images