Multi-field coupling robot self-adaptive inspection method and system based on creative technology stack

Abstract

The invention discloses a multi-field coupling robot self-adaptive inspection method and system based on a creative technology stack, and the method comprises the steps: reading a fixed historical parameter set in a previous inspection period, collecting the environment data of a current operation region, generating a control parameter set, calling thermal control and propulsion parameters in the control parameter set, and outputting real-time flow field data and attitude control quantity. Based on real-time flow field data and attitude control quantity, navigation speed and propulsive power are optimized, energy consumption data and path tracking data are synchronously recorded, a real-time signal-to-noise ratio is calculated by using environment real-time parameters, communication states are adjusted by combining communication bandwidth parameters in a control parameter set, and the environment real-time parameters and communication state data are synchronously received. Summarizing the data to complete data archiving; through heat-flow coupling control and multi-mode sensing fusion and environment adaptive communication and parameter iterative optimization cooperation, stable inspection in a complex flow field, zero-visibility defect identification and reliable return of noisy environment key data are realized.

Description

Technical Field

[0001] This invention relates to the field of adaptive inspection technology for robots, specifically a method and system for adaptive inspection of multi-field coupled robots based on the domestic information technology stack. Background Technology

[0002] With the increasing urgency of operation and maintenance needs for deep-water engineering structures such as hydropower dams, offshore oil and gas platforms, and underwater tunnels, underwater inspection technology is gradually evolving from traditional manual diving operations to intelligent robotic inspection. Early underwater inspections relied on manned submersibles, which were limited by environmental factors such as water depth, current field, and visibility, resulting in high operational risks, high costs, and low efficiency. Subsequently, remotely operated underwater vehicles (ROVs) gradually became more widespread, but they mostly used umbilical cable control, which limited their maneuverability and made it difficult to adapt to complex environments such as strong currents and high turbidity. In recent years, autonomous underwater robots have become mainstream, achieving wireless autonomous operation. However, their core technologies largely rely on foreign software and hardware systems, and they have shortcomings in areas such as multi-physics collaborative control and environmental adaptability. While the maturity of domestically developed information technology has provided technical support for the development of intelligent inspection equipment for critical infrastructure that is independently controllable, existing technologies still have the following deficiencies: 1. Existing underwater robots have multiple independent subsystems for thermal control, propulsion, etc., without a collaborative mechanism. When operating at high speeds against the flow field, the strong convection cooling effect is not utilized, resulting in redundant heat dissipation in the electronic compartment. The propulsion system lacks adaptive flow field compensation, has weak anti-turbulence and anti-adsorption capabilities, and is prone to deviating from the inspection path. When hovering at low speeds for inspection, insufficient natural heat dissipation leads to overheating and shutdown of electronic equipment. Overall energy efficiency and operational stability are difficult to balance, making it unsuitable for the refined inspection requirements under complex flow fields.

[0003] 2. Lack of environmental adaptability and self-evolution capabilities: Underwater acoustic communication does not incorporate real-time environmental parameters, such as noise, reverberation, and temperature, salinity, and depth data, to dynamically adjust transmission strategies, leading to frequent interruptions of critical inspection data in complex environments. Simultaneously, safety monitoring relies on a single sensor, which can easily misinterpret environmental condensation and interference signals as equipment leaks or structural defects, posing a risk of false alarms or missed detections. Summary of the Invention This invention provides a multi-field coupled robot adaptive inspection method and system based on the domestic information technology stack. By integrating thermal-fluid coupling control with multimodal perception, and coordinating environmental adaptive communication with parameter iterative optimization, it achieves stable inspection under complex flow fields, zero-visibility defect identification, and reliable data transmission in noisy environments, thereby solving the problems in the background technology.

[0004] To achieve the above objectives, the technical solution of the present invention is as follows: An adaptive inspection method for multi-field coupled robots based on the domestic information technology stack, comprising the following steps executed via computer equipment. S1, parameter initialization: read the historical parameter set solidified in the previous inspection cycle, collect basic environmental data of the current work area, and generate a set of robot control parameters including thermal control, propulsion, and communication thresholds; S2, inspection deployment adaptation, calls the thermal control and propulsion parameters in the control parameter set, collects flow field data in the inspection scenario in real time, dynamically adjusts attitude control quantity and thermal control execution parameters, and outputs real-time flow field data and attitude control quantity; S3, Inspection and Energy Efficiency Optimization, optimizes sailing speed and propulsion power based on real-time flow field data and attitude control parameters. It collects structural feature data and real-time environmental parameters through multiple sensors, and simultaneously records energy consumption data and path tracking data.

[0005] S4, environment adaptive communication, calculates the real-time signal-to-noise ratio using real-time environmental parameters, and adjusts the communication code rate, transmission mode and operating frequency band by combining the communication bandwidth parameters in the control parameter set, transmitting the structural characteristic data of S3, and outputting communication status data; S5, safety monitoring and response, synchronously receives real-time environmental parameters and communication status data, collects response values ​​from chemical and acoustic sensors, and outputs safety status judgment results after cross-verification.

[0006] S6, Data Archiving: Summarize the energy consumption data, path tracking data and safety status judgment results in S3, calculate the parameter optimization increment, update and generate a new set of control parameters and store it in a fixed manner, return to the recycling point based on the path parameters in the control parameter set, export the structural feature data in S3, and complete the data archiving.

[0007] Optionally, the specific implementation steps of S1 are as follows: S11, Historical parameter retrieval, reads the historical control parameter set stored in the previous inspection cycle for this inspection task. Extract the environmental adaptation baseline parameters, control strategy thresholds, and performance optimization records; where k represents the current inspection cycle. Represents the set of control parameters; S12, Current environmental basic data collection: The robot acquires core environmental feature data of the work area through environmental perception sensors to form an initial environmental dataset. S13, Parameter Fusion and Adaptation: This step involves collaboratively analyzing the historical parameters extracted in S11 with the current environmental dataset collected in S12. It adaptively adjusts the dimensions and thresholds of the control parameters to generate a current control parameter set that includes thermal control target parameters, propulsion reference parameters, and communication reference parameters. .

[0008] Optionally, the specific implementation steps of S2 are as follows: S21, core parameter call, extract the current control parameter set. The thermal control target parameters and propulsion reference parameters are used as the initial control basis for the deployment phase; S22, flow field data acquisition: Through the environmental perception sensors on the robot, key flow field characteristic data such as flow velocity and water temperature in the working scene are acquired in real time to form a real-time flow field dataset; S23, multi-field coordinated adjustment, based on the initial control parameters of S21 and the real-time flow field dataset of S22, dynamically adjusts the attitude stabilization control quantity through thermal-fluid coupling control logic, and simultaneously optimizes the thermal control execution parameters; S24, Adaptation parameter output, integrates optimized real-time flow field data, attitude control quantities and thermal control execution parameters to form a deployment adaptation result dataset.

[0009] Optionally, the specific implementation steps of S3 are as follows: S31, Input data reception, obtain the deployment adaptation result dataset output by S2, and extract the real-time flow field data, optimized attitude control quantity and thermal control execution parameters from it; S32, navigation and propulsion parameter optimization, based on engineering inspection targets, combined with real-time flow field data and attitude control variables from S31, dynamically adjusts the robot's navigation speed and propulsion power distribution strategy. S33, multi-dimensional data acquisition, uses environmental sensors on the robot to simultaneously collect structural feature data of the object being detected and real-time parameters of the working environment, including noise and reverberation; S34, Key Data Recording, captures and records energy consumption data and path tracking deviation data in real time during robot operation; S35, intermediate data output, integrates the structural feature data and real-time environmental parameters collected by S33 with the energy consumption data and path tracking deviation data recorded by S34 to form an intermediate dataset during the inspection process.

[0010] Optionally, the specific implementation steps of S4 are as follows: S41, Input data reception, acquire real-time environmental parameters including noise and reverberation characteristics from the intermediate dataset output by S3 during the inspection process, and extract the current control parameter set. The communication baseline parameters are used to construct a basic dataset for communication adaptation.

[0011] S42, Real-time signal-to-noise ratio calculation: Based on the channel capacity-related logic in the document, combined with the noise and reverberation characteristic data obtained in S41, the real-time signal-to-noise ratio under the current operating environment is inverted. S43, communication parameters are dynamically adjusted based on real-time signal-to-noise ratio and... The communication baseline parameters are used to adaptively optimize the communication code rate allocation, transmission waveform, and operating frequency band. S44, data transmission, based on the adjusted communication reference parameters, directional transmission of key structural feature data collected by S3; S45, communication status feedback output, monitors link stability and data transmission success rate in real time, and forms a communication status dataset.

[0012] Optionally, the specific implementation steps of S5 are as follows: S51, Data reception, synchronously acquires real-time environmental parameters from the intermediate dataset of the inspection process output by S3, and communication status dataset output by S4, to build a basic dataset for safety monitoring. S52, data acquisition, synchronously collects chemical sensor response values, including humidity and ion concentration, and acoustic sensor response values, including abnormal impact frequency and intensity, through the robot, to supplement the data for safety judgment dimensions; S53, multi-dimensional data cross-validation, based on dual-modal hysteresis decision logic, collaboratively analyzes the basic environmental and communication data of S51 with the dual-modal sensor data of S52 to identify real security risks.

[0013] S54, Security Status Determination Output: Based on the cross-validation results, outputs a clear security status logical value, forming a security determination result file.

[0014] S55 Emergency Response Execution: If the judgment result is abnormal, trigger the corresponding emergency action according to the risk type; if the judgment result is safe, continue to maintain the current inspection operation status.

[0015] Optionally, the specific implementation steps of S6 are as follows: S61, receive associated data, summarize energy consumption data, path tracking error data and safety status judgment results, and construct the basic dataset for data archiving.

[0016] S62, Parameter optimization calculation, based on the archived basic dataset, solves for the parameter optimization increment according to the preset parameter optimization logic. Update and generate a new generation of control parameter sets. ; S63, new parameters are permanently stored, including the optimized control parameter set. Write the data to the robot's non-volatile storage module to achieve persistent storage of parameters; S64, return path execution, call Based on the path planning parameters, and according to the preset recycling point coordinates, the robot is driven to perform a return motion and arrive at the designated recycling area; S65, core data export, extracts the full structural feature data collected by S3 from the storage module and exports it to the shore station system via wired or wireless transmission. S66, full data archiving, integrating and exporting structural feature data; The parameter set and process data records of this operation are classified and archived according to the preset format to form a complete operation data archive.

[0017] An adaptive inspection system for a multi-field coupled robot based on the domestic IT innovation technology stack includes an electronic cabin assembly module, a scheduling module, a memory, and a processor; The memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of any of the methods described above.

[0018] Optionally, the electronic cabin assembly module may internally house a computing unit, a thermal control unit, an attitude control unit, a sensor unit, a communication unit, and a scheduling unit. The electronic cabin assembly module includes a lightweight pressure-resistant shell, streamlined airflow channels and heat dissipation fins, and phase change thermally conductive filling material, providing a physical installation and protective carrier for each unit; The scheduling module includes a task scheduling unit, a display unit, and a data archiving unit. It is used to send task requirement parameters and control commands to the computing unit, receive inspection data and equipment status information uploaded by the vehicle-mounted computing module, display the 3D model of the work area and defect annotations, and complete data archiving.

[0019] Optionally, the computing unit is installed inside the electronic cabin assembly module, including a domestic edge computing chip, a real-time Linux / RTOS, a defect identification algorithm, and a parameter storage unit. It is used to issue task requirement parameters and control commands to other modules. The control commands include thermal control commands and attitude control quality. The computing parameters are optimized incrementally and written back to storage, and data is uploaded to the shore station module. The thermal control unit includes a distributed thermistor array and a cooling array, used to collect component temperature data of the computing unit, receive thermal control commands issued by the computing module, adjust the cabin temperature, and feed back the temperature adjustment results to the computing module. The attitude control unit includes a multi-vector brushless DC thruster and a thruster driver, which is used to receive attitude control commands issued by the computing unit, drive the thruster to achieve six degrees of freedom maneuver, collect thruster operating parameters and feed them back to the computing module, and adjust the propulsion strategy according to changes in the flow field. The sensor unit includes a high-resolution optical camera, a multi-beam imaging sonar, a CTD sensor, a chemical and acoustic sensor, and an extended operating interface; it is used to collect data on the structural features, flow field, water temperature, and noise environment of the operating area, and transmit the collected data to the computing module after preprocessing. The communication unit includes a short-baseline positioning device, a communication unit, a multi-beam sonar, and a visual odometer. It receives the task requirements from the computing unit, completes the positioning of the work area, adjusts the communication parameters to adapt to environmental interference, and feeds back the positioning and communication status to the computing unit.

[0020] As can be seen from the above technical solution compared with the prior art, the present invention has the following beneficial effects: 1. This invention integrates thermal-fluid coupling control logic with multimodal sensing technology to adjust attitude control quantity and thermal control parameters in a coordinated manner. At the same time, it relies on high-resolution three-dimensional imaging sonar to achieve defect identification under zero visibility conditions, realize robot attitude stability and refined inspection coordination under complex flow fields, and improve the continuity of inspection and the accuracy of defect identification.

[0021] 2. This invention dynamically adjusts the underwater acoustic communication waveform, bit rate, and frequency band through an environment-adaptive communication strategy. Combined with a full parameter write-back optimization mechanism, it iteratively controls the parameters to achieve reliable backhaul of key defect data in noisy environments such as shallow water with strong reverberation and autonomous evolution of the robot's environmental adaptability. This results in continuous optimization of communication stability and operational efficiency. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the method steps of the present invention; Figure 2 This is a schematic diagram of the connection process of the system modules and units of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0024] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention, but should not be used to limit the scope of the present invention.

[0025] Example 1: like Figure 1 As shown in this embodiment, at the operation and maintenance site of a hydropower dam project, the water depth in the work area can reach 300 meters, and there are complex environments such as strong current shear, water turbidity, GPS signal loss, and unit noise interference, while facing multiple core problems. Traditional inspection equipment, due to its heat-fluid decoupling design, is susceptible to environmental interference with its single sensor, which may misjudge condensation as leakage or miss tiny cracks. Furthermore, the parameters for each operation are fixed, making it impossible to accumulate past data to optimize subsequent tasks.

[0026] To address the challenges in the aforementioned scenarios, this invention designs a multi-field coupled robot adaptive inspection system based on the domestic information technology stack. Through core innovations such as thermal-fluid coordinated control, environmental adaptive communication, dual-modal health monitoring, and full lifecycle parameter write-back, it achieves efficient and reliable inspection operations in complex underwater environments. The following detailed description is based on specific embodiments.

[0027] This embodiment describes the system architecture of the present invention in detail, such as... Figure 2 As shown, this system forms the physical basis for realizing the aforementioned mathematical model and control logic. It is specifically designed to adapt to the complex water flow, high turbidity, and mixed metal-concrete environment within hydropower station dams. The system includes: The electronic compartment assembly module uses a lightweight, high-strength hard anodized aluminum alloy or carbon fiber composite material pressure-resistant shell, designed to withstand a pressure depth of 300 meters, and is filled with phase change thermal conductive material or air depending on heat dissipation requirements.

[0028] Meanwhile, the shell is designed with streamlined flow channels and heat dissipation fins to adapt to water flow. For high flow velocity areas such as dam inlets, its geometric parameters have been optimized by CFD to maximize convective heat transfer efficiency and use ambient water flow to carry away the heat generated by the high-power computing unit.

[0029] The thermal control unit includes a distributed high-precision thermistor array, which focuses on monitoring the high-performance edge computing module for crack detection and high-efficiency battery pack temperature. It is also equipped with an intelligent air-cooling or micro liquid-cooling circulation system, which is controlled by the computing unit.

[0030] During low-speed, fine-tuning inspections (such as hovering to photograph cracks), if natural convection is insufficient, the internal circulation rate will be automatically increased to ensure that the equipment does not overheat.

[0031] The attitude control unit, equipped with high-thrust brushless DC thrusters with a multi-vector layout, possesses strong resistance to interference from strong currents and is responsible for six-degree-of-freedom maneuvers. It is particularly enhanced for anti-adsorption and anti-turbulence control when navigating close to the dam surface. Real-time feedback of current and back electromotive force is used to estimate near-wall fluid resistance and assist in judging changes in the flow field.

[0032] The communication unit is equipped with a high-frequency short baseline positioning and high-bandwidth underwater acoustic communication device, which is adapted to shallow water and strong reverberation environment. Combined with multi-beam sonar and visual odometry (VIO), it can achieve high-precision positioning at the bottom of the dam where GPS signals are missing, and adjust the communication frequency in real time to avoid the noise frequency band of the power plant unit.

[0033] The sensor unit integrates environmental perception and expansion interfaces, and is equipped with a high-resolution optical camera, a turbidity removal algorithm, and two-dimensional / three-dimensional multibeam imaging sonar. It is used to construct a three-dimensional model of the dam surface and detect cracks and erosion in turbid water. Environmental perception: CTD sensor monitors water stratification in the reservoir area; and it is equipped with dissolved oxygen and turbidity sensors to assist in ecological monitoring.

[0034] Expansion Interface: A robotic arm interface is reserved to support the installation of cavitation noise monitors, trash rack cleaning tools, and concrete non-destructive testing probes, meeting the core needs of hydropower operation and maintenance.

[0035] The computing unit runs a real-time Linux / RTOS, deploys a deep learning-based dam defect identification algorithm, analyzes video streams in real time, uploads only key frames of suspected defects to reduce communication bandwidth requirements, and records path planning deviations and environmental flow field data for each inspection to optimize the approach control strategy for the next inspection of the same area.

[0036] The scheduling unit is used for human-computer interaction, displays the dam's 3D point cloud and defect annotations in real time, and supports seamless integration with the power plant operation and maintenance system.

[0037] By relying on domestic IT innovation technology support, integrated protection and thermal-fluid coupling structure design, multi-module collaborative scheduling and multi-sensor fusion perception, the system's autonomous controllability, extreme environment adaptability, inspection coverage and identification accuracy are improved, while ensuring data transmission stability and convenient operation and maintenance decision-making.

[0038] Example 2: like Figure 1 As shown, after the system configuration is completed, the dam inspection method based on device-level collaborative closed loop is described in detail in this embodiment. The collaborative closed loop workflow based on the device in Embodiment 1 is used to achieve efficient operation in the hydropower dam scenario through full-process optimization, and the core mathematical model is decomposed into each step.

[0039] S1: Parameter loading and environment adaptation, read the flow field characteristics and noise model of a specific area of ​​the dam recorded in the airborne black box from the previous dive; after confirmation by the shore station, load specific control parameters for the mission area, such as the intake, spillway or dam foundation.

[0040] Mathematical description:

[0041] DamZoneParams represents the preset parameters for the dam area. This indicates an adaptive adjustment of the increment. This represents the current set of control parameters for the inspection task; S2: Descent and Flow Field. As the robot descends along the dam surface, it may encounter water shear at different depths or suction flow caused by power station units.

[0042] Algorithm: Adaptive flow field compensation control is implemented, with the core being the real-time solution of the thermal-fluid coupling equilibrium point.

[0043] in, Indicates the heating power with thermal compensation. Indicates the current sailing speed. This indicates the target constant temperature value inside the cabin. The external ambient water temperature is represented by v, the speed by v, and the effective heat dissipation area of ​​the avionics compartment shell by A. This indicates the heat generated by the electronic equipment inside the electronics compartment.

[0044] The flow velocity vector is estimated in real time by using load changes fed back from the thrusters, and the attitude is dynamically adjusted to maintain a constant distance from the dam surface. At the same time, the high heat transfer coefficient brought by the strong water flow is utilized. Optimize the heat dissipation strategy for the electronic compartment.

[0045] S3: Close-fitting fine-tuning inspection and energy efficiency optimization, steady-state operation. For example, this inspection task is to conduct a "carpet-like" scan of the dam's concrete surface to find tiny cracks or exposed rebar.

[0046] While ensuring image / sonar imaging clarity and low-speed stability, we need to find a balance between energy consumption and heat dissipation, which means solving the following local optimization problem:

[0047] in, This represents the optimization weight coefficient. Indicates the platform's base power consumption. This represents the propulsion power function. Strategy: When high-power supplementary lighting is required in turbid waters, utilize the accompanying turbulence for natural cooling; in still waters, appropriately reduce the operating frequency to minimize heat generation.

[0048] S4: Environmentally driven communication anti-interference. If a suspected crack or trash rack blockage is detected, a high-resolution image or sonar map must be uploaded to monitor the reverberation level of the underwater acoustic channel and the noise level of the power plant unit in real time, and calculate the real-time signal-to-noise ratio. .

[0049] The current channel capacity limit is determined by inverting the Shannon formula, and an adaptive code rate R is set:

[0050] in, This represents the real-time communication bit rate, where t represents the current real-time time. The maximum physical bandwidth of the system is represented by SNR(t), and the time-varying signal-to-noise ratio is represented by SNR(t). denoted by the channel utilization efficiency coefficient, the correction coefficient of Shannon's formula, and the loss adapted to the actual communication scenario. B represents the communication channel bandwidth. In strong multipath shallow water / near-wall environments, it automatically switches to anti-multipath spread spectrum communication mode. When noise interference is strong, it avoids interference frequency bands by frequency hopping, reliably transmits key defect data, and ensures zero missed detections.

[0051] S5: Dual-modal safety decision, continuously monitoring internal humidity and pressure while listening for abnormal external sounds; Criteria: Combining visual obstacle avoidance and sonar ranging, the following safety logic gates are executed:

[0052] in, Represents the logical value of the system's security status. Indicates the humidity alarm threshold; This indicates the impact / acoustic anomaly threshold. This indicates the response value of the chemical sensor, specifically referring to the sensor output value for monitoring the humidity inside the cabin. This indicates the acoustic sensor response value, specifically the output value of the sensor monitoring external impacts / abnormal sounds. When too close to an obstacle, such as a protruding rebar or tree trunk, and with abnormal flow velocity, an emergency avoidance mechanism is triggered. Response: If water ingress into the cabin is detected, indicating a risk of high-pressure leakage deep within the dam, the buoyancy adjustment device is immediately inflated to achieve rapid ascent.

[0053] S6: Full parameter write-back and path memory; calculation: Before the task ends, evaluate the actual fit and energy efficiency of the current path planning, and calculate the performance functional. :

[0054] Where T represents the total duration of this inspection task recorded by the onboard computing module, which is automatically timed from task start to end. This represents the functional of the post-task evaluation cost. Represents the total instantaneous power. Indicates path tracking error. The evaluation weights are represented and stored in the computation unit, and can be configured through the shore station module; the flow field map and obstacle avoidance threshold are corrected using the gradient descent method.

[0055] in, This represents the updated parameter set. This represents the parameter set before the update; Indicates the learning rate. This represents the gradient operator with respect to the parameters. The parameters are calculated by the computing unit based on the mapping relationship between the cost functional Jactual and the parameter set, and the corrected parameters are written into the storage area, so that the robot becomes more and more familiar with the specific structure of the dam as it is inspected.

[0056] Step S7: Return and Data Archiving. The robot floats to the recovery platform and exports all inspection data via wired / wireless connection. It automatically generates a dam health status report and analyzes the crack expansion trend by comparing it with historical data.

[0057] By adapting historical parameters to current environmental data, dynamically adjusting thermal-fluid coupling, optimizing the balance between energy efficiency and precision, adaptively adjusting communication parameters, cross-validating multi-dimensional data, and iteratively updating parameters, the system achieves precise adaptation to operational scenarios, stable equipment operation under complex flow fields, comprehensive and efficient defect identification, reliable transmission of key data, effective prevention and control of safety risks, and autonomous evolution of subsequent inspection capabilities.

[0058] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0059] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0060] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the adaptive inspection methods for multi-field coupled robots based on the information technology stack in the above embodiments.

[0061] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0062] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0063] For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media.

[0064] The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0065] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0066] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0067] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0068] The embodiments of the present invention are given for the purposes of illustration and description. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. An adaptive inspection method for multi-field coupled robots based on domestic information technology stack, characterized in that, Perform the following steps using a computer device: S1, parameter initialization: read the historical parameter set solidified in the previous inspection cycle, collect basic environmental data of the current work area, and generate a set of robot control parameters including thermal control, propulsion, and communication thresholds; S2, inspection deployment adaptation, calls the thermal control and propulsion parameters in the control parameter set, collects flow field data in the inspection scenario in real time, dynamically adjusts attitude control quantity and thermal control execution parameters, and outputs real-time flow field data and attitude control quantity; S3, Inspection and Energy Efficiency Optimization, optimizes sailing speed and propulsion power based on real-time flow field data and attitude control variables. It collects structural feature data and real-time environmental parameters through multiple sensors, and simultaneously records energy consumption data and path tracking data. S4, environment adaptive communication, calculates the real-time signal-to-noise ratio using real-time environmental parameters, and adjusts the communication code rate, transmission mode and operating frequency band by combining the communication bandwidth parameters in the control parameter set, transmitting the structural characteristic data of S3, and outputting communication status data; S5, safety monitoring and response, synchronously receives real-time environmental parameters and communication status data, collects chemical sensor response values ​​and acoustic sensor response values, and outputs safety status judgment results after cross-verification; S6, Data Archiving: Summarize the energy consumption data, path tracking data and safety status judgment results in S3, calculate the parameter optimization increment, update and generate a new set of control parameters and store it in a fixed manner, return to the recycling point based on the path parameters in the control parameter set, export the structural feature data in S3, and complete the data archiving.

2. The adaptive inspection method for multi-field coupled robots based on the information technology stack as described in claim 1, characterized in that: The specific implementation steps of S1 are as follows: S11, Historical parameter retrieval, reads the historical control parameter set stored in the previous inspection cycle for this inspection task. Extract the environmental adaptation baseline parameters, control strategy thresholds, and performance optimization records; where k represents the current inspection cycle. Represents the set of control parameters; S12, Current environmental basic data collection: The robot acquires core environmental feature data of the work area through environmental perception sensors to form an initial environmental dataset. S13, Parameter Fusion and Adaptation: This step involves collaboratively analyzing the historical parameters extracted in S11 with the current environmental dataset collected in S12. It adaptively adjusts the dimensions and thresholds of the control parameters to generate a current control parameter set that includes thermal control target parameters, propulsion reference parameters, and communication reference parameters. .

3. The adaptive inspection method for multi-field coupled robots based on the information technology stack as described in claim 2, characterized in that: The specific implementation steps of S2 are as follows: S21, core parameter call, extract the current control parameter set. The thermal control target parameters and propulsion reference parameters are used as the initial control basis for the deployment phase; S22, flow field data acquisition: Through the environmental perception sensors on the robot, key flow field characteristic data such as flow velocity and water temperature in the working scene are acquired in real time to form a real-time flow field dataset; S23, multi-field coordinated adjustment, based on the initial control parameters of S21 and the real-time flow field dataset of S22, dynamically adjusts the attitude stabilization control quantity through thermal-fluid coupling control logic, and simultaneously optimizes the thermal control execution parameters; S24, Adaptation parameter output, integrates optimized real-time flow field data, attitude control quantities and thermal control execution parameters to form a deployment adaptation result dataset.

4. The adaptive inspection method for multi-field coupled robots based on the domestic information technology stack as described in claim 3, characterized in that: The specific implementation steps of S3 are as follows: S31, Input data reception, obtain the deployment adaptation result dataset output by S2, and extract the real-time flow field data, optimized attitude control quantity and thermal control execution parameters from it; S32, navigation and propulsion parameter optimization, based on engineering inspection targets, combined with real-time flow field data and attitude control variables from S31, dynamically adjusts the robot's navigation speed and propulsion power distribution strategy. S33, multi-dimensional data acquisition, uses environmental sensors on the robot to simultaneously collect structural feature data of the object being detected and real-time parameters of the working environment, including noise and reverberation; S34, Key Data Recording, captures and records energy consumption data and path tracking deviation data in real time during robot operation; S35, intermediate data output, integrates the structural feature data and real-time environmental parameters collected by S33 with the energy consumption data and path tracking deviation data recorded by S34 to form an intermediate dataset during the inspection process.

5. The adaptive inspection method for multi-field coupled robots based on the domestic information technology stack as described in claim 4, characterized in that: The specific implementation steps of S4 are as follows: S41, Input data reception, acquire real-time environmental parameters including noise and reverberation characteristics from the intermediate dataset output by S3 during the inspection process, and extract the current control parameter set. Based on the communication baseline parameters, construct a basic dataset for communication adaptation; S42, Real-time signal-to-noise ratio calculation: Based on the channel capacity-related logic in the document, combined with the noise and reverberation characteristic data obtained in S41, the real-time signal-to-noise ratio under the current operating environment is inverted. S43, communication parameters are dynamically adjusted based on real-time signal-to-noise ratio and... The communication baseline parameters are used to adaptively optimize the communication code rate allocation, transmission waveform, and operating frequency band. S44, data transmission, based on the adjusted communication reference parameters, directional transmission of key structural feature data collected by S3; S45, communication status feedback output, monitors link stability and data transmission success rate in real time, and forms a communication status dataset.

6. The adaptive inspection method for multi-field coupled robots based on the domestic information technology stack as described in claim 5, characterized in that: The specific implementation steps of S5 are as follows: S51, Data reception, synchronously acquires real-time environmental parameters from the intermediate dataset of the inspection process output by S3, and communication status dataset output by S4, to build a basic dataset for safety monitoring. S52, data acquisition, synchronously collects chemical sensor response values, including humidity and ion concentration, and acoustic sensor response values, including abnormal impact frequency and intensity, through the robot, to supplement the data for safety judgment dimensions; S53, multi-dimensional data cross-validation, based on dual-modal hysteresis decision logic, collaboratively analyzes the basic environment and communication data of S51 with the dual-modal sensor data of S52 to identify real security risks; S54, Security Status Determination Output: Based on the cross-validation results, outputs a clear security status logical value, forming a security determination result file; S55 Emergency Response Execution: If the judgment result is abnormal, trigger the corresponding emergency action according to the risk type; if the judgment result is safe, continue to maintain the current inspection operation status.

7. The adaptive inspection method for multi-field coupled robots based on the domestic information technology stack as described in claim 6, characterized in that: The specific implementation steps of S6 are as follows: S61, Related data reception, summarizing energy consumption data, path tracking error data and safety status judgment results, and constructing a basic dataset for data archiving; S62, Parameter optimization calculation, based on the archived basic dataset, solves for the parameter optimization increment according to the preset parameter optimization logic. Update and generate a new generation of control parameter sets. ; S63, new parameters are permanently stored, including the optimized control parameter set. Write the data to the robot's non-volatile storage module to achieve persistent storage of parameters; S64, return path execution, call Based on the path planning parameters, and according to the preset recycling point coordinates, the robot is driven to perform a return motion and arrive at the designated recycling area; S65, core data export, extracts the full structural feature data collected by S3 from the storage module and exports it to the shore station system via wired or wireless transmission. S66, full data archiving, integrating and exporting structural feature data; The parameter set and process data records of this operation are classified and archived according to the preset format to form a complete operation data archive.

8. An adaptive inspection system for a multi-field coupled robot based on an information technology stack, comprising an electronic cabin assembly module, a scheduling module, a memory, and a processor, wherein the memory stores a computer program, characterized in that, When the computer program is executed by the processor, it causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

9. The multi-field coupled robot adaptive inspection system based on the domestic information technology stack as described in claim 8, characterized in that: The electronic cabin assembly module internally includes a computing unit, a thermal control unit, an attitude control unit, a sensor unit, a communication unit, and a scheduling unit; The electronic cabin assembly module includes a lightweight pressure-resistant shell, streamlined airflow channels and heat dissipation fins, and phase change thermally conductive filling material, providing a physical installation and protective carrier for each unit; The scheduling module includes a task scheduling unit, a display unit, and a data archiving unit. It is used to send task requirement parameters and control commands to the computing unit, receive inspection data and equipment status information uploaded by the vehicle-mounted computing module, display the 3D model of the work area and defect annotations, and complete data archiving.

10. The adaptive inspection method for multi-field coupled robots based on the information technology stack as described in claim 9, characterized in that: The computing unit is installed inside the electronic cabin assembly module and includes a domestically produced edge computing chip, a real-time Linux / RTOS, a defect identification algorithm, and a parameter storage unit. It is used to issue task requirement parameters and control commands to other modules. The control commands include thermal control commands and attitude control quality. It calculates parameter optimization increments and writes them back to storage, and uploads data to the shore station module. The thermal control unit includes a distributed thermistor array and a cooling array, used to collect component temperature data of the computing unit, receive thermal control commands issued by the computing module, adjust the cabin temperature, and feed back the temperature adjustment results to the computing module. The attitude control unit includes a multi-vector brushless DC thruster and a thruster driver, which is used to receive attitude control commands issued by the computing unit, drive the thruster to achieve six degrees of freedom maneuver, collect thruster operating parameters and feed them back to the computing module, and adjust the propulsion strategy according to changes in the flow field. The sensor unit includes a high-resolution optical camera, a multi-beam imaging sonar, a CTD sensor, chemical and acoustic sensors, and an extended operating interface. It is used to collect data on the structural features, flow field, water temperature and noise environment of the work area, and transmits the collected data to the computing module after preprocessing. The communication unit includes a short-baseline positioning device, a communication unit, a multi-beam sonar, and a visual odometer. It receives the task requirements from the computing unit, completes the positioning of the work area, adjusts the communication parameters to adapt to environmental interference, and feeds back the positioning and communication status to the computing unit.

Images