Robot operation control method and system and robot

Abstract

The invention relates to the technical field of machine control, in particular to a robot operation control method and system and a robot. The method comprises the following steps: acquiring spatial data of a breeding work ship; performing multi-source sensor network deployment according to the spatial data of the cultivation work ship to generate a multi-source sensor network; performing sensor data integration on the multi-source sensor network to generate cultivation work ship environment sensing integrated data; performing real-time environment updating on the cultivation work ship environment sensing integrated data to generate cultivation work ship real-time environment monitoring data; performing spatio-temporal data synchronization on the real-time environment monitoring data of the breeding work ship to generate environment synchronization data of the breeding work ship; and according to the cultivation work ship environment synchronization data, environment model construction is carried out to generate a comprehensive cultivation environment model. Through real-time environment sensing, dynamic path planning, intelligent feeding control and anomaly detection, the stability and the intelligent level of robot operation control are improved.

Description

Technical Field

[0001] The present invention relates to the field of machine control technology, and in particular to a robot operation control method, system and robot. Background Art

[0002] Early aquaculture vessels mainly relied on manual loading and processing. With the development of marine aquaculture, traditional manual operations have become increasingly inefficient, and it is difficult to ensure the consistency and safety of processing. The main technical bottleneck at this stage is the lack of automated equipment, and the high labor intensity and poor working environment caused by manual operation need to be solved urgently. Entering the 21st century, with the development of artificial intelligence and machine learning technologies, the robot systems on aquaculture vessels have begun to have a higher level of intelligence. Advanced sensors and image recognition technologies are applied to robot systems, enabling them to perform loading and processing tasks more accurately. At the same time, the system can monitor environmental changes in real time and make dynamic adjustments, further improving the safety and stability of production. In recent years, with the integration of the Internet of Things (IoT) and big data technologies, the ship-borne processing and loading robots of aquaculture vessels have gradually realized comprehensive intelligent control and remote operation. The robot system can not only complete complex processing tasks autonomously, but also optimize the production process through data analysis, improve resource utilization and economic benefits. However, the current traditional robot control methods often perceive the environment in a static manner and have difficulty adapting to dynamically changing environments. At the same time, robot feeding control often ignores the intelligent analysis of reserve data, which can easily lead to insufficient or excessive feeding, resulting in low stability and intelligence level of robot operation control. Summary of the invention

[0003] Based on this, it is necessary to provide a robot operation control method, system and robot to solve at least one of the above technical problems.

[0004] To achieve the above object, a robot operation control method is provided, the method comprising the following steps:

[0005] Step S1: Acquire spatial data of aquaculture vessels; deploy a multi-source sensor network according to the spatial data of aquaculture vessels to generate a multi-source sensor network; integrate sensor data of the multi-source sensor network to generate environmental perception integrated data of aquaculture vessels; perform real-time environmental updates on the environmental perception integrated data of aquaculture vessels to generate real-time environmental monitoring data of aquaculture vessels;

[0006] Step S2: synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture vessel to generate the environmental synchronization data of the aquaculture vessel; construct an environmental model based on the environmental synchronization data of the aquaculture vessel to generate a comprehensive aquaculture environment model; perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture vessel through the comprehensive aquaculture environment model to generate a robot loading motion path planning diagram;

[0007] Step S3: obtaining the robot feeding reserve data and the robot initial mechanical arm angle control data; dividing the robot feeding reserve data into reserve modes according to a preset feeding reserve threshold value, generating a high-reserve feeding mode and a low-reserve feeding mode; performing a path difficulty analysis on the robot feeding motion path planning diagram, generating feeding path difficulty data; performing multi-modal feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data, generating a robot multi-mode control instruction;

[0008] Step S4: Monitor the robot's feeding and feeding through the robot's multi-mode control instructions to generate robot feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection to execute the robot's normal operation control operation.

[0009] The present invention can comprehensively and accurately capture the environmental information in the ship by acquiring the spatial data of the aquaculture vessel and deploying a multi-source sensor network, ensuring that the basic data for the robot operation is sufficient. The sensor data is integrated and updated in real time, so that the environmental perception information is always kept up to date, and the robot's response speed and accuracy to environmental changes are improved. Real-time monitoring data can help the robot avoid collisions and other safety hazards and enhance the safety of the operation process. Through spatiotemporal data synchronization and environmental model construction, the robot can perform dynamic path planning based on the latest environmental information and adapt to environmental changes. The comprehensive aquaculture environment model can accurately calculate the optimal feeding motion path, improve the efficiency and accuracy of the robot's operation, and reduce the feeding time. The optimized path planning reduces the robot's movement time and energy consumption, and improves the overall operation efficiency. By dividing the feeding reserve data into modes, intelligent control of high reserve and low reserve modes is realized, and the feeding strategy is adjusted according to actual needs to improve resource utilization. The difficulty of the path is analyzed, and the feeding mode can be adjusted according to the complexity of the path to ensure that the robot can successfully complete the task under various environmental conditions. Multimodal feeding operation control allows the robot to select different feeding modes according to actual conditions, which increases the flexibility and adaptability of operation. By real-time transmission and anomaly detection of the monitoring data of the robot's feeding, abnormal situations in the system can be discovered and handled in a timely manner, reducing the occurrence of failures. The ability to detect anomalies and control operations in real time enhances the stability and reliability of the system and reduces the risks caused by operating errors or equipment failures. Through the anomaly detection function of the central control system, accurate normal operation control can be performed to ensure the robot's operating effect and stable operation. Therefore, the present invention improves the stability and intelligence level of the robot's operation control through real-time environmental perception, dynamic path planning, intelligent feeding control and anomaly detection.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Acquire the spatial data of the aquaculture vessel;

[0012] Step S12: deploying a multi-source sensor network according to the spatial data of the aquaculture vessel to generate a multi-source sensor network, wherein the multi-source sensor network deployment includes the deployment of a lidar sensor, an infrared camera sensor, and a temperature and humidity sensor;

[0013] Step S13: collecting sensor data from the multi-source sensor network to obtain a multi-source sensor data packet; performing preliminary data processing on the multi-source sensor data packet to generate a multi-source sensor preliminary processing data packet, wherein the preliminary data processing includes data denoising and data deduplication;

[0014] Step S14: Transmitting the multi-source sensing preliminary processing data packet to the central control system through the 5G network to integrate the multi-source sensing data and generate the aquaculture vessel environmental perception integrated data; estimating the environmental state of the aquaculture vessel environmental perception integrated data and generating environmental state estimation data;

[0015] Step S15: Real-time environmental status monitoring and updating of the environmental perception integrated data of the aquaculture vessel is performed through the environmental status estimation data to generate real-time environmental monitoring data of the aquaculture vessel.

[0016] The present invention collects and obtains the spatial layout and environmental data of the aquaculture workboat. These data include the size of the workboat, the location of each functional area, and other key environmental information, providing basic data for the subsequent deployment of the sensor network to ensure that the sensor coverage is comprehensive and reasonable. According to the collected spatial data, various types of sensor networks are deployed on the aquaculture workboat, including lidar sensors, infrared camera sensors, and temperature and humidity sensors. Through the collaborative work of multiple sensors, the environmental conditions of the aquaculture workboat can be fully monitored to ensure the diversity and comprehensiveness of the data. The data obtained by the deployed sensor network is collected and preliminarily processed (such as denoising and deduplication) to improve the data quality, reduce useless data and duplicate data, and provide a clean and reliable data source for subsequent data integration and analysis. The 5G network is used to transmit the preliminarily processed data packets to the central control system, where the multi-source sensor data is integrated to generate the environmental perception integrated data of the aquaculture workboat, and the environmental state is estimated. The high-speed 5G network is used to realize real-time data transmission and processing, and the response speed and data processing efficiency of the system are improved. Environmental state estimation provides a comprehensive understanding of the workboat environment. According to the environmental status estimation data, the environmental monitoring data of the aquaculture vessel is updated in real time, so as to realize real-time monitoring of the aquaculture vessel environment, timely discover and deal with potential problems, and ensure the stable operation of the aquaculture vessel.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: preprocessing the real-time environmental monitoring data of the aquaculture vessel to generate standard real-time environmental monitoring data of the aquaculture vessel, wherein the data preprocessing includes filling missing values, processing abnormal values ​​and standardizing data;

[0019] Step S22: performing spatiotemporal data synchronization on the real-time environmental monitoring data of the standard aquaculture vessel to generate aquaculture vessel environmental synchronization data;

[0020] Step S23: constructing an environmental model based on the aquaculture vessel environment synchronization data to generate a comprehensive aquaculture environment model;

[0021] Step S24: Perform dynamic path planning of robot loading motion on the spatial data of the aquaculture vessel through a comprehensive aquaculture environment model to generate a robot loading motion path planning diagram.

[0022] The present invention adopts specific methods (such as mean filling, interpolation, etc.) to fill missing values ​​in monitoring data to ensure the integrity of data. Detect and process abnormal values ​​(such as values ​​beyond a reasonable range) in monitoring data, and the methods adopted include deleting abnormal values, replacing abnormal values, etc. Convert data of different dimensions to the same scale so that the data can be compared and processed on the same horizontal line, thereby improving the stability and accuracy of the model. Perform spatiotemporal synchronization processing on the pre-processed real-time environmental monitoring data to ensure the consistency of the data in time and space. The goal of this step is to generate a unified, spatiotemporally aligned environmental monitoring data set. Using the synchronized breeding ship environmental data, a comprehensive breeding environment model is constructed. This model contains the dynamic changes of various environmental parameters (such as temperature, humidity, water quality, etc.) and can accurately reflect the environmental conditions in the breeding ship. Based on the comprehensive breeding environment model, the spatial data of the breeding ship is analyzed to plan the optimal path for the robot to load materials. The path planning takes into account the dynamic changes of environmental factors to ensure that the robot can efficiently complete the loading task. The path planning map finally generated will be used to guide the actual operation of the robot. Data preprocessing ensures the accuracy and integrity of environmental monitoring data, laying the foundation for subsequent processing and analysis. Spatiotemporal data synchronization ensures the consistency of data in time and space, improving the reliability of the model. Environmental model construction provides a comprehensive understanding and prediction capability of the environmental conditions in the aquaculture vessel. Robot path planning optimizes the robot's working path, improves feeding efficiency and accuracy, and reduces resource waste and operation time.

[0023] Preferably, step S23 includes the following steps:

[0024] Step S231: performing environmental data cluster analysis on the aquaculture vessel environmental synchronization data to generate environmental cluster analysis data; performing key environmental feature screening on the aquaculture vessel environmental synchronization data according to the environmental cluster analysis data to generate a key environmental feature set;

[0025] Step S232: extracting spatiotemporal features of the key environmental feature set by using a convolutional neural network algorithm and a long short-term memory neural network algorithm to generate a spatiotemporal feature matrix of the workboat environment; using the spatiotemporal feature matrix of the workboat environment to fuse multi-scale environmental changes of the synchronous data of the aquaculture workboat environment to generate multi-scale environmental fusion data;

[0026] Step S233: Perform environmental state modeling according to the multi-scale environmental fusion data to generate an environmental state model; perform parameter tuning on the environmental state model to generate optimized environmental model parameters;

[0027] Step S234: Optimizing the environmental state model using the optimized environmental model parameters, thereby generating a comprehensive breeding environment model.

[0028] The present invention reduces data redundancy and improves the efficiency of data processing and analysis through data cluster analysis and key feature screening. The use of convolutional neural networks and long short-term memory neural networks for spatiotemporal feature extraction can more accurately capture complex patterns in environmental data and improve the prediction ability of the model. Through the fusion of multi-scale environmental changes, the model can adapt to environmental changes at different time and space scales, enhancing the stability and robustness of the model. The optimized environmental model parameters make the model's consumption of computing resources and time more reasonable, and improve resource utilization efficiency. The comprehensive aquaculture environment model provides comprehensive and accurate environmental status information, provides strong support for aquaculture management decisions, and reduces decision-making risks. By constructing and optimizing the environmental model, accurate environmental data support is provided for robot feeding path planning, and the level of automated management of aquaculture vessels is improved. Accurate environmental prediction and optimized feeding path planning reduce resource waste and operation time, and reduce the operating cost of aquaculture vessels. The application of the comprehensive environmental model helps to maintain the optimal aquaculture environment and improve aquaculture output and quality.

[0029] Preferably, step S24 includes the following steps:

[0030] Step S241: performing environmental data analysis on the spatial data of the aquaculture vessel through a comprehensive aquaculture environment model to generate analyzed environmental data of the aquaculture vessel;

[0031] Step S242: spatially rasterizing the analyzed environmental data of the aquaculture vessel to generate spatial rasterized data of the aquaculture vessel; constructing an internal map based on the spatial rasterized data of the aquaculture vessel to generate an internal rasterized map of the aquaculture vessel;

[0032] Step S243: performing obstacle detection on the internal rasterized map of the aquaculture workboat to generate obstacle detection data for the internal obstacle of the aquaculture workboat; performing obstacle avoidance path planning on the internal rasterized map of the aquaculture workboat by using the obstacle detection data for the internal obstacle of the aquaculture workboat to generate a robot obstacle avoidance planning path planning;

[0033] Step S244: confirming the robot loading target position data based on the preset work requirements of the aquaculture workboat; performing task allocation path planning on the internal rasterized map of the aquaculture workboat according to the robot loading target position data, and generating the robot target task allocation planning path;

[0034] Step S245: overlap the robot obstacle avoidance planning path planning and the robot target task allocation planning path to generate a robot feeding motion path planning diagram.

[0035] The present invention analyzes environmental data through a comprehensive breeding environment model, can accurately identify and extract various environmental data of the breeding ship, and improves the accuracy and reliability of data analysis. Through spatial rasterization and internal map construction, the internal space of the breeding ship is structured, space utilization is optimized, and more efficient management and operation are facilitated. Obstacle detection is performed using the internal rasterized map, which can accurately identify obstacles in the breeding ship, provide timely obstacle avoidance information, and improve the safety and reliability of the system. Obstacle avoidance path planning is performed through obstacle detection data to ensure that the robot can effectively avoid obstacles in a complex environment, and ensure the smooth and safe loading of the robot. Task allocation path planning is performed according to preset work requirements and target position data to realize intelligent allocation of robot tasks and improve work efficiency and accuracy. Through the overlap and splicing of obstacle avoidance path planning and task allocation path, the optimal robot loading motion path planning diagram is generated to ensure that the robot can complete the loading task with the shortest path and high efficiency. Combining the above steps, the system can automatically and intelligently complete environmental analysis, obstacle detection and path planning, greatly improving the overall operation efficiency of the breeding ship. Automated path planning and environment analysis reduce reliance on manual operations, lower labor costs, and reduce errors and risks caused by manual operations.

[0036] Preferably, step S3 comprises the following steps:

[0037] Step S31: Acquire robot feeding reserve data and robot initial mechanical arm angle control data;

[0038] Step S32: Compare the robot's material loading reserve data with a preset material loading reserve threshold. When the robot's material loading reserve data is greater than or equal to the material loading reserve threshold, the corresponding robot is marked as a high-reserve material loading mode; when the robot's material loading reserve data is less than the material loading reserve threshold, the corresponding robot is marked as a low-reserve material loading mode.

[0039] Step S33: Analyze the path difficulty of the robot feeding motion path planning diagram to generate feeding path difficulty data; perform the first feeding operation control of the robot on the initial robot arm angle control data according to the high reserve feeding mode based on the feeding path difficulty data to generate the first feeding operation control data of the robot;

[0040] Step S34: Based on the feeding path difficulty data, the robot's initial mechanical arm angle control data is controlled according to the low-reserve feeding mode to perform the second feeding operation control of the robot, and the second feeding operation control data of the robot is generated; the robot control instruction is generated through the robot's first feeding operation control data and the robot's second feeding operation control data to obtain the robot's multi-mode control instruction.

[0041] The present invention obtains and analyzes the robot feeding reserve data, and the system can dynamically adjust the robot working mode according to the reserve level to ensure the reasonable allocation and efficient use of resources. According to the feeding reserve data and the preset threshold, the high reserve and low reserve modes are automatically distinguished, so that the robot can adopt the most suitable feeding strategy under different reserve levels to improve the feeding efficiency. The difficulty of the feeding path is analyzed to identify the difficult points and obstacles in the path, which can provide a safer and more efficient path planning for the robot. According to the path difficulty data, the angle and movement mode of the robot mechanical arm are intelligently adjusted to adapt to different working environments and feeding requirements, and the accuracy and flexibility of the operation are improved. By distinguishing between high reserve and low reserve modes, the system can realize the multi-mode operation of the robot under different working conditions, and improve the adaptability and robustness of the system. Real-time analysis and adjustment of the robot feeding mode and path planning enable the system to quickly respond to environmental changes and changes in the reserve level, and improve the response speed of the overall system. Through intelligent control and path planning, the error caused by human operation is reduced, and the reliability and stability of the system operation are improved. The path difficulty analysis and intelligent obstacle avoidance functions improve the safety of the system operation and reduce the failure rate and accident rate of the robot in a complex environment. Multi-mode control and real-time adjustment mechanisms improve the stability and continuity of the system under different working conditions, ensuring the smooth operation of the system. Through flexible robotic arm angle control and multi-mode feeding strategies, the system can adapt to the complex and changing working environment in the aquaculture vessel, improving the adaptability of the system.

[0042] Preferably, step S33 includes the following steps:

[0043] Step S331: dividing the path space plane of the robot feeding motion path planning diagram to generate path transverse plane space data and path longitudinal plane space data;

[0044] Step S332: Performing a path curvature analysis on the path transverse spatial data to generate path transverse curvature data; performing an elevation resistance analysis on the path longitudinal spatial data to generate path longitudinal elevation resistance data; performing a path difficulty analysis based on the path transverse curvature data and the path longitudinal elevation resistance data to generate feeding path difficulty data;

[0045] Step S333: Analyze the impact of the high-reserve feeding mode on the robot's operating performance based on the feeding path difficulty data, and generate high-reserve robot performance impact data; perform high-performance priority angle control on the robot's initial manipulator arm angle control data through the high-reserve robot performance impact data, and generate the robot's first feeding operation control data.

[0046] The present invention can understand the geometric characteristics and changes of the path in detail by performing spatial plane division and curvature analysis on the path, thereby realizing more accurate path planning and adjustment. The comprehensive analysis of the curvature of the path cross section and the elevation resistance of the longitudinal surface helps to identify the difficult points in the path, provides a scientific basis for the subsequent path optimization and adjustment, and optimizes the path selection in the feeding process. By comprehensively analyzing the curvature and elevation resistance data of the path, the difficulty data of the feeding path is generated, so that the system can better adapt to the complex path environment and improve the effectiveness of path planning. Based on the path difficulty data, the robot operation performance in the high reserve mode is analyzed, and the control parameters of the robot arm can be optimized to ensure the efficient operation of the robot under high reserve conditions. Through high-performance priority angle control, the robot can achieve the best operation performance in the high reserve mode, and improve the feeding efficiency and operation accuracy. The detailed analysis and optimization of the path can reduce the operational difficulties of the robot when performing tasks, and improve the smoothness and stability of the operation. The optimized path and angle control can improve the quality of the feeding work, reduce errors and failures, and improve the overall working effect of the system. Through comprehensive analysis of the path and robot performance, the system can be flexibly adjusted according to different working conditions and environmental changes, enhancing the adaptability of the system. Through precise path analysis and optimized control, the possible waste of resources in the robot loading process is reduced and the utilization efficiency of resources is improved.

[0047] Preferably, step S34 includes the following steps:

[0048] Step S341: Analyze the impact of the low-reserve feeding mode on the robot's operating energy consumption according to the feeding path difficulty data, and generate low-reserve robot energy consumption impact data; perform low-energy priority angle control on the robot's initial mechanical arm angle control data according to the low-reserve robot performance impact data, and generate the robot's second feeding operation control data;

[0049] Step S342: performing instruction packaging on the robot first loading operation control data and the robot second loading operation control data to generate a robot mode control packaging instruction;

[0050] Step S343: Perform control instruction simulation on the robot mode control encapsulation instruction to generate a robot multi-modal control instruction.

[0051] By analyzing the energy consumption impact data in the low reserve mode, the system can optimize the energy consumption management in the operation of the robot, reduce unnecessary energy consumption, and improve energy efficiency. Low energy consumption priority angle control enables the robot to complete the feeding task with lower energy consumption in the low reserve mode, improves energy efficiency, and reduces operating costs. By encapsulating the operation control data in different modes, a comprehensive robot mode control instruction is generated to ensure that the robot can execute the optimal operation strategy under different working conditions. The encapsulated control instructions can adapt to different working modes more flexibly, improving the flexibility and adaptability of the robot operation. By simulating the control instructions, the effect of the instructions can be verified and optimized in advance, ensuring that the robot can perform efficient tasks as expected in actual operation. The simulation can identify potential problems and risks and make adjustments, reduce errors in actual operation, and improve the reliability and stability of the system. Through intelligent energy consumption control and efficient instruction generation, the operation time and cost are reduced, and the economic benefits of the overall system are improved. The comprehensive control instructions and optimized energy consumption management make the robot more efficient in performing tasks, and improve the completion efficiency and quality of the feeding task. Through multi-modal control and simulation optimization, the intelligent development of robot control technology has been promoted and the overall intelligence level of the system has been improved. Automatic generation and optimization of control instructions reduce dependence on human intervention, reduce errors and uncertainties caused by human operation, and improve the automation level of the system.

[0052] In this specification, a robot operation control system is provided, which is used to execute the above-mentioned robot operation control method, and the robot operation control system includes:

[0053] The environmental perception module is used to obtain the spatial data of the aquaculture workboat; deploy a multi-source sensor network based on the spatial data of the aquaculture workboat to generate a multi-source sensor network; integrate the sensor data of the multi-source sensor network to generate environmental perception integrated data of the aquaculture workboat; perform real-time environmental updates on the environmental perception integrated data of the aquaculture workboat to generate real-time environmental monitoring data of the aquaculture workboat;

[0054] The path planning module is used to synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture workboat and generate the environmental synchronization data of the aquaculture workboat; to construct the environmental model based on the environmental synchronization data of the aquaculture workboat and generate a comprehensive aquaculture environment model; to perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture workboat through the comprehensive aquaculture environment model and generate a path planning diagram of the robot loading motion;

[0055] An operation control module is used to obtain the robot's feeding reserve data and the robot's initial mechanical arm angle control data; divide the robot's feeding reserve data into reserve modes according to a preset feeding reserve threshold value, and generate a high-reserve feeding mode and a low-reserve feeding mode; perform a path difficulty analysis on the robot's feeding motion path planning diagram, and generate feeding path difficulty data; perform multi-modal feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data, and generate a robot multi-mode control instruction;

[0056] The control feedback module is used to monitor the robot's feeding and feeding through the robot's multi-mode control instructions, generate the robot's feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection, so as to execute the robot's normal operation control operation.

[0057] The present invention also provides a robot, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, for use in the operation control method of the robot as described above.

[0058] The beneficial effect of the present invention is that through the deployment and data integration of a multi-source sensor network, a comprehensive perception of the environment of the aquaculture vessel is achieved, accurate and real-time environmental monitoring data is provided, and the accuracy of environmental management is improved. The real-time updated environmental monitoring data ensures the immediate understanding of the internal environment of the aquaculture vessel, helps to quickly respond to environmental changes and potential problems, and improves the real-time nature of aquaculture management. Through the synchronization of spatiotemporal data and the construction of a comprehensive environmental model, dynamic path planning for the robot's feeding movement can be achieved, the path selection is optimized, and obstacles and efficiency losses in the feeding process are reduced. The reserve mode is divided according to the feeding reserve data and threshold, so that the robot can adopt the most suitable working mode under different reserve conditions, thereby improving resource utilization and work efficiency. Accurate data support is provided for the analysis of the difficulty of the feeding path, and the working performance and efficiency of the robot are optimized through the multi-modal feeding operation control of high reserve and low reserve modes. The generated robot multi-mode control instructions can adapt to different working environments and task requirements, and improve the flexibility and adaptability of the robot operation. The real-time transmission of the robot feeding and feeding monitoring data and the abnormality detection of the central control system can timely discover and handle abnormal situations in the robot operation, ensuring the stability and safety of the system. Comprehensive environmental data, intelligent path planning and anomaly detection mechanisms improve the overall reliability of the system and reduce the possibility of failures and operational interruptions. Optimized path planning and intelligent reserve mode management improve the efficiency and accuracy of robot loading and shorten the task completion time. Through the automated path planning and monitoring system, the reliance on manual operation is reduced, the possibility of human error is reduced, and the automation level of the system is improved. Therefore, the present invention improves the stability and intelligence level of robot operation control through real-time environmental perception, dynamic path planning, intelligent loading control and anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic diagram of the steps of a robot operation control method;

[0060] Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0061] Figure 3 for Figure 2 Detailed implementation steps of step S33;

[0062] Figure 4 for Figure 2 Detailed implementation steps of step S34;

[0063] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0064] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0066] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0067] To achieve this, please refer to Figures 1 to 4 , a robot operation control method, the method comprising the following steps:

[0068] Step S1: Acquire spatial data of aquaculture vessels; deploy a multi-source sensor network according to the spatial data of aquaculture vessels to generate a multi-source sensor network; integrate sensor data of the multi-source sensor network to generate environmental perception integrated data of aquaculture vessels; perform real-time environmental updates on the environmental perception integrated data of aquaculture vessels to generate real-time environmental monitoring data of aquaculture vessels;

[0069] Step S2: synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture vessel to generate the environmental synchronization data of the aquaculture vessel; construct an environmental model based on the environmental synchronization data of the aquaculture vessel to generate a comprehensive aquaculture environment model; perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture vessel through the comprehensive aquaculture environment model to generate a robot loading motion path planning diagram;

[0070] Step S3: obtaining the robot feeding reserve data and the robot initial mechanical arm angle control data; dividing the robot feeding reserve data into reserve modes according to a preset feeding reserve threshold value, generating a high-reserve feeding mode and a low-reserve feeding mode; performing a path difficulty analysis on the robot feeding motion path planning diagram, generating feeding path difficulty data; performing multi-modal feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data, generating a robot multi-mode control instruction;

[0071] Step S4: Monitor the robot's feeding and feeding through the robot's multi-mode control instructions to generate robot feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection to execute the robot's normal operation control operation.

[0072] The present invention can comprehensively and accurately capture the environmental information in the ship by acquiring the spatial data of the aquaculture vessel and deploying a multi-source sensor network, ensuring that the basic data for the robot operation is sufficient. The sensor data is integrated and updated in real time, so that the environmental perception information is always kept up to date, and the robot's response speed and accuracy to environmental changes are improved. Real-time monitoring data can help the robot avoid collisions and other safety hazards and enhance the safety of the operation process. Through spatiotemporal data synchronization and environmental model construction, the robot can perform dynamic path planning based on the latest environmental information and adapt to environmental changes. The comprehensive aquaculture environment model can accurately calculate the optimal feeding motion path, improve the efficiency and accuracy of the robot's operation, and reduce the feeding time. The optimized path planning reduces the robot's movement time and energy consumption, and improves the overall operation efficiency. By dividing the feeding reserve data into modes, intelligent control of high reserve and low reserve modes is realized, and the feeding strategy is adjusted according to actual needs to improve resource utilization. The difficulty of the path is analyzed, and the feeding mode can be adjusted according to the complexity of the path to ensure that the robot can successfully complete the task under various environmental conditions. Multimodal feeding operation control allows the robot to select different feeding modes according to actual conditions, which increases the flexibility and adaptability of operation. By real-time transmission and anomaly detection of the monitoring data of the robot's feeding, abnormal situations in the system can be discovered and handled in a timely manner, reducing the occurrence of failures. The ability to detect anomalies and control operations in real time enhances the stability and reliability of the system and reduces the risks caused by operating errors or equipment failures. Through the anomaly detection function of the central control system, accurate normal operation control can be performed to ensure the robot's operating effect and stable operation. Therefore, the present invention improves the stability and intelligence level of the robot's operation control through real-time environmental perception, dynamic path planning, intelligent feeding control and anomaly detection.

[0073] In the embodiment of the present invention, reference Figure 1The above is a schematic flow chart of the steps of a robot operation control method of the present invention. In this example, the robot operation control method includes the following steps:

[0074] Step S1: Acquire spatial data of aquaculture vessels; deploy a multi-source sensor network according to the spatial data of aquaculture vessels to generate a multi-source sensor network; integrate sensor data of the multi-source sensor network to generate environmental perception integrated data of aquaculture vessels; perform real-time environmental updates on the environmental perception integrated data of aquaculture vessels to generate real-time environmental monitoring data of aquaculture vessels;

[0075] In an embodiment of the present invention, the spatial layout data of the aquaculture vessel is obtained by using high-precision measuring instruments (such as laser radar, photogrammetry, etc.). The collected spatial data is processed and corrected to ensure its accuracy and completeness, and the processed spatial data is stored in a database for subsequent use. Based on the spatial data, the environmental monitoring needs of the aquaculture vessel are analyzed, including the monitoring area, the required sensor types (such as temperature sensors, humidity sensors, gas sensors, etc.) and their quantity. Specific sensors are selected and their installation locations are considered to cover the target area to the greatest extent. The layout positions of the sensors are determined according to the spatial data to ensure that the sensors can effectively cover the entire aquaculture area, and the network topology is designed to achieve data transmission and fusion. The sensors are installed on the aquaculture vessel according to the design plan to ensure their stability and normal operation. The sensor network is configured, including setting the communication protocol, node address and data transmission path, and performing system testing to ensure that the sensor network can collect and transmit data normally. Data is collected from each sensor node, and the collected data is preprocessed, including data cleaning, format conversion, etc. Data fusion algorithms (such as weighted average, Kalman filtering, etc.) are used to integrate data from different sensors into a comprehensive environmental perception data set. Analyze the integrated data, extract environmental change trends and key indicators, generate environmental perception reports, and provide them to relevant personnel for decision support. Continuously monitor sensor data, update real-time data, process and analyze real-time data, update environmental monitoring data, and automatically adjust the environmental control measures of the aquaculture vessel (such as temperature adjustment, humidity control, etc.) according to real-time data changes.

[0076] Step S2: synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture vessel to generate the environmental synchronization data of the aquaculture vessel; construct an environmental model based on the environmental synchronization data of the aquaculture vessel to generate a comprehensive aquaculture environment model; perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture vessel through the comprehensive aquaculture environment model to generate a robot loading motion path planning diagram;

[0077] In the embodiment of the present invention, by assigning a timestamp to the data of each sensor, it is ensured that the data has a time dimension. The environmental data collected at different time points are time-aligned, and the missing data are filled using time interpolation or data interpolation methods. The spatial data is spatially aligned with the time-synchronized data to ensure the consistency of the data in space. The spatiotemporal data is synthesized to form a comprehensive data set containing time and space dimensions, that is, the environmental synchronization data of the aquaculture vessel. The environmental synchronization data is analyzed to identify the key factors and change patterns in the environment. A specific modeling method (such as a statistical model, a machine learning model, a physical model, etc.) is selected to construct an environmental model. The selected model is trained using the environmental synchronization data, and the model parameters are optimized so that it can accurately describe the dynamic changes of the aquaculture environment. The constructed environmental model is verified to ensure that it can accurately predict and simulate the actual environmental conditions. Various environmental factors (such as temperature, humidity, gas concentration, etc.) are integrated into a comprehensive model, considering their mutual influence and comprehensive effects. The model parameters are optimized to ensure its robustness and accuracy under different environmental conditions. The comprehensive aquaculture environment model is applied to actual environmental monitoring and prediction tasks. The spatial data of the aquaculture vessel is analyzed using a comprehensive environmental model to identify the optimal path and task area for robot loading. Select a specific path planning algorithm (such as Dijkstra algorithm, rapidly expanding random tree (RRT), etc.) to perform dynamic path planning. Dynamically adjust the path planning according to the environmental model to cope with environmental changes and obstacles, and generate a motion path planning diagram for the robot to load materials, including detailed path instructions and action requirements. Design a motion path diagram to show the robot's motion path, target points, and obstacle avoidance areas in the aquaculture vessel, and convert the path planning diagram into a format that the robot control system can recognize and execute (such as path files, instruction sets, etc.). Verify and test the generated path to ensure that the robot can smoothly perform the loading task according to the planned path.

[0078] Step S3: obtaining the robot feeding reserve data and the robot initial mechanical arm angle control data; dividing the robot feeding reserve data into reserve modes according to a preset feeding reserve threshold value, generating a high-reserve feeding mode and a low-reserve feeding mode; performing a path difficulty analysis on the robot feeding motion path planning diagram, generating feeding path difficulty data; performing multi-modal feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data, generating a robot multi-mode control instruction;

[0079] In an embodiment of the present invention, the current loading reserve data, including the storage amount, storage location and other information, is obtained from the robot or the storage system. The angle control data of the robot arm in the initial state is obtained, including the angle, speed and acceleration of each joint. The collected data is stored in a database or a data cache system for subsequent processing and analysis. According to system requirements and practical applications, the threshold of the reserve mode is set. For example, the threshold of the high reserve mode is set to a storage amount exceeding a certain level, while the low reserve mode is set to a storage amount below the level. The loading reserve data is compared with the preset threshold, and the data is divided into a high reserve loading mode and a low reserve loading mode according to the level of the reserve, and the data of different modes are labeled for subsequent operation control and analysis. The features in the motion path, such as the path length, curvature, obstacle density, etc., are analyzed to determine the complexity of the path. Based on the path features, a scoring system or algorithm is used to quantitatively analyze the difficulty of the path, involving calculating factors such as the slope, width, curvature and other factors of the path, and generating the loading path difficulty data, including the difficulty score and related information of each path. Identify the control strategy to be adopted based on the reserve mode (high reserve or low reserve) and the path difficulty. Develop control strategies for high reserve mode and low reserve mode respectively. For example: High reserve mode: When the path difficulty is low, you can choose a faster running speed and efficient path planning; when the path difficulty is high, you can choose a slower speed and adopt an obstacle avoidance strategy. Low reserve mode: Due to the small reserve, the path needs to be optimized to reduce the travel and improve the accuracy of loading. Generate multi-mode control instructions for the robot based on the developed control strategy. These instructions should include the motion trajectory, speed, acceleration and other control parameters of the robot arm. Send the generated control instructions to the robot system and perform the relevant loading operations. Monitor the robot loading process in real time to ensure that it runs according to the expected path and mode. Verify the effect of the loading process to ensure that it meets the expected loading mode and path difficulty requirements. If there is a deviation, adjust the control strategy and regenerate the control instructions.

[0080] Step S4: Monitor the robot's feeding and feeding through the robot's multi-mode control instructions to generate robot feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection to execute the robot's normal operation control operation.

[0081] In an embodiment of the present invention, the robot is operated to perform feeding tasks according to the multi-mode control instructions generated previously. Ensure that the instructions cover all actions of the robot arm, including motion path, speed control, feeding accuracy, etc. During the feeding process, the operation data of the robot is collected in real time. These data include: the position and angle of the robot arm, the feeding amount and feeding rate, the sensor data during the feeding process (such as force sensor, position sensor, etc.), operation status and error log. The collected real-time data is recorded in a local database or data cache system to ensure the integrity and accuracy of the data. The recorded data is preliminarily processed, such as data cleaning and formatting, to ensure that it is suitable for further analysis. Analyze the monitoring data to evaluate whether the robot's feeding process meets expectations, including feeding accuracy, feeding rate, operation stability, etc. The processed feeding and feeding monitoring data is transmitted to the central control system through a secure network protocol. Ensure that the data is not tampered with or lost during transmission. In the central control system, the received feeding and feeding monitoring data is integrated with other relevant data (such as environmental monitoring data, path planning data, etc.) to form a complete operation data set. The integrated data is analyzed using the set anomaly detection algorithm. These algorithms include statistical analysis, machine learning models, rule engines, etc. to identify abnormal patterns or behaviors that do not meet expectations. The detection system identifies abnormal situations, such as abnormal loading volume, deviation in robot arm movement, and running speed that does not meet expectations. Once an abnormal situation is detected, the system should generate an alarm and notify the relevant operator or maintenance personnel so that the problem can be handled in time. Based on the results of abnormal detection, the robot control strategy and parameters are automatically adjusted to correct the problems found. For example, adjust the movement trajectory of the robot arm, modify the loading rate, etc. In the case that automatic adjustment cannot solve the problem, the operator should manually intervene and adjust according to the abnormal situation to ensure that the robot resumes normal operation. Record the abnormal handling process and results, and update the feedback information to the central control system to optimize future abnormal detection and control strategies.

[0082] Preferably, step S1 comprises the following steps:

[0083] Step S11: Acquire the spatial data of the aquaculture vessel;

[0084] Step S12: deploying a multi-source sensor network according to the spatial data of the aquaculture vessel to generate a multi-source sensor network, wherein the multi-source sensor network deployment includes the deployment of a lidar sensor, an infrared camera sensor, and a temperature and humidity sensor;

[0085] Step S13: collecting sensor data from the multi-source sensor network to obtain a multi-source sensor data packet; performing preliminary data processing on the multi-source sensor data packet to generate a multi-source sensor preliminary processing data packet, wherein the preliminary data processing includes data denoising and data deduplication;

[0086] Step S14: Transmitting the multi-source sensing preliminary processing data packet to the central control system through the 5G network to integrate the multi-source sensing data and generate the aquaculture vessel environmental perception integrated data; estimating the environmental state of the aquaculture vessel environmental perception integrated data and generating environmental state estimation data;

[0087] Step S15: Real-time environmental status monitoring and updating of the environmental perception integrated data of the aquaculture vessel is performed through the environmental status estimation data to generate real-time environmental monitoring data of the aquaculture vessel.

[0088] In an embodiment of the present invention, the spatial parameters of the interior and surrounding environment of the workboat are measured by using a three-dimensional laser scanner or other spatial data acquisition tools. Including the internal spatial layout of the hull, the size of each area, the location of equipment, etc. The acquired spatial data is organized into electronic formats, such as point cloud data, CAD drawings, etc., to facilitate subsequent processing and analysis. Determine the installation position of the laser radar to maximize the coverage of key areas inside the workboat, such as breeding ponds, equipment areas, etc. Install and calibrate the laser radar to ensure that it can accurately measure distances and generate high-precision spatial data. Install infrared cameras at key locations on the workboat to monitor temperature changes and perform visual inspections. Set the viewing angle and resolution of the camera to obtain the required visual data. Deploy temperature and humidity sensors in different areas to monitor the temperature and humidity of the environment. Ensure that the sensor installation location can represent the temperature and humidity conditions of the entire breeding environment. Collect data in real time from the laser radar, infrared camera, and temperature and humidity sensor. The collected data includes distance data, infrared images, temperature and humidity values, etc. Use filtering algorithms (such as mean filtering, median filtering, etc.) to remove noise from sensor data. Identify and remove duplicate data records to improve the validity of the data. Transmit the preliminarily processed data packets to the central control system via the 5G network. Ensure the security and integrity of data transmission. In the central control system, integrate data from different sensors to generate environmental perception integrated data. Use data fusion technologies, such as Kalman filtering and Bayesian inference, to improve the comprehensive accuracy of data. Based on the integrated data, use environmental state estimation algorithms (such as machine learning models, statistical models, etc.) to estimate the current environmental state. Generate environmental state estimation data, including the changing trend of ambient temperature and humidity, possible abnormal conditions, etc. Based on the environmental state estimation data, monitor the environmental state of the aquaculture vessel in real time. Update the environmental state monitoring data to provide the latest status information of the current environment. Display real-time environmental monitoring data in the central control system, including charts, alarms, etc. Generate automated feedback and control suggestions based on real-time monitoring data to optimize the aquaculture environment and operation strategies.

[0089] Preferably, step S2 comprises the following steps:

[0090] Step S21: preprocessing the real-time environmental monitoring data of the aquaculture vessel to generate standard real-time environmental monitoring data of the aquaculture vessel, wherein the data preprocessing includes filling missing values, processing abnormal values ​​and standardizing data;

[0091] Step S22: performing spatiotemporal data synchronization on the real-time environmental monitoring data of the standard aquaculture vessel to generate aquaculture vessel environmental synchronization data;

[0092] Step S23: constructing an environmental model based on the aquaculture vessel environment synchronization data to generate a comprehensive aquaculture environment model;

[0093] Step S24: Perform dynamic path planning of robot loading motion on the spatial data of the aquaculture vessel through a comprehensive aquaculture environment model to generate a robot loading motion path planning diagram.

[0094] In an embodiment of the present invention, by detecting missing values ​​in real-time environmental monitoring data, a specific filling method is selected, such as mean filling, median filling, interpolation or model-based filling (such as KNN filling), etc., to fill in missing data. Statistical analysis or machine learning algorithms (such as Z-score, IQR method, isolation forest, etc.) are used to identify outliers in the data, and outliers can be deleted, replaced with reasonable values, or corrected according to the situation. The data is standardized to meet unified standards, such as a mean of 0, a standard deviation of 1, or the data is scaled to a specific range (such as 0-1). Ensure that the timestamps of all environmental monitoring data are consistent for time series analysis. For data with inconsistent timestamps, interpolation methods (such as linear interpolation and spline interpolation) are used for time alignment. According to the spatial layout of the aquaculture ship, the data of different sensors are spatially aligned. The sensor data is fused into a unified spatial model to generate data with a consistent spatial reference. The data that has been synchronized in time and space is integrated into the synchronized data of the aquaculture ship environment, ready for subsequent analysis. According to the characteristics of environmental synchronization data, select specific environmental modeling methods (such as physical models, statistical models, machine learning models, etc.). Use historical data and real-time data to train the model and establish a comprehensive aquaculture environment model. Verify and evaluate the model to ensure its accuracy and stability in practical applications, adjust model parameters according to the verification results, and optimize model performance. Based on the training and verification results, generate a comprehensive aquaculture environment model for subsequent path planning and decision support. Use the spatial data of the aquaculture workboat to analyze the layout and space constraints inside the workboat, and determine the target area, path requirements, and obstacle locations for the robot's loading movement. According to the spatial data and environmental model, select a specific path planning algorithm (such as Dijkstra algorithm, RRT algorithm, etc.). Optimize the planned path to reduce movement time and improve efficiency, generate a dynamic path planning diagram for the robot's loading movement, and display the path planning diagram on the control interface or robot system for actual operation.

[0095] Preferably, step S23 includes the following steps:

[0096] Step S231: performing environmental data cluster analysis on the aquaculture vessel environmental synchronization data to generate environmental cluster analysis data; performing key environmental feature screening on the aquaculture vessel environmental synchronization data according to the environmental cluster analysis data to generate a key environmental feature set;

[0097] Step S232: extracting spatiotemporal features of the key environmental feature set by using a convolutional neural network algorithm and a long short-term memory neural network algorithm to generate a spatiotemporal feature matrix of the workboat environment; using the spatiotemporal feature matrix of the workboat environment to fuse multi-scale environmental changes of the synchronous data of the aquaculture workboat environment to generate multi-scale environmental fusion data;

[0098] Step S233: Perform environmental state modeling according to the multi-scale environmental fusion data to generate an environmental state model; perform parameter tuning on the environmental state model to generate optimized environmental model parameters;

[0099] Step S234: Optimizing the environmental state model using the optimized environmental model parameters, thereby generating a comprehensive breeding environment model.

[0100] In an embodiment of the present invention, cluster analysis is performed on environmental synchronization data by selecting a specific clustering algorithm (such as K-means, DBSCAN, hierarchical clustering, etc.). The parameters of the clustering algorithm (such as the K value of K-means, ε and MinPts of DBSCAN) are set to effectively group the data. The clustering results are analyzed, environmental clustering analysis data is generated, and patterns and regions of different environmental states are identified. Key environmental features are extracted from the environmental clustering analysis data, which can effectively describe the state and changes of the breeding environment. Feature selection techniques (such as principal component analysis PCA, feature importance ranking, LASSO regression, etc.) are used to screen out features that have an important impact on the environmental state, and a key environmental feature set containing these key environmental features is generated, providing a basis for subsequent feature extraction and modeling. A specific CNN architecture is designed for extracting spatial features. CNN can extract local and global spatial information from environmental data. The CNN model is trained using a key environmental feature set to extract spatial features of environmental data. A specific LSTM architecture is designed for extracting time series features. LSTM can handle time dependencies in environmental data. The LSTM model is trained using the spatial features extracted by CNN and the time series of environmental data to extract time series features. The outputs of CNN and LSTM are combined to generate a spatiotemporal feature matrix of the ship environment to represent the spatiotemporal features of the environmental data. The spatiotemporal feature matrix of the ship environment is analyzed at multiple scales to consider environmental changes at different time and spatial scales. Multiscale fusion techniques (such as wavelet transform, multiscale convolutional network, etc.) are used to fuse environmental features at different scales to generate multiscale environmental fusion data and provide comprehensive environmental state information. Specific modeling methods (such as regression model, time series model, graph model, etc.) are selected for environmental state modeling. The environmental state model is trained using multiscale environmental fusion data to establish a mathematical or computational model of the environmental state and generate an environmental state model to describe and predict the environmental state of the aquaculture ship. The parameters of the environmental state model are adjusted using optimization algorithms (such as grid search, random search, Bayesian optimization, etc.). The model performance is evaluated on the validation set to ensure that the adjustment of the model parameters improves the prediction accuracy and stability of the model, and the optimized environmental model parameters are obtained for further model optimization. The optimized environmental model parameters are applied to the environmental state model for model optimization. Specific techniques such as gradient descent and genetic algorithms are used to further adjust the model to ensure its optimal performance. Based on the optimized model parameters and optimization process, a comprehensive aquaculture environment model is generated. The comprehensive aquaculture environment model is finally verified and tested to ensure its accuracy and effectiveness in practical applications.

[0101] Preferably, step S24 includes the following steps:

[0102] Step S241: performing environmental data analysis on the spatial data of the aquaculture vessel through a comprehensive aquaculture environment model to generate analyzed environmental data of the aquaculture vessel;

[0103] Step S242: spatially rasterizing the analyzed environmental data of the aquaculture vessel to generate spatial rasterized data of the aquaculture vessel; constructing an internal map based on the spatial rasterized data of the aquaculture vessel to generate an internal rasterized map of the aquaculture vessel;

[0104] Step S243: performing obstacle detection on the internal rasterized map of the aquaculture workboat to generate obstacle detection data for the internal obstacle of the aquaculture workboat; performing obstacle avoidance path planning on the internal rasterized map of the aquaculture workboat by using the obstacle detection data for the internal obstacle of the aquaculture workboat to generate a robot obstacle avoidance planning path planning;

[0105] Step S244: confirming the robot loading target position data based on the preset work requirements of the aquaculture workboat; performing task allocation path planning on the internal rasterized map of the aquaculture workboat according to the robot loading target position data, and generating the robot target task allocation planning path;

[0106] Step S245: overlap the robot obstacle avoidance planning path planning and the robot target task allocation planning path to generate a robot feeding motion path planning diagram.

[0107] In an embodiment of the present invention, the spatial data of the aquaculture workboat is parsed by using a comprehensive aquaculture environment model to generate parsed environmental data of the aquaculture workboat, which includes a detailed description of the internal environment of the aquaculture workboat (such as temperature, humidity, spatial layout, etc.), and the parsed environmental data is organized into a unified data format to facilitate subsequent processing and analysis (such as two-dimensional or three-dimensional data format). The parsed environmental data of the aquaculture workboat is spatially rasterized, and the continuous environmental data is converted into data in a grid format. The specific grid size is selected according to the resolution of the environmental data and the application requirements. According to the spatial rasterized data, an internal rasterized map of the aquaculture workboat is constructed to represent the spatial layout and environmental information inside the workboat. The generated rasterized map can be a two-dimensional map or a three-dimensional map, depending on the requirements and application scenarios. An obstacle detection algorithm (such as a detection method based on image processing, laser radar data analysis, etc.) is applied to perform obstacle detection on the internal rasterized map, generate internal obstacle detection data of the aquaculture workboat, and identify the obstacle area in the map. Using obstacle detection data, path planning algorithms (such as Dijkstra algorithm, etc.) are applied to perform obstacle avoidance path planning, generate robot obstacle avoidance planning paths, and ensure that the robot can avoid obstacles during the loading process. According to the preset work requirements of the aquaculture workboat, the target position data of the robot loading (such as material storage location, loading point, etc.) is confirmed, and the target position data is converted into coordinate information on the map. According to the target position data, the path planning algorithm is used to generate task allocation path planning for the robot. The path from the starting position to the target position is planned on the internal rasterized map, considering the priority and constraints of the task. The robot obstacle avoidance planning path and the task allocation path are overlapped and integrated to ensure that the two paths are consistent on the map, and the integrated path is optimized to improve movement efficiency and reduce path conflicts. The obstacle avoidance path planning and the task allocation path are spliced ​​together to generate a complete robot loading motion path planning map. The final generated path planning map includes the complete path from the starting point to the target point, and ensures the continuity and operability of the path.

[0108] As an example of the present invention, refer to Figure 2 As shown, in this example, step S3 includes:

[0109] Step S31: Acquire robot feeding reserve data and robot initial mechanical arm angle control data;

[0110] Step S32: Compare the robot's material loading reserve data with a preset material loading reserve threshold. When the robot's material loading reserve data is greater than or equal to the material loading reserve threshold, the corresponding robot is marked as a high-reserve material loading mode; when the robot's material loading reserve data is less than the material loading reserve threshold, the corresponding robot is marked as a low-reserve material loading mode.

[0111] Step S33: Analyze the path difficulty of the robot feeding motion path planning diagram to generate feeding path difficulty data; perform the first feeding operation control of the robot on the initial robot arm angle control data according to the high reserve feeding mode based on the feeding path difficulty data to generate the first feeding operation control data of the robot;

[0112] Step S34: Based on the feeding path difficulty data, the robot's initial mechanical arm angle control data is controlled according to the low-reserve feeding mode to perform the second feeding operation control of the robot, and the second feeding operation control data of the robot is generated; the robot control instruction is generated through the robot's first feeding operation control data and the robot's second feeding operation control data to obtain the robot's multi-mode control instruction.

[0113] In an embodiment of the present invention, the current loading reserve data is obtained from the robot system. This usually includes the amount of materials stored by the robot, the inventory level, etc., to ensure that the data format is unified and convenient for subsequent processing (such as JSON, CSV, etc.). The initial angle control data of the manipulator is obtained from the robot control system. This includes the angle, position, motion state, etc. of the manipulator. Ensure that the format of the manipulator angle data meets the requirements of the control system (such as angle value, coordinate system information, etc.). Determine the loading reserve threshold, which is usually determined by the system configuration or empirical data, and compare the robot loading reserve data with the preset loading reserve threshold. When the loading reserve data is greater than or equal to the threshold, the corresponding robot is marked as a high-reserve loading mode. When the loading reserve data is less than the threshold, the corresponding robot is marked as a low-reserve loading mode. Using the data in the path planning map, analyze the difficulty of the path (such as obstacle density, path complexity, slope, etc.), generate the loading path difficulty data, and describe the difficulty of walking the path. According to the high-reserve loading mode, the manipulator angle control data is adjusted to optimize the operating efficiency. Generate the robot's first loading operation control data based on the adjusted data to control the robot's operation in high reserve mode. According to the low reserve loading mode, adjust the robot arm angle control data to consider the operation strategy under insufficient reserve conditions. Generate the robot's second loading operation control data based on the adjusted data to control the robot's operation in low reserve mode. Synthesize the robot's first loading operation control data and the robot's second loading operation control data to generate a comprehensive control instruction, generate a robot multi-mode control instruction, and guide the robot's operation behavior in different reserve modes.

[0114] As an example of the present invention, refer to Figure 3 As shown, in this example, step S33 includes:

[0115] Step S331: dividing the path space plane of the robot feeding motion path planning diagram to generate path transverse plane space data and path longitudinal plane space data;

[0116] Step S332: Performing a path curvature analysis on the path transverse spatial data to generate path transverse curvature data; performing an elevation resistance analysis on the path longitudinal spatial data to generate path longitudinal elevation resistance data; performing a path difficulty analysis based on the path transverse curvature data and the path longitudinal elevation resistance data to generate feeding path difficulty data;

[0117] Step S333: Analyze the impact of the high-reserve feeding mode on the robot's operating performance based on the feeding path difficulty data, and generate high-reserve robot performance impact data; perform high-performance priority angle control on the robot's initial manipulator arm angle control data through the high-reserve robot performance impact data, and generate the robot's first feeding operation control data.

[0118] In an embodiment of the present invention, the robot loading motion path planning diagram is divided into transverse and longitudinal spaces. The transverse space is the transverse perspective of the path, and the longitudinal space is the longitudinal perspective of the path. The path transverse space data and the path longitudinal space data are generated, which respectively represent the spatial information of the path in the transverse and longitudinal directions. The projection of the path on the horizontal plane is extracted to obtain the transverse path data. The projection of the path on the vertical plane is extracted to obtain the longitudinal path data. The curvature of the path transverse space data is analyzed. The curvature describes the curvature or curve change of the path on the horizontal plane, and the path transverse curvature data is generated to indicate the curvature difficulty of the path. Specifically, the curvature calculation formula (such as curvature calculation) or tool is used to analyze the transverse space data. The curvature data generally includes a numerical representation or a curve graph of the curvature degree. The influence of the elevation change in the path longitudinal space data on the resistance of the path is analyzed. The elevation resistance describes the influence of the height change of the path on the vertical plane on the movement, and the path longitudinal elevation resistance data is generated to indicate the elevation resistance degree of the path. Specifically, the elevation data analysis tool is used to calculate the elevation resistance of the path, and the elevation resistance data can be displayed in the form of a numerical value or a resistance graph. Combine the path cross curvature data and the path longitudinal elevation resistance data to analyze the path difficulty and generate the loading path difficulty data. Take the path curvature and elevation resistance into consideration. Specifically, weighted average, comprehensive scoring and other methods can be used to combine the cross curvature and longitudinal elevation resistance to generate difficulty data. According to the loading path difficulty data, analyze the impact of the high-reserve loading mode on operation performance. Analyze the impact of the path difficulty on the robot's operating efficiency and performance, generate high-reserve robot performance impact data, describe the impact of different difficulty levels on robot performance, and use performance analysis tools or models (such as simulation, regression analysis) to analyze the impact of difficulty on robot performance in high-reserve mode. According to the high-reserve robot performance impact data, adjust the robot's initial robot arm angle control data, give priority to high-performance operation, generate the robot's first loading operation control data, optimize the robot arm angle control to improve performance, and the generated control data is specifically the angle value, control signal or instruction set, which is directly used in the control system.

[0119] As an example of the present invention, refer to Figure 4 As shown, in this example, step S34 includes:

[0120] Step S341: Analyze the impact of the low-reserve feeding mode on the robot's operating energy consumption according to the feeding path difficulty data, and generate low-reserve robot energy consumption impact data; perform low-energy priority angle control on the robot's initial mechanical arm angle control data according to the low-reserve robot performance impact data, and generate the robot's second feeding operation control data;

[0121] Step S342: performing instruction packaging on the robot first loading operation control data and the robot second loading operation control data to generate a robot mode control packaging instruction;

[0122] Step S343: Perform control instruction simulation on the robot mode control encapsulation instruction to generate a robot multi-modal control instruction.

[0123] In an embodiment of the present invention, the impact analysis of the robot's operating energy consumption in the low-reserve loading mode is performed based on the loading path difficulty data. Analyze how the difficulty of the path affects the robot's energy consumption in the low-reserve mode, generate low-reserve robot energy consumption impact data, describe the impact of different path difficulties on the robot's energy consumption, and specifically use energy consumption models or simulation tools to combine the path difficulty data to calculate the robot's energy consumption in the low-reserve mode. The energy consumption impact data can be displayed in the form of numerical values ​​or charts to indicate energy consumption changes under different path conditions. According to the low-reserve robot energy consumption impact data, the robot's initial mechanical arm angle control data is adjusted, giving priority to low-energy operation, generating the robot's second loading operation control data, optimizing the mechanical arm angle to reduce energy consumption, and specifically using an energy optimization algorithm or control model to adjust the mechanical arm angle to achieve low-energy operation. The generated control data includes the mechanical arm angle setting and control instructions, which are suitable for energy consumption optimization in the low-reserve mode. Integrate and encapsulate the robot's first loading operation control data (high reserve mode) and the robot's second loading operation control data (low reserve mode) to generate robot mode control encapsulation instructions, including control strategies for different reserve modes. Specifically, use instruction encapsulation tools or programming languages ​​to integrate the control data in the two modes into an instruction set, and ensure that the format of the encapsulated instructions meets the requirements of the robot control system (such as instruction files in XML or JSON format). Use simulation tools or software to simulate the robot mode control encapsulation instructions, verify the effect and accuracy of the instructions in actual applications, and generate robot multimodal control instructions based on the simulation results to ensure that the instructions can adapt to different loading modes and environmental conditions. Specifically, use simulation software (such as Gazebo, V-REP) or programming languages ​​(such as Python, MATLAB) to simulate, run the simulation, observe the operation of the robot, check the correctness and performance of the instructions, optimize and adjust the control instructions according to the simulation results, and generate the final robot multimodal control instructions.

[0124] In this specification, a robot operation control system is provided, which is used to execute the above-mentioned robot operation control method, and the robot operation control system includes:

[0125] The environmental perception module is used to obtain the spatial data of the aquaculture workboat; deploy a multi-source sensor network based on the spatial data of the aquaculture workboat to generate a multi-source sensor network; integrate the sensor data of the multi-source sensor network to generate environmental perception integrated data of the aquaculture workboat; perform real-time environmental updates on the environmental perception integrated data of the aquaculture workboat to generate real-time environmental monitoring data of the aquaculture workboat;

[0126] The path planning module is used to synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture workboat and generate the environmental synchronization data of the aquaculture workboat; to construct the environmental model based on the environmental synchronization data of the aquaculture workboat and generate a comprehensive aquaculture environment model; to perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture workboat through the comprehensive aquaculture environment model and generate a path planning diagram of the robot loading motion;

[0127] An operation control module is used to obtain the robot's feeding reserve data and the robot's initial mechanical arm angle control data; divide the robot's feeding reserve data into reserve modes according to a preset feeding reserve threshold value, and generate a high-reserve feeding mode and a low-reserve feeding mode; perform a path difficulty analysis on the robot's feeding motion path planning diagram, and generate feeding path difficulty data; perform multi-modal feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data, and generate a robot multi-mode control instruction;

[0128] The control feedback module is used to monitor the robot's feeding and feeding through the robot's multi-mode control instructions, generate the robot's feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection, so as to execute the robot's normal operation control operation.

[0129] The present invention also provides a robot, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, for use in the operation control method of the robot as described above.

[0130] The beneficial effect of the present invention is that through the deployment and data integration of a multi-source sensor network, a comprehensive perception of the environment of the aquaculture vessel is achieved, accurate and real-time environmental monitoring data is provided, and the accuracy of environmental management is improved. The real-time updated environmental monitoring data ensures the immediate understanding of the internal environment of the aquaculture vessel, helps to quickly respond to environmental changes and potential problems, and improves the real-time nature of aquaculture management. Through the synchronization of spatiotemporal data and the construction of a comprehensive environmental model, dynamic path planning for the robot's feeding movement can be achieved, the path selection is optimized, and obstacles and efficiency losses in the feeding process are reduced. The reserve mode is divided according to the feeding reserve data and threshold, so that the robot can adopt the most suitable working mode under different reserve conditions, thereby improving resource utilization and work efficiency. Accurate data support is provided for the analysis of the difficulty of the feeding path, and the working performance and efficiency of the robot are optimized through the multi-modal feeding operation control of high reserve and low reserve modes. The generated robot multi-mode control instructions can adapt to different working environments and task requirements, and improve the flexibility and adaptability of the robot operation. The real-time transmission of the robot feeding and feeding monitoring data and the abnormality detection of the central control system can timely discover and handle abnormal situations in the robot operation, ensuring the stability and safety of the system. Comprehensive environmental data, intelligent path planning and anomaly detection mechanisms improve the overall reliability of the system and reduce the possibility of failures and operational interruptions. Optimized path planning and intelligent reserve mode management improve the efficiency and accuracy of robot loading and shorten the task completion time. Through the automated path planning and monitoring system, the reliance on manual operation is reduced, the possibility of human error is reduced, and the automation level of the system is improved. Therefore, the present invention improves the stability and intelligence level of robot operation control through real-time environmental perception, dynamic path planning, intelligent loading control and anomaly detection.

[0131] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0132] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A robot operation control method, characterized in that: The following steps are involved: Step S1: Acquire spatial data of aquaculture vessels; deploy a multi-source sensor network based on the spatial data of aquaculture vessels to generate a multi-source sensor network; Integrate sensor data from multi-source sensor networks to generate integrated data on the environment perception of aquaculture vessels; perform real-time environmental updates on the integrated data on the environment perception of aquaculture vessels to generate real-time environmental monitoring data on aquaculture vessels; Step S2: synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture vessel to generate the environmental synchronization data of the aquaculture vessel; construct an environmental model based on the environmental synchronization data of the aquaculture vessel to generate a comprehensive aquaculture environment model; perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture vessel through the comprehensive aquaculture environment model to generate a robot loading motion path planning diagram; Step S3: obtaining the robot feeding reserve data and the robot initial mechanical arm angle control data; dividing the robot feeding reserve data into reserve modes according to a preset feeding reserve threshold, and generating a high reserve feeding mode and a low reserve feeding mode; Perform path difficulty analysis on the robot feeding motion path planning diagram to generate feeding path difficulty data; perform multi-mode feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data to generate robot multi-mode control instructions; Step S4: Monitor the robot's feeding and feeding through the robot's multi-mode control instructions to generate robot feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection to execute the robot's normal operation control operation.

2. The robot operation control method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire the spatial data of the aquaculture vessel; Step S12: deploying a multi-source sensor network according to the spatial data of the aquaculture vessel to generate a multi-source sensor network, wherein the multi-source sensor network deployment includes the deployment of a lidar sensor, an infrared camera sensor, and a temperature and humidity sensor; Step S13: collecting sensor data from the multi-source sensor network to obtain a multi-source sensor data packet; performing preliminary data processing on the multi-source sensor data packet to generate a multi-source sensor preliminary processing data packet, wherein the preliminary data processing includes data denoising and data deduplication; Step S14: Transmitting the multi-source sensing preliminary processing data packet to the central control system through the 5G network to integrate the multi-source sensing data and generate the aquaculture vessel environmental perception integrated data; estimating the environmental state of the aquaculture vessel environmental perception integrated data and generating environmental state estimation data; Step S15: Real-time environmental status monitoring and updating of the environmental perception integrated data of the aquaculture vessel is performed through the environmental status estimation data to generate real-time environmental monitoring data of the aquaculture vessel.

3. The robot operation control method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: preprocessing the real-time environmental monitoring data of the aquaculture vessel to generate standard real-time environmental monitoring data of the aquaculture vessel, wherein the data preprocessing includes filling missing values, processing abnormal values ​​and standardizing data; Step S22: performing spatiotemporal data synchronization on the real-time environmental monitoring data of the standard aquaculture vessel to generate aquaculture vessel environmental synchronization data; Step S23: constructing an environmental model based on the aquaculture vessel environment synchronization data to generate a comprehensive aquaculture environment model; Step S24: Perform dynamic path planning of robot loading motion on the spatial data of the aquaculture vessel through a comprehensive aquaculture environment model to generate a robot loading motion path planning diagram.

4. The robot operation control method according to claim 3, characterized in that: Step S23 includes the following steps: Step S231: performing environmental data cluster analysis on the aquaculture vessel environmental synchronization data to generate environmental cluster analysis data; performing key environmental feature screening on the aquaculture vessel environmental synchronization data according to the environmental cluster analysis data to generate a key environmental feature set; Step S232: extracting spatiotemporal features of the key environmental feature set by using a convolutional neural network algorithm and a long short-term memory neural network algorithm to generate a spatiotemporal feature matrix of the workboat environment; using the spatiotemporal feature matrix of the workboat environment to fuse multi-scale environmental changes of the synchronous data of the aquaculture workboat environment to generate multi-scale environmental fusion data; Step S233: Perform environmental state modeling according to the multi-scale environmental fusion data to generate an environmental state model; perform parameter tuning on the environmental state model to generate optimized environmental model parameters; Step S234: Optimizing the environmental state model using the optimized environmental model parameters, thereby generating a comprehensive breeding environment model.

5. The robot operation control method according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: performing environmental data analysis on the spatial data of the aquaculture vessel through a comprehensive aquaculture environment model to generate analyzed environmental data of the aquaculture vessel; Step S242: spatially rasterizing the analyzed environmental data of the aquaculture vessel to generate spatial rasterized data of the aquaculture vessel; constructing an internal map based on the spatial rasterized data of the aquaculture vessel to generate an internal rasterized map of the aquaculture vessel; Step S243: performing obstacle detection on the internal rasterized map of the aquaculture workboat to generate obstacle detection data for the internal obstacle of the aquaculture workboat; performing obstacle avoidance path planning on the internal rasterized map of the aquaculture workboat by using the obstacle detection data for the internal obstacle of the aquaculture workboat to generate a robot obstacle avoidance planning path planning; Step S244: confirming the robot loading target position data based on the preset work requirements of the aquaculture workboat; performing task allocation path planning on the internal rasterized map of the aquaculture workboat according to the robot loading target position data, and generating the robot target task allocation planning path; Step S245: overlap the robot obstacle avoidance planning path planning and the robot target task allocation planning path to generate a robot feeding motion path planning diagram.

6. The robot operation control method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Acquire robot feeding reserve data and robot initial mechanical arm angle control data; Step S32: Compare the robot's material loading reserve data with a preset material loading reserve threshold. When the robot's material loading reserve data is greater than or equal to the material loading reserve threshold, the corresponding robot is marked as a high-reserve material loading mode; when the robot's material loading reserve data is less than the material loading reserve threshold, the corresponding robot is marked as a low-reserve material loading mode. Step S33: Analyze the path difficulty of the robot feeding motion path planning diagram to generate feeding path difficulty data; perform the first feeding operation control of the robot on the initial robot arm angle control data according to the high reserve feeding mode based on the feeding path difficulty data to generate the first feeding operation control data of the robot; Step S34: Based on the feeding path difficulty data, the robot's initial mechanical arm angle control data is controlled according to the low-reserve feeding mode to perform the second feeding operation control of the robot, and the second feeding operation control data of the robot is generated; the robot control instruction is generated through the robot's first feeding operation control data and the robot's second feeding operation control data to obtain the robot's multi-mode control instruction.

7. The robot operation control method according to claim 6, characterized in that: Step S33 includes the following steps: Step S331: dividing the path space plane of the robot feeding motion path planning diagram to generate path transverse plane space data and path longitudinal plane space data; Step S332: Performing a path curvature analysis on the path transverse spatial data to generate path transverse curvature data; performing an elevation resistance analysis on the path longitudinal spatial data to generate path longitudinal elevation resistance data; performing a path difficulty analysis based on the path transverse curvature data and the path longitudinal elevation resistance data to generate feeding path difficulty data; Step S333: Analyze the impact of the high-reserve feeding mode on the robot's operating performance based on the feeding path difficulty data, and generate high-reserve robot performance impact data; perform high-performance priority angle control on the robot's initial manipulator arm angle control data through the high-reserve robot performance impact data, and generate the robot's first feeding operation control data.

8. The robot operation control method according to claim 6, characterized in that: Step S34 includes the following steps: Step S341: Analyze the impact of the low-reserve feeding mode on the robot's operating energy consumption according to the feeding path difficulty data, and generate low-reserve robot energy consumption impact data; perform low-energy priority angle control on the robot's initial mechanical arm angle control data according to the low-reserve robot performance impact data, and generate the robot's second feeding operation control data; Step S342: performing instruction packaging on the robot first loading operation control data and the robot second loading operation control data to generate a robot mode control packaging instruction; Step S343: Perform control instruction simulation on the robot mode control encapsulation instruction to generate a robot multi-modal control instruction.

9. A robot operation control system, characterized in that: The robot operation control method according to claim 1 is used to execute the robot operation control system, the robot operation control system comprising: The environmental perception module is used to obtain the spatial data of the aquaculture workboat; deploy a multi-source sensor network based on the spatial data of the aquaculture workboat to generate a multi-source sensor network; integrate the sensor data of the multi-source sensor network to generate environmental perception integrated data of the aquaculture workboat; perform real-time environmental updates on the environmental perception integrated data of the aquaculture workboat to generate real-time environmental monitoring data of the aquaculture workboat; The path planning module is used to synchronize the spatiotemporal data of the real-time environmental monitoring data of the aquaculture workboat and generate the environmental synchronization data of the aquaculture workboat; to construct the environmental model based on the environmental synchronization data of the aquaculture workboat and generate a comprehensive aquaculture environment model; to perform dynamic path planning of the robot loading motion on the spatial data of the aquaculture workboat through the comprehensive aquaculture environment model and generate a path planning diagram of the robot loading motion; An operation control module is used to obtain the robot's feeding reserve data and the robot's initial mechanical arm angle control data; divide the robot's feeding reserve data into reserve modes according to a preset feeding reserve threshold value, and generate a high-reserve feeding mode and a low-reserve feeding mode; perform a path difficulty analysis on the robot's feeding motion path planning diagram, and generate feeding path difficulty data; perform multi-modal feeding operation control on the high-reserve feeding mode and the low-reserve feeding mode based on the feeding path difficulty data, and generate a robot multi-mode control instruction; The control feedback module is used to monitor the robot's feeding and feeding through the robot's multi-mode control instructions, generate the robot's feeding and feeding monitoring data; transmit the robot's feeding and feeding monitoring data to the central control system for feeding operation abnormality detection, so as to execute the robot's normal operation control operation.

10. A robot, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, and is used to execute the operation control method of the robot as claimed in any one of claims 1 to 8.

Images