Data transmission method, robot and storage medium

Abstract

The invention relates to the technical field of data transmission, and discloses a data transmission method, a robot and a storage medium. The method comprises the steps of obtaining first health data; preprocessing the first health data to obtain second health data; the second health data are dynamically marked, the priority of the second health data is determined, third health data are obtained, and the third health data comprise the data type and the data priority; sorting the third health data according to the data priority, and packaging the sorted third health data to obtain fourth health data; performing grading processing on the fourth health data to obtain target health data; and uploading the target health data. In this way, the data processing and reporting efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of data transmission, and in particular to a data transmission method, a robot and a storage medium. Background Art

[0002] Current robot health data reporting technologies are mainly based on traditional monitoring systems, which usually collect certain predefined health indicators, such as voltage, current, and temperature. These systems have the following technical defects: (1) low efficiency of data collection and processing; (2) rigid data reporting strategies and lack of flexibility; (3) limited scalability and compatibility of the system. Summary of the invention

[0003] The implementation methods of the present application mainly solve the technical problem of low efficiency in robot data collection and processing.

[0004] To solve the above technical problems, a technical solution adopted in the implementation mode of the present application is: providing a data transmission method, applied to a robot, the method comprising: obtaining first health data; preprocessing the first health data to obtain second health data; dynamically marking the second health data, and determining the priority of the second health data to obtain third health data, the third health data including a data type and a data priority; sorting the third health data according to the data priority, and encapsulating the sorted third health data to obtain fourth health data; hierarchically processing the fourth health data to obtain target health data; uploading the target health data.

[0005] Optionally, the preprocessing of the first health data to obtain the second health data includes: obtaining the task type of the robot, and determining the operating state of the robot according to the task type; when the operating state is a high-speed moving state, selecting a first filter to process the first health data to obtain the second health data; when the operating state is a load operating state, selecting a second filter to process the first health data to obtain the second health data; when the operating state is a stable or low-speed state, selecting a third filter to process the first health data to obtain the second health data.

[0006] Optionally, the method further includes: acquiring environmental conditions of the robot; and adjusting key parameters of a filter according to the task type and the environmental conditions, the filter including the first filter, the second filter, and the third filter.

[0007] Optionally, the dynamic labeling of the second health data includes: obtaining historical health data of the robot, the task type of the robot, and the environmental conditions of the robot; inputting the historical health data into a preset model for multiple iterative training to obtain a pre-trained model to identify the data type of the historical health data; wherein the historical health data includes voltage, current, and temperature, and the data type includes normal, abnormal, and critical; when the first health data is pre-processed according to the task type and the environmental conditions to obtain the second health data, the second health data is input into the pre-trained model for prediction to obtain a prediction result; and re-labeling the second health data according to the prediction result to update the data type of the second health data.

[0008] Optionally, determining the priority of the second health data to obtain the third health data includes: defining a basic priority for the second health data; obtaining the task type of the robot and performing an importance check on the task type; if the task type is a critical task, increasing the basic priority of the data associated with the critical task in the second health data; if the task type is a non-critical task, reducing the basic priority of the data associated with the non-critical task in the second health data.

[0009] Optionally, the method also includes: obtaining the environmental conditions of the robot and performing risk detection on the environmental conditions; if the environmental conditions are high-risk environments, increasing the basic priority of the data associated with the high-risk environment in the second health data; if the environmental conditions are low-risk environments, reducing the basic priority of the data associated with the low-risk environment in the second health data.

[0010] Optionally, the grading processing of the fourth health data to obtain target health data includes: setting a grading threshold according to the data priority; comparing the priority of the data in the fourth health data with the grading threshold; if the priority of the data in the fourth health data is higher than the grading threshold, determining that the data higher than the grading threshold is the target health data; if the priority of the data in the fourth health data is lower than the grading threshold, determining that the data lower than the grading threshold is non-target health data.

[0011] Optionally, uploading the target health data includes: uploading the target health data to a server, so that the server performs data analysis after receiving the target health data to obtain an analysis result, if the analysis result is normal, a health report is generated according to the analysis result, if the analysis result is abnormal, an alarm is automatically triggered.

[0012] To solve the above technical problems, another technical solution adopted in the embodiment of the present application is: to provide a robot, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described above.

[0013] In order to solve the above technical problems, another technical solution adopted in the implementation mode of the present application is: providing a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a robot, the robot executes the method described above.

[0014] Different from the related art, the present application obtains the first health data; pre-processes the first health data to obtain the second health data; dynamically marks the second health data, and determines the priority of the second health data to obtain the third health data, wherein the third health data includes a data type and a data priority; sorts the third health data according to the data priority, and encapsulates the sorted third health data to obtain the fourth health data; hierarchically processes the fourth health data to obtain the target health data; and uploads the target health data. Based on this, according to the type of task currently performed by the robot and the current environmental conditions, the high-quality second health data obtained after pre-processing is dynamically analyzed and processed to obtain the third health data with the data type and data priority re-marked. Among them, the data type and data priority in the third health data are dynamically modified with the current task type and environmental conditions. Then, the encapsulated fourth health data is hierarchically processed again to filter out the target health data and upload it to the server. In this way, the present application can dynamically mark, modify priority, sort, and grade health data according to the type of task currently performed by the robot and the current environmental conditions, thereby realizing the screening of health data and reporting high-priority health data in real time, reducing the amount of data transmission and improving the efficiency of data processing and reporting. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] One or more embodiments are exemplarily described by corresponding drawings, which do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and the figures in the drawings do not constitute proportional limitations unless otherwise stated.

[0016] Figure 1 This is a schematic diagram of a specific application scenario of a data transmission method provided in an embodiment of the present application;

[0017] Figure 2 is a flow chart of a data transmission method provided in an embodiment of the present application;

[0018] Figure 3 It is a flowchart of a preprocessing process provided by an embodiment of the present application;

[0019] Figure 4 is a flow chart of a dynamic marking process provided by an embodiment of the present application;

[0020] Figure 5 is a flowchart of a priority determination process provided by an embodiment of the present application;

[0021] Figure 6 is a flowchart of a priority determination process provided by another embodiment of the present application;

[0022] Figure 7 A flowchart of a data classification process provided in an embodiment of the present application;

[0023] Figure 8 It is a schematic diagram of the hardware structure of a robot for executing the data transmission method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0025] It should be noted that, if there is no conflict, the various features in the embodiments of the present application can be combined with each other, all within the scope of protection of the present application. Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those generally understood by those skilled in the art of the present application. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not used to limit the present application. The term "and / or" used in this specification includes any and all combinations of one or more related listed items.

[0026] See also Figure 1 , Figure 1 1 is a schematic diagram of a specific application scenario of a data transmission method provided in an embodiment of the present application. Specifically, the application scenario includes an upper computer 10 and a lower computer 20. The upper computer 10 and the lower computer 20 are in communication connection.

[0027] Among them, the upper computer 10 is usually a device for monitoring, controlling and managing the lower computer 20 (robot), and is usually responsible for sending control instructions and processing and storing data sent back from the lower computer 20. Specifically, the upper computer 10 can be a server, a personal computer (PC), an industrial computer (IPC), and a cloud platform. In this embodiment, the upper computer 10 is a server, and its main functions include: command sending, data processing, real-time monitoring, human-computer interaction interface, storage and analysis, etc. Communicate with the lower computer 20 through the network to remotely monitor and control the behavior of the lower computer 20.

[0028] The lower computer 20 is usually a device that directly performs tasks. In this embodiment, the lower computer 20 is a robot, which usually includes an actuator, a sensor, a control system, and a communication module. The lower computer 20 receives instructions sent by the upper computer 10 and performs corresponding operation tasks, while collecting and feeding back data to the upper computer 10.

[0029] Specifically, see 2, Figure 2 1 is a flow chart of a data transmission method provided in an embodiment of the present application. The method is applied to a robot, and the method includes steps S101-S106:

[0030] S101: Acquire first health data.

[0031] In the robot system, the installation and configuration of sensors is an important step to ensure that the system can monitor key parameters (such as voltage, current and temperature) in real time. These parameters are critical to the performance and safety of the robot, so appropriate sensors need to be used and accurately installed and configured. It includes two processes: sensor installation and configuration and data acquisition strategy setting.

[0032] Specifically, for the sensor installation and configuration process:

[0033] Taking the acquisition of health data of voltage, current and temperature as an example, voltage sensors, current sensors and temperature sensors will be equipped for data collection. Among them, voltage sensors usually use voltage dividers, voltage transmitters or Hall effect sensors to measure the voltage inside the robot. Voltage sensors are usually installed in key positions of the power system, such as between the positive and negative poles of the battery pack or in the motor drive circuit. They also need to be properly connected to the data acquisition unit (such as a microcontroller or data acquisition card) to ensure that the sensor output is closely matched with the input of the acquisition unit, and the data line is short and shielded to reduce noise interference. Current sensors can be in the form of shunts, Hall effect sensors or current transmitters. Current sensors are usually installed in key positions of robot motors, battery outputs or power transmission lines, and the direction of the current and the polarity of the sensor need to be considered to ensure the accuracy of the measurement. When connecting, care should be taken to reduce electromagnetic interference, and filtering circuits can be used to optimize signal quality when necessary. Temperature sensors include thermocouples, thermistors (NTC or PTC) and semiconductor temperature sensors. Thermistors are suitable for temperature monitoring in robots due to their high sensitivity and fast response characteristics. Temperature sensors are installed near key components that may generate heat, such as battery packs, motors, or transmission systems. To obtain more accurate temperature data, the sensor should be as close to the heat source as possible, and thermal conductive materials should be used when necessary to enhance the sensor's response speed. In addition, the temperature sensor is usually connected to the data acquisition unit through an analog or digital signal interface.

[0034] The data collection strategy setting process includes the data collection frequency setting strategy and the data storage and caching strategy:

[0035] The data acquisition frequency of each sensor should be set according to the working environment and task requirements of the robot. For example, in a high-load working environment, the acquisition frequency of current and temperature sensors should be relatively high to capture possible overheating or overload conditions in a timely manner. Generally speaking, the sampling frequency of voltage and current sensors can be set above 1kHz, while the sampling frequency of temperature sensors can be relatively low, set between 10Hz and 100Hz according to demand. During operation, the system can dynamically adjust the acquisition frequency of the sensor according to the actual workload. For example, when the robot enters a high-power mode (such as acceleration or heavy-load operation), the data acquisition frequency can be temporarily increased to ensure safe monitoring. When the robot is in standby mode, the frequency can be reduced to reduce the amount of data and power consumption.

[0036] In addition, to optimize system performance, sensor data can be first stored in the robot's local cache and then uploaded to the central processing unit or server regularly for further analysis. The cache strategy should be set according to the timeliness and importance of the data. For example, voltage and current data may need to be uploaded in real time, while temperature data can be uploaded in batches. The storage strategy should take into account the persistence and traceability of the data. Critical safety data (such as overheating alarms or voltage anomalies) should be stored in local storage first and backed up to the remote server regularly. In addition, it is also necessary to consider the limitations of storage space and reasonably arrange the data storage cycle and elimination mechanism to ensure stable operation of the system.

[0037] Through reasonable sensor installation and configuration, as well as scientific data acquisition strategy setting, the robot system can achieve efficient real-time monitoring and safety protection, ensure reliable operation in various working environments, and obtain the first health data in real time. Among them, the first health data can be the voltage, current and temperature data described above, or other important health data, which is not limited in this embodiment. In addition, new sensors can be added to obtain new health data, and this embodiment does not limit the type of sensor.

[0038] S102: Preprocess the first health data to obtain second health data.

[0039] After the robot acquires the first health data through different sensors, it performs a preprocessing operation on the first health data. This embodiment performs the preprocessing operation by providing a state-aware filtering method. State-aware filtering is a method for dynamically selecting the most appropriate filtering algorithm according to the operating state of the robot. Unlike the traditional fixed filtering method, state-aware filtering takes into account the diversity of the robot's operating state and can effectively utilize computing resources while maintaining data integrity and accuracy.

[0040] See also Figure 3 , Figure 3 : is a flow chart of a preprocessing process provided by an embodiment of the present application. Figure 3 As shown, preprocessing the first health data to obtain second health data includes steps S1021-S1024:

[0041] S1021: Obtain the task type of the robot, and determine the operating state of the robot according to the task type.

[0042] The robot automatically obtains the task type of the currently executed task, and distinguishes the current running state of the robot by judging the task type. For example, when the task type is a fast moving task (such as navigation, path tracking or obstacle avoidance, etc.), the current running state of the robot is determined to be a high-speed moving state. For another example, when the task type is to carry goods or perform a specific task, the load will directly affect the output of the sensor, and the current running state of the robot is determined to be a load running state. For another example, when the task type is to perform a fine operation task, such as precise positioning, detailed detection or slow movement in a complex environment, the current running state of the robot is determined to be a stable or low-speed state. It can be understood that the preprocessing method provided in this embodiment can judge the current running state according to the task type currently performed by the robot, and its task type includes but is not limited to the task type described above. It can be understood that in a complex multi-task scenario, the robot may need to combine multiple states, and the switching between different states may also require a transitional filtering strategy. For example, when the robot switches from a high-speed moving state to a load running state, a transitional filtering strategy may be required to smooth data changes to prevent sudden noise or abnormal data from affecting system decisions, which is not limited in this embodiment.

[0043] S1022: When the running state is a high-speed moving state, select a first filter to process the first health data to obtain the second health data.

[0044] When the current operating state is a high-speed moving state, the robot is usually subjected to large mechanical vibrations, and the environmental noise may increase significantly. This will affect the accuracy of the sensor, such as the readings of sensors such as accelerometers and gyroscopes may be interfered with by high-frequency noise. Based on this, the first health data is processed using a first filter to obtain the second health data. For example, the first filter is a recursive least squares filter (RLS). RLS is an adaptive filter that can quickly adjust its weights in an uncertain dynamic environment to suppress noise. It minimizes the sum of squared errors through recursive updates, thereby having a good suppression effect on sudden noise. It can be understood that the RLS filter can respond to environmental changes in real time and quickly adjust the filter parameters, which makes it very suitable for high-speed dynamic changing environments. It can effectively reduce the impact of noise on sensor data without affecting the response speed, ensuring the data accuracy of the robot when it moves at high speed.

[0045] S1023: When the operating state is a load operating state, select a second filter to process the first health data to obtain the second health data.

[0046] When the current operating state is the load operating state, the load will directly affect the output of the sensor. For example, additional weight may cause tilt, thereby affecting the readings of the accelerometer and the inclinometer. In this case, the sensor output data may contain uncertainty caused by the load change. Based on this, the first healthy data is processed using the second filter to obtain the second healthy data. For example, the second filter is a Kalman filter with dynamic adjustment effects. The Kalman filter can dynamically adjust the filter parameters to correct the errors in the sensor data by optimizing the robot's state prediction model and actual observation values. The filter combines the feedback of prediction and actual observation, and can handle dynamic changes with uncertainty in the robot. It can be understood that the Kalman filter can effectively reduce the impact of the load on the sensor data through its prediction and correction mechanism, thereby providing more accurate and reliable sensor readings and ensuring the stability and safety of the robot in the load operating state.

[0047] S1024: When the operating state is stable or in a low-speed state, select a third filter to process the first health data to obtain the second health data.

[0048] When the current operating state is stable or low speed, the robot is usually performing delicate operation tasks, such as precise positioning, detailed detection or slow movement in a complex environment. In this state, the dynamic changes of the sensor are less and the noise level is relatively low. Based on this, the first health data is processed using a third filter to obtain the second health data. For example, the third filter is a moving average filter or an exponential smoothing filter. The moving average filter smoothes fluctuations by averaging data over a period of time, which is suitable for processing low-frequency noise. The exponential smoothing filter achieves a smoothing effect by giving a higher weight to the most recent data. It can be understood that these simple filtering algorithms can provide sufficient smoothing effects while having a low computational burden, and are suitable for noise suppression in low dynamic environments. These algorithms have low computational complexity and are very suitable for resource-limited robot systems, and can ensure the smoothness and consistency of data in a stable working environment.

[0049] It should be noted that state-aware filtering can also be combined with machine learning algorithms to further optimize the selection of filters by continuously learning the robot's operating mode and environmental changes. This intelligent filtering method can significantly improve the adaptability and robustness of the robot system, enabling it to maintain efficient and stable operation in various complex environments.

[0050] In the embodiment of the present application, the running state of the robot (such as speed, acceleration, load, etc.) can be monitored in real time through state-aware filtering, and the appropriate filter can be automatically selected according to the predefined state classification and filtering algorithm strategy. This method can improve the adaptability of the robot in various complex environments, better reduce the impact of noise, effectively avoid the waste of computing resources, and improve the reliability and accuracy of sensor data.

[0051] In some embodiments, the data transmission method of the present invention further includes:

[0052] Obtaining environmental conditions of the robot;

[0053] According to the task type and the environmental conditions, key parameters of filters are adjusted, and the filters include the first filter, the second filter, and the third filter.

[0054] It is understandable that, because robots usually operate in complex and changing environments, the data collected by sensors may be interfered by various noises, affecting the accuracy and stability of the robot system. This embodiment provides an adaptive filtering method that allows the parameters of the filter to be dynamically adjusted according to real-time needs and conditions. Through adaptive filtering, the robot can respond to these challenges more flexibly, thereby improving overall performance.

[0055] Specifically, the key parameters of the filter are adjusted according to the task type and environmental conditions.

[0056] It can be understood that environmental noise refers to any non-target signal in the external environment that may interfere with the sensor signal. These noise sources may come from surrounding electronic equipment, mechanical vibrations, temperature fluctuations, and even light changes. Under different environmental conditions, the noise level may vary greatly, so fixed filtering parameters are difficult to provide the best filtering effect in all scenarios. Based on this, a key function of the adaptive filtering method is to automatically adjust the parameters of the filter when the noise level changes are sensed. Specifically, when the sensor detects a high level of environmental noise, the filter can increase its filtering strength, which is usually achieved by increasing the size of the filter window or adjusting the filter coefficient. The larger the filter window, the more historical data will be used to smooth the current signal, which helps to suppress short-term noise fluctuations. However, when the noise level is low and the environment is relatively stable, too strong filtering may cause a lag in signal response. At this time, the filter can reduce the filtering strength, so that the robot system can respond to changes in sensor input more quickly and provide more accurate and timely feedback.

[0057] It is understandable that the type of robot task and the urgency of the current task will also significantly affect the requirements for filter parameters. Different tasks have different requirements for data accuracy and response speed. For example, when performing high-precision measurement tasks, any noise interference will significantly affect the reliability of the results, so a higher filtering strength is required to ensure the stability and accuracy of the data. Based on this, in some critical tasks, such as when the robot performs precision assembly or navigation operations, the system may need a faster response speed to process real-time data. At this time, the adaptive filter method can reduce the size of the filter window or adjust the filter coefficient so that the system can react faster. Although this may increase the impact of noise, this trade-off is necessary for urgent tasks or high-risk scenarios. In addition, in addition to single-dimensional adjustments, adaptive filters can also comprehensively consider the dual factors of environment and task. For example, when performing critical tasks in a noisy environment, the filter can dynamically balance the filtering strength and response speed to suppress noise while maintaining the sensitivity of the system as much as possible.

[0058] It should be noted that adaptive filtering methods can usually be implemented through a variety of algorithms, the most common of which are based on adaptive algorithms such as LMS (Least Mean Squares) and RLS (Recursive Least Squares). These algorithms can analyze the difference between the input signal and the target signal in real time, and adjust the weight or coefficient of the filter to make the output signal closer to the desired result. In practical applications, the design of adaptive filters needs to take into account the consumption of computing resources, real-time requirements, and system complexity. For example, for embedded systems with limited computing power, it may be necessary to balance the filtering effect and algorithm complexity during design.

[0059] By flexibly adjusting the parameters of the filter, the adaptive filtering in this embodiment can help the robot system to better perform tasks in a complex and changing environment, which is beneficial to improving the noise reduction processing effect of the filter, and improving the reliability of the data and the overall performance of the system.

[0060] S103: Dynamically mark the second health data and determine the priority of the second health data to obtain third health data, where the third health data includes a data type and a data priority.

[0061] See also Figure 4 , Figure 4 This is a flow chart of a dynamic marking process provided by an embodiment of the present application. Figure 4 As shown, dynamically marking the second health data includes steps S1031-S1034:

[0062] S1031: Acquire historical health data of the robot, the task type of the robot, and the environmental conditions of the robot.

[0063] First, obtain the historical health data of the robot. It can be understood as sensor data collected by the robot during its past operation, which reflects the status of the robot and its changes over time. Key data includes: Voltage: Monitors the status of the robot's power supply system, such as battery voltage, power supply voltage, etc. Current: Monitors the current consumption of various parts of the robot to help identify potential overload or short circuit conditions. Temperature: Monitors the temperature of various parts of the robot to prevent overheating or insufficient cooling. This data may be collected at different time periods, tasks, and environmental conditions, and stored in a database for subsequent analysis.

[0064] Secondly, the task type of the robot is obtained. It can be understood that the specific description of the task type can be found in the description of the task type in the above step S1021.

[0065] Finally, the environmental conditions of the robot are obtained. It can be understood that the specific description of the environmental conditions can be found in the description of the environmental conditions in the above step S102.

[0066] S1032: Input the historical health data into a preset model for multiple iterative training to obtain a pre-trained model to identify the data type of the historical health data; wherein the historical health data includes voltage, current, and temperature, and the data type includes normal, abnormal, and critical.

[0067] It can be understood that a preset model refers to a machine learning model with a defined structure, such as a deep neural network, support vector machine, random forest, etc. The collected historical health data is input into this model for training, with the goal of allowing the model to learn how to distinguish and identify different data types. Iterative training means that the model continuously optimizes its predictive ability by updating weights or parameters multiple times. In each iteration, the model analyzes the input data and adjusts internal parameters (such as weights) to better predict the data type. Through multiple iterations, the model gradually improves its ability to identify normal, abnormal, and critical states.

[0068] S1033: When the first health data is preprocessed according to the task type and the environmental conditions to obtain the second health data, the second health data is input into the pre-trained model for prediction to obtain a prediction result.

[0069] First, according to the task type and the environmental conditions, the first health data is preprocessed to obtain the second health data. It can be understood that the specific preprocessing operations are described in the above step S102. It should be noted that on the basis of the above preprocessing operations, other preprocessing operations can be added to process the first health data to obtain the second health data. For example, the data is converted into a standard range through a normalization operation so that the model can process data of different magnitudes. For another example, useful features are extracted from the raw data through feature extraction operations, such as the current change rate, temperature increment, etc. These features can better represent the state of the system.

[0070] Secondly, the second health data is input into the pre-trained model for prediction to obtain a prediction result. It can be understood that the preprocessed data is called the second health data. These data are input into the previously trained model, and the model will predict the data based on what it has learned during the training process. The goal of the prediction is to determine whether the data belongs to the "normal", "abnormal" or "critical" type. For example, normal: the data reflects the state of the robot in normal operation without obvious abnormalities. For another example, abnormal: the data shows certain unusual characteristics, which may indicate potential problems, such as sensor failure, minor mechanical problems, etc. For another example, critical: the data indicates that the robot is at serious risk and immediate measures need to be taken, such as equipment overheating, short circuit, etc.

[0071] S1034: Relabel the second health data according to the prediction result to update the data type of the second health data.

[0072] It can be understood that the second health data will be relabeled according to the prediction results of the model. If a data point is predicted to be abnormal or critical, it will be marked as the corresponding type. This process can help identify data points that need special attention and take appropriate measures. After the label is updated, the type information of these data will be stored or updated in the database. This process not only improves the accuracy of data management, but also provides richer historical data support for future analysis.

[0073] In addition, the dynamic labeling process also includes feedback mechanisms and model optimization, as well as continuous data accumulation and learning. It can be understood that the robot will monitor and feedback these prediction results in real time, and compare the actual situation with the prediction. By comparison, if the model prediction is found to be wrong or inaccurate, the robot can be retrained based on the new data to gradually optimize the model's predictive performance. And over time, the robot will accumulate more and more labeled data, which will be further used to optimize the model and improve its adaptability in diverse tasks and environments. Through continuous feedback and learning, the robot will become more intelligent and able to handle more complex scenarios.

[0074] The embodiment of the present application can accurately identify the state of health data through an algorithm model. If data in an abnormal state appears, the label of the data is dynamically adjusted. For example, the model is first trained based on historical health data so that the model learns how to identify the labels (normal, abnormal, critical) corresponding to the health data. According to different task types and environmental conditions, new health data is obtained. At this time, this is predicted by the trained model to find abnormal data and re-label it. In this way, it is possible to dynamically label data according to the task type and environmental conditions of the robot, which is beneficial to the subsequent data reporting process and improves the flexibility of the data reporting process.

[0075] In some embodiments, the priority of the second health data is determined based on the type of task the robot is currently performing, so that the most important data can be processed and focused on first in critical tasks. Figure 5 , Figure 5 : is a flowchart of a priority determination process provided by an embodiment of the present application. Figure 5 As shown, determining the priority of the second health data to obtain the third health data includes steps S1035-S1038:

[0076] S1035: Define a basic priority for the second health data.

[0077] The base priority is the initial level of importance assigned to each type of sensor data without considering the current task type and environment. This base priority can be defined based on several factors: Type of data: For example, voltage, current, and temperature data may each represent different priorities because they have different impacts on the robot's operation. Historical performance of the data: If a certain type of data has been frequently associated with failures in past analyses, its base priority may be set higher. Frequency of data collection: Data collected more frequently generally have a higher priority because they provide more timely status information.

[0078] By defining basic priorities, the robot can initially filter out which data should receive more attention and which can be temporarily ignored.

[0079] S1036: Obtain the task type of the robot, and perform importance detection on the task type.

[0080] Task type refers to the operation task that the robot is currently performing. Task types can be divided into critical tasks and non-critical tasks, and each type may have different requirements for the robot. Critical tasks may be operations that need to be performed accurately, such as welding, precision assembly, handling emergencies, etc.; while non-critical tasks may be some daily or low-risk operations, such as patrolling, cleaning, etc. The robot needs to be able to identify and classify these task types because they will directly affect the data priority adjustment in subsequent steps.

[0081] Importance detection refers to the robot's judgment of the criticality of the current task. If the task is classified as a critical task, the system needs to treat it with a higher priority; if the task is classified as a non-critical task, it will be treated with a relatively lower priority. This detection can be based on predefined task classification rules, or through machine learning models to automatically judge the importance of the task based on historical data and task results.

[0082] S1037: If the task type is a critical task, increase the basic priority of the data associated with the critical task in the second health data.

[0083] Critical tasks require accurate and timely data support, so data associated with these tasks should be given higher priority. Specific operations include:

[0084] Prioritize critical data: For example, in welding tasks, temperature sensor data may become very critical because the welding temperature must be precisely controlled. In this case, the priority of temperature data will be increased.

[0085] Resource allocation: The system may allocate more computing resources or network bandwidth to process these high-priority data, ensuring that they are analyzed more quickly and accurately.

[0086] Real-time monitoring: High-priority data may be incorporated into more frequent monitoring and analysis cycles to ensure that any anomalies in the mission execution process can be detected immediately.

[0087] S1038: If the task type is a non-critical task, reduce the basic priority of the data associated with the non-critical task in the second health data.

[0088] Non-critical tasks typically do not require extreme accuracy or timeliness, so the data associated with these tasks can be given a lower priority. This has several benefits:

[0089] Save resources: Lowering the priority of this data can reduce the use of computing resources and bandwidth, and use more resources for mission-critical data processing.

[0090] Reduced processing burden: By reducing priority, the system can reduce the frequency and depth of processing of non-critical data, examining or analyzing it only when necessary.

[0091] Delayed processing: For low-priority data, the system can choose to delay processing, or process the data only after other tasks are completed.

[0092] The third health data is the health data set after adjusting the priority through the above steps S1031-S1038. It represents the key data set that the system needs to process and monitor with priority under the current task and environmental conditions. These data will become the focus of the robot's real-time operation and health management.

[0093] It should be noted that this priority adjustment is not a one-time thing, but a dynamic process. As tasks change or the environment changes, the system may need to continuously monitor and adjust the priority of health data. This ensures that the system is always in the best operating state. In addition, by focusing on high-priority data, the system can detect potential failures earlier. For example, when performing critical tasks, if a sensor data is abnormal, the system can immediately issue an alarm and take measures to prevent the failure from affecting the completion of the task. Finally, by reasonably adjusting the data priority, the system can optimize resource utilization, reduce unnecessary data processing burden, and ensure that the data of critical tasks is fully processed. This is especially important in complex multi-tasking environments and helps improve the overall work efficiency of the robot.

[0094] In some embodiments, see Figure 6 , Figure 6 FIG. 1 is a flowchart of a priority determination process provided by another embodiment of the present application. Figure 6 As shown, the data transmission method of the present invention further includes steps S1039-S10311:

[0095] S1039: Obtain environmental conditions of the robot, and perform risk detection on the environmental conditions.

[0096] The various conditions of the robot's environment are monitored in real time through sensors on the robot (such as ambient temperature sensors, ambient humidity sensors, ambient air pressure sensors, ambient light sensors, etc.). Environmental conditions are classified into high-risk environments and low-risk environments. Among them, high-risk environments: such as high temperature, low temperature, high humidity, high radiation, strong magnetic field, dangerous chemicals, etc. These environments may have a significant impact on the operation of the robot, and even endanger the normal operation or life of the robot. Low-risk environment: such as mild temperature, normal humidity, no harmful substances, etc., the robot can work normally in these environments, and the external environment has little impact on the robot. It can be understood that machine learning or rule algorithms are used to analyze environmental condition data to evaluate the risk level of the current environment. Among them, a comprehensive judgment can be made based on historical data, empirical rules or real-time data.

[0097] S10310: If the environmental condition is a high-risk environment, increase the basic priority of the data associated with the high-risk environment in the second health data.

[0098] The priority of the secondary health data associated with high-risk environments will be increased. For example, if the ambient temperature is very high, the priority of the temperature-related sensor data will be increased to ensure that the robot can obtain timely feedback on temperature changes and take necessary countermeasures. In addition, if the current robot is performing a critical task (such as an emergency rescue mission), the priority of the data related to the task needs to be further increased to ensure the reliable execution of the task.

[0099] S10311: If the environmental condition is a low-risk environment, reduce the basic priority of the data associated with the low-risk environment in the second health data.

[0100] In low-risk environments, the priority of the second health data related to the environmental conditions can be appropriately reduced. For example, in a mild environment, the importance of temperature data is relatively reduced, and its weight in the overall data priority can be reduced. In addition, if the robot is performing a non-critical task, the priority of the data related to the task can be further reduced to save computing resources and energy.

[0101] The embodiment of the present application determines the final third health data priority by taking into account all environmental conditions, task types and other factors. Data with higher priorities will be processed and responded to first to ensure the safe and efficient operation of the robot in various environments. In addition, the priority of the data is dynamically adjusted according to the real-time changes in environmental conditions. For example, when the robot enters a new environment or the environmental conditions change significantly, the priority needs to be re-evaluated and adjusted. It should be noted that, according to the operation of the system after the priority adjustment, feedback data is collected, the adjustment effect is evaluated, and the priority evaluation model and algorithm are optimized, and through the continuous training and optimization of the machine learning model, the system can better adapt to various environmental conditions and improve the accuracy and timeliness of priority decisions. It can be understood that through these steps, the robot can flexibly adjust the priority of its data processing in different environments to ensure safety in high-risk environments and maximize efficiency in low-risk environments.

[0102] S104: Sort the third health data according to the data priority, and encapsulate the sorted third health data to obtain fourth health data.

[0103] Specifically, encapsulating data is a key step to ensure information structuring and efficient transmission during data upload. This embodiment adopts the TLV (Type-Length-Value) format and encapsulates data according to the RDUC (Robot Data Upload Convention) protocol. The advantages of its TLV format are: Type: specifies the category of data, such as temperature, position, speed, etc., so that the receiving end can quickly identify the data content. Length: specifies the length of the value field to ensure the consistency and accuracy of data parsing. Value: The actual sensor data or calculation results are encapsulated in a field of specified length to maintain data integrity.

[0104] Secondly, the encapsulation process is specifically as follows: Data selection and sorting: Sort the data according to the data priority. Data conversion and formatting: Convert the original data or processed data into a format suitable for uploading, for example, the value may need to be converted from floating point to integer or fixed format string. Encapsulation: Encapsulate the converted data in TLV structure according to the RDUC protocol to obtain the fourth health data.

[0105] S105: Perform hierarchical processing on the fourth health data to obtain target health data.

[0106] See also Figure 7 , Figure 7 A flowchart of a data classification process provided in an embodiment of the present application, wherein the fourth health data is classified to obtain target health data, specifically including the following steps S1051-S1054:

[0107] S1051: Setting a classification threshold according to the data priority.

[0108] First, it is necessary to clarify the priority range of health data. Priority can usually be expressed by numerical values ​​(such as 0 to 100), levels (such as low, medium, high), or other quantitative methods. Priority setting can be based on factors such as the urgency of the task, environmental risks, and the importance of the data.

[0109] Determine threshold criteria: Static threshold: Based on a pre-set fixed threshold. For example, data with a priority of 70 or above is considered high priority data, and data below 70 is considered low priority data. Dynamic threshold: Dynamically adjust the threshold based on real-time conditions (such as environmental changes, task changes). For example, in a high-risk environment, the threshold can be lowered to increase the amount of data considered high priority.

[0110] S1052: Compare the priority of the data in the fourth health data with the classification threshold.

[0111] A priority calculation is performed on each item of the fourth health data, and the priority of each item of data is compared with a set grading threshold to determine whether the data is a high priority or a low priority.

[0112] S1053: If the priority of the data in the fourth health data is higher than the classification threshold, determine that the data higher than the classification threshold is the target health data.

[0113] If the priority of a certain data in the fourth health data is higher than the classification threshold, the data is regarded as target health data. Such data is critical to the current operation of the robot and may require priority processing, analysis, or response. Target health data usually requires a higher processing priority and may trigger corresponding system responses, such as alarms, adjustment of operation strategies, or preventive measures.

[0114] S1054: If the priority of the data in the fourth health data is lower than the classification threshold, determine that the data below the classification threshold is non-target health data.

[0115] If the priority of a certain data in the fourth health data is lower than the classification threshold, the data is considered as non-target health data. Non-target health data is less important in the current situation and may be delayed or ignored. By identifying non-target health data, the system can optimize resource allocation and focus more computing power, energy, and bandwidth on processing target health data.

[0116] It should be noted that the robot can dynamically adjust the classification threshold according to the external environment, task changes and system feedback. For example, when the robot enters a more dangerous environment, the threshold can be appropriately lowered to more sensitively capture and process more health data. In addition, the priority calculation model and classification threshold setting can be continuously optimized based on actual operation conditions and feedback data to improve the system's resilience and efficiency.

[0117] In the embodiment of the present application, by setting priorities and grading thresholds, the robot can effectively distinguish which health data is most critical in the current situation and which data can be temporarily ignored or delayed. This mechanism helps to improve the real-time and responsiveness of the system, ensuring that resources can be concentrated to process the most important data at critical moments, ensuring the safety of the robot and the smooth execution of tasks.

[0118] S106: Upload the target health data.

[0119] The uploading of the target health data includes:

[0120] Upload the target health data to a host computer (e.g., a server) so that the server receives the target health data and performs data analysis to obtain an analysis result. If the analysis result is normal, a health report is generated based on the analysis result. If the analysis result is abnormal, an alarm is automatically triggered.

[0121] Among them, the embodiments of the present application also provide a network monitoring mechanism, a data upload strategy, and exception handling and feedback operations to ensure the stable operation of the data upload process and the timely transmission of data.

[0122] Specifically, the network monitoring mechanism includes: real-time monitoring: deploying network monitoring tools to collect key network indicators such as network bandwidth, latency and packet loss rate in real time; data response: automatically adjusting data upload strategies when network parameters change to ensure optimal data transmission.

[0123] Data upload strategies include: Priority management: When network conditions are good, the data packet size can be increased to increase the frequency of data transmission; when network conditions are poor, the data packet size can be reduced to reduce the frequency of transmission, but ensure that the target health data is sent first. Miniaturization of target health data packets: Optimize the encapsulation of target health data and reduce the size of target health data packets, so that it can be uploaded quickly and reliably even when network conditions are poor.

[0124] The exception handling and feedback mechanism includes: Exception monitoring: Real-time detection of abnormal situations during data upload, such as consecutive failed upload attempts, unexpected high-latency responses, etc. Automatic retry mechanism: Automatically retry unsuccessfully uploaded data packets, and the retry strategy is adjusted according to the network status and data importance.

[0125] Secondly, on the server side, the server receives the data packet and parses the TLV format, extracts the data from each sensor, and performs data analysis and decision making on the data. It can be understood that by applying data analysis algorithms or machine learning models, data trends are analyzed and potential problems are identified. A health report is generated based on the analysis results. If a major problem is detected, an alarm is automatically triggered and the maintenance team is notified.

[0126] Finally, the server side also includes the user interface and interactive processes, including monitoring interface display, maintenance and control response. It can be understood that the health status of the robot is displayed in real time through the monitoring interface, including the current value of each indicator, historical trends and any alarms or notifications. In the user operation interface, technicians are allowed to remotely perform maintenance operations or adjust configurations based on the robot's health reports and alarms.

[0127] The embodiment of the present application obtains the first health data; pre-processes the first health data to obtain the second health data; dynamically marks the second health data, and determines the priority of the second health data to obtain the third health data, wherein the third health data includes a data type and a data priority; sorts the third health data according to the data priority, and encapsulates the sorted third health data to obtain the fourth health data; hierarchically processes the fourth health data to obtain the target health data; and uploads the target health data. In this way, the embodiment of the present application can dynamically mark, modify the priority, sort and hierarchically operate the health data according to the task type currently performed by the robot and the current environmental conditions, thereby realizing the screening of health data, so as to report the health data with high priority in real time, reduce the amount of data transmission, and improve the efficiency of data processing and reporting. Since the present application can dynamically adjust the reporting of the target health data according to different task types and environmental conditions during the data reporting process, the flexibility of the data reporting process is improved. In addition, when a new sensor is added to collect new health data, the method provided by the present application can also adaptively and dynamically process the new health data, thereby improving the scalability and compatibility of the data reporting process.

[0128] The present application also provides a robot, see Figure 8 , which shows a schematic diagram of the hardware structure of a robot capable of executing the method described in the above embodiment. The robot 300 includes: at least one processor 310; and a memory 320 in communication with the at least one processor 310, Figure 8In the embodiment, a processor 310 is taken as an example. The memory 320 stores instructions that can be executed by the at least one processor 310. The instructions are executed by the at least one processor 310 so that the at least one processor 310 can execute the data transmission method described in the above embodiment. The processor 310 and the memory 320 can be connected via a bus or other means. Figure 8 The example of connecting through bus is taken in the following.

[0129] The memory 320 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the data transmission method in the embodiment of the present application. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, implementing the data transmission method described in the above embodiment.

[0130] The memory 320 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created according to the use of the computing device, etc. In addition, the memory 320 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 320 may optionally include a memory remotely arranged relative to the processor 310, and these remote memories may be connected to the computing device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0131] The one or more modules are stored in the memory 320, and when executed by the one or more processors 310, the data transmission method described in the above embodiment is executed.

[0132] The above product can execute the method provided in the embodiment of the present application, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the data transmission method described in the above embodiment of the present application.

[0133] The embodiment of the present application also provides a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more processors so that the at least one processor can execute the data transmission method described in the above embodiment. For example, the non-volatile computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CDROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0134] It should be noted that the above described embodiments are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0135] Through the description of the above implementation methods, ordinary technicians in this field can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, it can also be implemented by hardware. Ordinary technicians in this field can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Under the concept of the present application, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes in different aspects of the present application as described above, which are not provided in detail for the sake of simplicity. Although the present application has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A data transmission method, applied to a robot, characterized in that: The method comprises: Get First Health data; Preprocessing the first health data to obtain second health data; Dynamically marking the second health data and determining the priority of the second health data to obtain third health data, wherein the third health data includes a data type and a data priority; sorting the third health data according to the data priority, and encapsulating the sorted third health data to obtain fourth health data; performing hierarchical processing on the fourth health data to obtain target health data; Upload the target health data.

2. The data transmission method according to claim 1, characterized in that: The preprocessing of the first health data to obtain second health data includes: Acquire the task type of the robot, and determine the operating state of the robot according to the task type; When the running state is a high-speed moving state, selecting a first filter to process the first health data to obtain the second health data; When the operating state is a load operating state, selecting a second filter to process the first health data to obtain the second health data; When the operating state is a stable or low-speed state, a third filter is selected to process the first health data to obtain the second health data.

3. The data transmission method according to claim 2, characterized in that: The method further comprises: Obtaining environmental conditions of the robot; According to the task type and the environmental conditions, key parameters of filters are adjusted, and the filters include the first filter, the second filter, and the third filter.

4. The data transmission method according to claim 1, characterized in that: The dynamically marking the second health data includes: Obtaining historical health data of the robot, the task type of the robot, and the environmental conditions of the robot; Input the historical health data into a preset model for multiple iterations of training to obtain a pre-trained model to identify the data type of the historical health data; wherein the historical health data includes voltage, current, and temperature, and the data type includes normal, abnormal, and critical; When the first health data is preprocessed according to the task type and the environmental conditions to obtain the second health data, the second health data is input into the pre-trained model for prediction to obtain a prediction result; The second health data is relabeled according to the prediction result to update the data type of the second health data.

5. The data transmission method according to claim 1, characterized in that: The determining the priority of the second health data to obtain the third health data includes: defining a basic priority level for the second health data; Obtaining the task type of the robot, and performing importance detection on the task type; If the task type is a critical task, increasing the basic priority of the data associated with the critical task in the second health data; If the task type is a non-critical task, the basic priority of the data associated with the non-critical task in the second health data is reduced.

6. The data transmission method according to claim 5, characterized in that: The method further comprises: Acquiring environmental conditions of the robot and performing risk detection on the environmental conditions; If the environmental condition is a high-risk environment, increasing the basic priority of the data associated with the high-risk environment in the second health data; If the environmental condition is a low-risk environment, the basic priority of data associated with the low-risk environment in the second health data is reduced.

7. The data transmission method according to claim 1, characterized in that: The step of performing hierarchical processing on the fourth health data to obtain target health data includes: Setting a classification threshold according to the data priority; comparing the priority of the data in the fourth health data with the classification threshold; If the priority of the data in the fourth health data is higher than the classification threshold, determining that the data higher than the classification threshold is the target health data; If the priority of the data in the fourth health data is lower than the classification threshold, it is determined that the data below the classification threshold is non-target health data.

8. The data transmission method according to claim 1, characterized in that: The uploading of the target health data includes: The target health data is uploaded to the server so that the server performs data analysis after receiving the target health data to obtain the analysis result. If the analysis result is normal, a health report is generated according to the analysis result. If the analysis result is abnormal, an alarm is automatically triggered.

9. A robot, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

10. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by a robot, enable the robot to perform the method according to any one of claims 1 to 8.

Images