Integrated control system based on intelligent robot

Abstract

The invention relates to the technical field of intelligent robot control, and discloses an integrated control system based on an intelligent robot, which comprises the steps of deploying multi-dimensional sensing equipment on a mechanical part and an electrical system, and integrating mechanical part operation parameters, electrical system operation parameters and environment operation parameters. Extracting features to calculate a time domain feature coefficient and a frequency domain feature coefficient, connecting a robot management system to obtain historical health data, calculating a historical time domain feature coefficient and a historical frequency domain feature coefficient, fusing a plurality of calculation results of the machine data arrangement module and the reference construction module, and constructing a comprehensive model of the robot health coefficient; and outputting the health state prediction condition of the robot. According to the system, real-time sensing and prediction of the health state of a user are realized, out-of-plan shutdown and secondary damage are reduced, and the robustness and the adaptive capacity of the system during performance degradation of components are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent robot control technology, specifically to an integrated control system based on intelligent robots. Background Technology

[0002] Intelligent robots are high-tech devices that integrate advanced sensors, artificial intelligence algorithms, and mechanical engineering. They are capable of making autonomous decisions by perceiving their environment and performing a range of complex tasks. These robots not only possess human-like interfaces, such as voice recognition and facial expression simulation, but they can also learn new skills and adapt to different working environments. For example, in manufacturing, intelligent robots can perform precision assembly; in the medical field, they can assist in surgery or care for patients; and in the home, they can serve as cleaning assistants or provide companionship. With continuous technological advancements, intelligent robots are becoming increasingly sophisticated, and their application areas are constantly expanding, foreshadowing profound changes in future human society.

[0003] Existing robot control systems have significant defects and shortcomings. They lack the ability to perceive, predict, and adapt to their own health status. Current systems rely heavily on simple threshold alarms to monitor hardware status, which makes it impossible to capture early signals of progressive physical wear and tear on hardware. This leads to a passive situation of "ignoring minor problems and stopping abruptly when major problems occur." This delayed response can easily cause unplanned downtime and secondary damage, and further exposes the system's lack of robustness when facing component performance degradation. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an integrated control system based on intelligent robots. This system enables real-time perception and prediction of the robot's own health status. Furthermore, through a hierarchical response mechanism, it avoids the limitation that small robot states are not easily detected and may evolve into larger failures. This reduces unplanned downtime and secondary damage, and significantly improves the system's robustness and adaptability when component performance deteriorates.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an integrated control system based on an intelligent robot, comprising a machine data collection module, a machine data processing module, a benchmark construction module, a health assessment module, and a lifespan prediction module; Machine data collection module: used to deploy multi-dimensional sensing devices in mechanical components and electrical systems, and integrate the operating parameters of mechanical components, electrical systems, and environmental operating parameters; Machine data processing module: Based on the data acquired by the machine data collection module, extracts features and calculates time-domain feature coefficients. and frequency domain characteristic coefficients ; The baseline construction module is used to connect to the robot management system to obtain historical health data and calculate historical time-domain feature coefficients based on the historical health data. and historical frequency domain characteristic coefficients ; Health assessment module: used to integrate the calculation results of the machine data processing module and the benchmark construction module to build a comprehensive model of the robot's health coefficient; Lifetime prediction module: Used to output the robot's health status prediction based on the comprehensive model of robot health coefficient.

[0006] Preferably, the machine data collection module deploys vibration sensors, torque sensors, and displacement sensors at the reducer, bearings, and joints of the mechanical components to collect vibration spectrum, instantaneous torque, and gap change data of the mechanical components and integrate the operating parameters of the mechanical components. The machine data collection module also collects motor stator current and driver output voltage through current sensors and voltage monitoring modules, and integrates electrical system operating parameters by monitoring winding temperature and circuit board temperature in conjunction with temperature sensors. The machine data collection module also integrates visual sensors, sound sensors, GPS positioning systems, and IMU inertial measurement units to acquire data on abnormal appearance, abnormal noise, operational changes, and load changes of the robot hardware, as well as environmental operating parameters.

[0007] Preferably, the time-domain characteristic coefficients The calculation formula is: ; In the formula, Represents the time-domain characteristic coefficients. This represents the total number of data points collected by a single sensor within a data window. Represents the first in the data window One data point; Represents all data within the data window The average value.

[0008] Preferably, the frequency domain characteristic coefficients The calculation formula is: ; In the formula, Represents frequency domain characteristic coefficients, This represents the total number of frequency components in the spectrum obtained after the time-domain signal from a single sensor has undergone FFT transformation. The first in the representative spectrum The center frequency corresponding to each component The first in the representative spectrum Frequency components The corresponding amplitude.

[0009] Preferably, the historical time-domain feature coefficients The calculation formula is: ; In the formula, Represents historical time-domain characteristic coefficients. This represents the lower limit of historical time-domain characteristics. This represents the upper limit of historical time-domain characteristics.

[0010] Preferably, the lower limit of the historical time domain features The calculation formula is: ; The upper limit of historical time domain features The calculation formula is: ; In the formula, Representing the The time-domain characteristic coefficients calculated from each historical time window Represents the total number of historical windows. Represents a constant; Representing all history The average value; Representing history The standard deviation.

[0011] Preferably, historical frequency domain characteristic coefficients The calculation formula is: ; In the formula, Represents historical frequency domain characteristic coefficients. It represents the lower limit of historical frequency domain characteristics. This represents the upper limit of historical frequency domain characteristics.

[0012] Preferably, the lower limit of the historical frequency domain features The calculation formula is: ; The upper limit of the historical frequency domain features The calculation formula is: ; In the formula, Represents historical frequency domain characteristic coefficients. Representing the Frequency domain characteristic coefficients calculated from historical time windows Represents the total number of historical windows. Represents a constant; Representing history The arithmetic mean; Representing history The standard deviation.

[0013] Preferably, the expression for the comprehensive model of the robot's health coefficient is: ; In the expression, The result represents the calculation of the robot's health coefficient.

[0014] Preferably, when the robot's health coefficient A value of 1.5 or higher indicates that the robot is in excellent health and operating normally. When the robot's health coefficient is greater than or equal to 1 but less than 1.5, it indicates that the robot is experiencing performance fluctuations and has early signs of failure. When the robot's health coefficient is less than 1.5, it indicates that the robot's health has deteriorated and there are clear signs of failure. Troubleshooting and maintenance of the robot are necessary.

[0015] Compared with the prior art, the present invention provides an integrated control system based on intelligent robots, which has the following beneficial effects: This invention first deploys multi-dimensional sensors in mechanical components and electrical systems and synchronizes timestamps through a machine data acquisition module, breaking through the limitations of single-parameter monitoring and comprehensively capturing multi-dimensional signals such as vibration, current, and temperature, providing a data foundation for sensing progressive physical losses. Then, a machine data processing module calculates time-domain characteristic coefficients. and frequency domain characteristic coefficients The raw data is transformed into a quantitative indicator sensitive to early wear and tear, replacing a simple threshold. This allows it to capture subtle performance degradation signals. Furthermore, historical time-domain characteristic coefficients are calculated using a benchmark building module that combines historical health data covering multiple operating conditions. and historical frequency domain characteristic coefficients A dynamic benchmark range was constructed to avoid the limitations of fixed thresholds, making health assessments more aligned with actual operating scenarios. Ultimately, a comprehensive assessment was achieved by integrating robot health coefficients based on time and frequency domain deviations through the health assessment module. The lifespan prediction module then implemented a graded response based on the robot health coefficients, upgrading the alarm from exceeding the standard to a progressive warning, identifying early signs of failure in advance, and providing a clear basis for maintenance decisions. This not only enabled real-time perception and prediction of the robot's own health status but also avoided the limitation of small robot states being difficult to detect and evolving into larger failures through a graded response mechanism. This reduced unplanned downtime and secondary damage, significantly improving the system's robustness and adaptability in the face of component performance degradation. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 An integrated control system based on intelligent robots includes a machine data collection module, a machine data processing module, a benchmark construction module, a health assessment module, and a lifespan prediction module. Machine data collection module: used to deploy multi-dimensional sensing devices in mechanical components and electrical systems, and integrate the operating parameters of mechanical components, electrical systems, and environmental operating parameters; The machine data collection module deploys vibration sensors, torque sensors, and displacement sensors at the reducers, bearings, and joints of mechanical components to collect vibration spectrum, instantaneous torque, and gap change data of mechanical components and integrate the operating parameters of mechanical components. The machine data collection module also collects motor stator current and driver output voltage through current sensors and voltage monitoring modules, and integrates electrical system operating parameters by monitoring winding temperature and circuit board temperature in conjunction with temperature sensors. The machine data collection module also integrates visual sensors, sound sensors, GPS positioning systems, and IMU inertial measurement units to acquire data on abnormal appearance, abnormal noise, operational changes, and load changes of the robot hardware, as well as environmental operating parameters. By deploying multi-dimensional sensors in mechanical components, electrical systems, and the environment, the limitations of traditional single-parameter monitoring are overcome, covering multi-dimensional early signals of progressive hardware loss. In the machine data collection module, each sensor uses the robot's main control high-precision real-time clock to generate and carry a Unix timestamp, thereby aligning the time dimension of each sensor and avoiding feature analysis errors caused by time deviation. This lays the foundation for subsequent multimodal feature fusion. The real-time clock needs to be calibrated regularly to avoid time drift caused by long-term operation. At the same time, the timestamp accuracy should match the data acquisition frequency to maintain data consistency. Machine data processing module: Based on the data acquired by the machine data collection module, extracts features and calculates time-domain feature coefficients. and frequency domain characteristic coefficients ; Time-domain characteristic coefficients The calculation formula is: ; In the formula, Represents the time-domain characteristic coefficients. This represents the total number of data points collected by a single sensor within a data window. Represents the first in the data window One data point; Represents all data within the data window The average value; This represents the difference between each data point and its mean, reflecting the deviation of the data point from its mean. The time-domain characteristic coefficients are based on the ratio of the fourth moment to the second moment, which can sensitively capture the impact anomalies in the signal and accurately reflect the early wear of mechanical parts. Frequency domain characteristic coefficients The calculation formula is: ; In the formula, Represents frequency domain characteristic coefficients, This represents the total number of frequency components in the spectrum obtained after the time-domain signal from a single sensor has undergone FFT transformation. The first in the representative spectrum The center frequency corresponding to each component The first in the representative spectrum Frequency components The corresponding amplitude; Frequency domain characteristic coefficients characterize the distribution of signal energy in the frequency domain by comparing frequency components with amplitude, and can identify the performance degradation of electrical systems. The benchmark construction module extracts historical health data of the robot under historical health or normal operating conditions from the robot management system. The extracted historical health data can cover various typical working conditions of the robot to ensure data robustness, and calculates historical time-domain characteristic coefficients based on the historical health data. and historical frequency domain characteristic coefficients ; Historical time-domain characteristic coefficients The calculation formula is: ; In the formula, Represents historical time-domain characteristic coefficients. This represents the lower limit of historical time-domain characteristics. Represents the upper limit of historical time-domain characteristics; Lower limit of historical time domain characteristics The calculation formula is: ; Upper limit of historical time domain features The calculation formula is: ; In the formula, Representing the The time-domain characteristic coefficients calculated for each historical time window represent a statistical characteristic of the signal in the time domain within a specific time window. Represents the total number of historical windows. This represents a constant used to define the tolerance of the health range in the time domain; Representing all history The average value represents the center line or typical value of the time-domain characteristics of the robot in a healthy state; Representing history The standard deviation represents the maximum allowable normal deviation of the time-domain characteristic coefficients under robot health conditions; Historical frequency domain characteristic coefficients The calculation formula is: ; In the formula, Represents historical frequency domain characteristic coefficients. It represents the lower limit of historical frequency domain characteristics. This represents the upper limit of historical frequency domain characteristics; Lower bound of historical frequency domain characteristics The calculation formula is: ;

[0019] Upper limit of historical frequency domain characteristics The calculation formula is: ; In the formula, Represents historical frequency domain characteristic coefficients. Representing the The frequency domain characteristic coefficients calculated for each historical time window represent the distribution characteristics of signal energy in the frequency domain within the corresponding time window. Represents the total number of historical windows. This represents a constant used to define the tolerance of the healthy range in the frequency domain; Representing history The arithmetic mean represents the center line or typical value of the frequency domain characteristics under the robot's healthy state; Representing history The standard deviation represents the inherent, normal fluctuation range of the frequency domain characteristic coefficients under healthy conditions; Calculate historical time-domain characteristic coefficients using historical health data. and historical frequency domain characteristic coefficients This approach quantifies health status into comparable upper and lower bounds, avoiding limitations imposed by subjective experience or fixed thresholds. It covers historical data from various typical operating conditions to ensure the robustness of the benchmark, reducing misjudgments caused by changes in operating conditions, and utilizes constants. and Adjusting the range of healthy values ​​can balance the false alarm rate and the false negative rate; Health assessment module: used to integrate the calculation results of the machine data processing module and the benchmark construction module to build a comprehensive model of the robot's health coefficient; The expression for the comprehensive model of robot health coefficient is as follows: ; In the expression, The calculation result of the robot health coefficient integrates the deviation between the time domain and the frequency domain into a single health coefficient, which comprehensively reflects the hardware status and avoids the one-sidedness of a single feature. The magnitude of the robot health coefficient directly reflects the health status, which is convenient for operators to understand and make subsequent decisions, and intuitively quantifies the health level of the robot. Lifespan prediction module: Used to output the robot's health status prediction based on a comprehensive model of robot health coefficients, specifically: When the robot's health coefficient A value of 1.5 or higher indicates that the robot is in excellent health and operating normally. When the robot's health coefficient is greater than or equal to 1 but less than 1.5, it indicates that the robot is experiencing performance fluctuations and has early signs of failure. When the robot's health coefficient is less than 1.5, it indicates that the robot's health status has deteriorated and there are clear signs of failure. The robot must be troubleshooted and maintained. The lifespan prediction module classifies the robot's health coefficient prediction results into levels, upgrading the traditional alarm to a progressive early warning system. It captures early fault signals in advance, preventing minor problems from becoming major ones. By setting clear handling strategies, it reduces unplanned downtime and secondary damage, lowers maintenance costs, and systematically solves the problems of limited perception, delayed early warning, and weak robustness of traditional control systems. It intuitively quantifies the robot's health status, ultimately realizing the transformation from passive alarm to proactive health management.

[0020] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An integrated control system based on an intelligent robot, characterized in that: It includes a machine data collection module, a machine data processing module, a benchmark construction module, a health assessment module, and a lifespan prediction module; Machine data collection module: used to deploy multi-dimensional sensing devices in mechanical components and electrical systems, and integrate the operating parameters of mechanical components, electrical systems, and environmental operating parameters; Machine data processing module: Based on the data acquired by the machine data collection module, extracts features and calculates time-domain feature coefficients. and frequency domain characteristic coefficients ; The baseline construction module is used to connect to the robot management system to obtain historical health data and calculate historical time-domain feature coefficients based on the historical health data. and historical frequency domain characteristic coefficients ; Health assessment module: used to integrate the calculation results of the machine data processing module and the benchmark construction module to build a comprehensive model of the robot's health coefficient; Lifetime prediction module: Used to output the robot's health status prediction based on the comprehensive model of robot health coefficient.

2. The integrated control system based on an intelligent robot according to claim 1, characterized in that: The machine data collection module deploys vibration sensors, torque sensors, and displacement sensors at the reducer, bearings, and joints of the mechanical components to collect vibration spectrum, instantaneous torque, and gap change data of the mechanical components and integrate the operating parameters of the mechanical components. The machine data collection module also collects motor stator current and driver output voltage through current sensors and voltage monitoring modules, and integrates electrical system operating parameters by monitoring winding temperature and circuit board temperature in conjunction with temperature sensors. The machine data collection module also integrates visual sensors, sound sensors, GPS positioning systems, and IMU inertial measurement units to acquire data on abnormal appearance, abnormal noise, operational changes, and load changes of the robot hardware, as well as environmental operating parameters.

3. The integrated control system based on an intelligent robot according to claim 2, characterized in that: The time-domain characteristic coefficients The calculation formula is: ； In the formula, Represents the time-domain characteristic coefficients. This represents the total number of data points collected by a single sensor within a data window. Represents the first in the data window One data point; Represents all data within the data window The average value.

4. An integrated control system based on an intelligent robot according to claim 2, characterized in that: The frequency domain characteristic coefficients The calculation formula is: ； In the formula, Represents frequency domain characteristic coefficients, This represents the total number of frequency components in the spectrum obtained after the time-domain signal from a single sensor has undergone FFT transformation. The first in the representative spectrum The center frequency corresponding to each component The first in the representative spectrum Frequency components The corresponding amplitude.

5. An integrated control system based on an intelligent robot according to claim 1, characterized in that: The historical time-domain feature coefficients The calculation formula is: ； In the formula, Represents historical time-domain characteristic coefficients. This represents the lower limit of historical time-domain characteristics. This represents the upper limit of historical time-domain characteristics.

6. An integrated control system based on an intelligent robot according to claim 5, characterized in that: The lower limit of historical time domain features The calculation formula is: ； The upper limit of historical time domain features The calculation formula is: ； In the formula, Representing the The time-domain characteristic coefficients calculated from each historical time window Represents the total number of historical windows. Represents a constant; Representing all history The average value; Representing history The standard deviation.

7. An integrated control system based on an intelligent robot according to claim 1, characterized in that: Historical frequency domain characteristic coefficients The calculation formula is: ； In the formula, Represents historical frequency domain characteristic coefficients. It represents the lower limit of historical frequency domain characteristics. This represents the upper limit of historical frequency domain characteristics.

8. An integrated control system based on an intelligent robot according to claim 7, characterized in that: The lower limit of the historical frequency domain features The calculation formula is: ； The upper limit of the historical frequency domain features The calculation formula is: ； In the formula, Represents historical frequency domain characteristic coefficients. Representing the Frequency domain characteristic coefficients calculated from historical time windows Represents the total number of historical windows. Represents a constant; Representing history The arithmetic mean; Representing history The standard deviation.

9. An integrated control system based on an intelligent robot according to claim 8, characterized in that: The expression for the comprehensive model of the robot's health coefficient is as follows: ; In the expression, The result represents the calculation of the robot's health coefficient.

10. An integrated control system based on an intelligent robot according to claim 9, characterized in that: When the robot health coefficient A value of 1.5 or higher indicates that the robot is in excellent health and operating normally. When the robot's health coefficient is greater than or equal to 1 but less than 1.5, it indicates that the robot is experiencing performance fluctuations and has early signs of failure. When the robot's health coefficient is less than 1.5, it indicates that the robot's health has deteriorated and there are clear signs of failure. Troubleshooting and maintenance of the robot are necessary.

Images