Industrial robot tool change operation monitoring system based on digital twinning

Abstract

The application discloses an industrial robot arm tool changing operation monitoring system based on digital twinning, and relates to the technical field of digital twinning. The system integrates a floating platform model, a robot arm model and a tool changing device model to create a complete digital twinning model, and sets up a perception sensor group to collect real physical world data of the platform, the robot arm and the tool changing device in real time. These data are transmitted to the digital twinning model in real time for processing and simulation, so that the virtual and physical world states can be bidirectionally synchronized. This highly integrated monitoring method can timely identify the resonance risk between the platform and the robot arm and the change in the tool locking trend. Through effective risk warning and adjustment mechanisms, the system can ensure the accuracy and safety of the tool changing operation, avoid tool changing failure or equipment damage caused by resonance or locking abnormalities, thereby improving production efficiency and reducing downtime.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, specifically to an industrial robotic arm tool changing operation monitoring system based on digital twins. Background Technology

[0002] With the continuous development of technology, intelligent manufacturing has gradually become mainstream in global industrial production, encompassing multiple areas such as robotics, sensor technology, and artificial intelligence. Specifically in the field of robotics, industrial robotic arms are widely used in manufacturing, assembly, inspection, and packaging, especially in industries such as automotive, electronics, and aerospace. Further refining the process, tool changing is a crucial aspect of robotic arm operation, particularly in CNC machine tools and precision machining, where the accuracy and efficiency of tool changing directly impact production quality and cost. Successful tool changing requires the robotic arm to complete tool replacement efficiently and accurately, but factors such as vibration and resonance often lead to operational failures or instability.

[0003] While many robotic arm systems have achieved automatic tool changing, several problems remain, particularly the resonance issue that occurs during tool changing. Due to frequency coupling and resonance between the floating platform and the robotic arm, excessive vibration of the end effector often occurs, affecting tool locking accuracy and potentially leading to tool locking failure. This is especially pronounced under the influence of floating platform structure and environmental factors such as wind speed and temperature variations. Traditional monitoring systems mostly consider only the vibration characteristics of the robotic arm itself, neglecting the influence of the platform frequency, and thus cannot effectively monitor and adjust the resonance coupling between the two. This design deficiency makes the system prone to serious problems such as tool lock-up and alignment deviations when encountering high-frequency superposition and unstable environmental factors, impacting production efficiency and product quality. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an industrial robotic arm tool changing operation monitoring system based on digital twins, which solves the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a digital twin-based industrial robotic arm tool changing operation monitoring system, comprising...

[0006] Model integration module: By integrating the floating platform model, robotic arm model, and tool changer model, a digital twin model is created, and a group of sensing sensors is set up to collect real data in real time and transmit it to the digital twin model;

[0007] Digital twin synchronization module: Based on real data, it executes a two-way synchronization mechanism in the digital twin model to synchronize the state of the real physical world and the state of the virtual physical world, extracts the perception data, and then performs preprocessing to obtain a standard perception data set.

[0008] Resonance Coupling Simulation Module: By extracting standard resonance data sets, the module calculates and outputs the resonance risk factor R, and based on the resonance risk factor R, it calculates and outputs the tool locking trend index S. At the same time, it sets a risk threshold Sth for preliminary comparative evaluation.

[0009] Environmental Adaptation Adjustment Module: When environmental adaptation adjustment is triggered, the module extracts a standard environmental perception data set, calculates and outputs an environmental impact coefficient N, and then calculates and outputs the micro-vibration frequency adjustment amount △Fm at the end of the robotic arm based on the environmental impact coefficient N.

[0010] Tool changing operation risk analysis module: By integrating the adjusted resonance risk factor R' and the tool locking trend index S', a comprehensive risk index Palarm is calculated and output, and a tool changing early warning threshold Pth is set for secondary comparative evaluation.

[0011] Preferably, the model integration module includes an integration unit and a real data acquisition unit;

[0012] The integration unit acquires the floating platform model, the robotic arm model, and the tool changing device model, and integrates them to create a digital twin model.

[0013] The floating platform model is modeled using a finite element modeling tool. The floating platform structure includes platform dimensions and connection relationships. Structural dynamics analysis is used to extract simulation frequencies, damping coefficients, and vibration transmission paths. The platform structure and structural dynamics are then combined to obtain the floating platform model, which is input into a digital twin engine as technical modal response features.

[0014] The robotic arm model is constructed using ROS and GAzebo / Webots to create a physical model of the robotic arm, which includes a multi-joint structure and degrees of freedom. Lagrange is used for dynamic modeling to establish the angular velocity, angular acceleration, and end effector trajectory of each key component of the robotic arm.

[0015] The tool changing device model is constructed by using the discrete element modeling method (DEM) to simulate the three-stage motion trajectory of tool installation, alignment and locking, and using a hybrid time-domain and frequency-domain model to describe the behavior deviation under re-vibration.

[0016] The real data acquisition unit collects real data from the floating platform, robotic arm, and tool changing device in real time by setting up a sensor group, and transmits the real data to the constructed digital twin model via wireless communication.

[0017] The sensor group includes an IMU inertial detection unit, a laser interferometer, a joint position encoder, a temperature sensor, a humidity sensor, a wind speed sensor, an industrial camera, and a vision recognition system.

[0018] The real data includes the three-axis acceleration and angular velocity of the floating platform and the robotic arm, high-frequency micro-vibration, real-time attitude of the robotic arm, temperature, humidity, wind speed, and tool changing status.

[0019] Preferably, the digital twin synchronization module includes a data processing input unit, a synchronization update and communication hub unit, and a sensing data extraction unit;

[0020] The data processing input unit receives real data in real time in the digital twin model and standardizes the data format of the real data. The data format standardization includes timestamp synchronization and multi-source heterogeneous data fusion. The digital twin model is driven based on the real data after data format standardization. The digital twin model driving method includes data driving and model prediction.

[0021] The data-driven approach uses real-time, real-world data to drive the state of the virtual model in the digital twin model.

[0022] The model prediction uses the virtual model state to predict the physical reality state 2 seconds from now.

[0023] The timestamp synchronization uses ROS time to synchronize the timestamps of all parameters in the real data to the same time.

[0024] The multi-source heterogeneous data fusion uses a multimodal fusion algorithm to fuse real data from different sources.

[0025] The synchronization update and communication hub unit uses the DDS / MQTT protocol as the main communication hub skeleton and executes a two-way synchronization mechanism to synchronize the state of the real physical world and the state of the virtual physical world.

[0026] The bidirectional synchronization mechanism obtains real data from the physical world in real time to simulate the state of the virtual model, and then predicts the state of the physical world 2 seconds later through the virtual model state simulation, and feeds back the control commands to the robotic arm control system.

[0027] Preferably, the sensing data extraction unit performs virtual-real synchronization and predictive feedback based on the created digital twin model, extracts sensing data in real time, and preprocesses the sensing data to obtain a standard sensing dataset. The preprocessing includes dimensionless standardization to eliminate the dimensional influence of all parameters in the sensing data. The standard sensing dataset includes an environmental sensing dataset and a resonance sensing dataset.

[0028] The environmental perception dataset includes temperature Wd(t) at time t, humidity Sd(t) at time t, wind speed Fs(t) at time t, and wind speed frequency Fw(t) at time t.

[0029] The resonance sensing dataset includes the resonant frequency Fp(t) of the floating platform at time t, the micro-vibration frequency Fm(t) of the end effector of the robotic arm at time t, and the vibration amplitude Ap(t) of the floating platform at time t.

[0030] Preferably, the resonance coupling simulation module includes a resonance risk analysis unit, a tool locking trend analysis unit, and a locking evaluation unit;

[0031] The resonance risk analysis unit constructs a resonance risk factor calculation formula in the digital twin model, extracts the resonance perception dataset and inputs it into the resonance risk factor calculation formula, calculates and outputs the resonance risk factor R, and analyzes the resonance risk between the floating platform and the robotic arm.

[0032] The resonance risk factor R is calculated and output using the following resonance risk factor calculation formula;

[0033]

[0034] In the formula, R(t) represents the resonance risk factor at time t, cos represents the cosine function, φ(t) represents the phase difference of frequency coupling at time t, γ represents the resonance sensitivity coefficient, e represents the exponential function, β represents the adjustment factor, and Ath represents the vibration threshold of the floating platform, which is dimensionless.

[0035] Preferably, the tool locking trend analysis unit calculates and outputs the tool locking trend index S by extracting the resonance risk factor R(t) at the current time t and combining it with the ambient temperature, analyzes the cumulative risk of tool locking, and performs dynamic simulation feedback based on the resonance risk assessment of the floating platform and the robotic arm and environmental changes.

[0036] The tool locking tendency index S is calculated and output using the following algorithm formula;

[0037]

[0038] In the formula, S(t) represents the tool locking tendency index at time t, Wdmax represents the upper limit of safe temperature, d represents the differential variable, and dt represents the time differential variable.

[0039] Preferably, the locking assessment unit extracts the time points of tool locking failure under different combinations of frequency, temperature and floating platform acceleration, obtains the risk threshold Sth through a reverse calculation formula, and conducts a preliminary comparison and assessment between the tool locking trend index S(t) at time t and the risk threshold Sth to analyze the current cumulative risk status of the tool. The specific assessment content is as follows:

[0040] When the tool locking trend index S(t) at time t is less than the risk threshold Sth, it means that the locking trend is within a controllable range, the current tool change is safe, and no adjustment is required.

[0041] When the tool locking trend index S(t) at time t is greater than or equal to the risk threshold Sth, it indicates that the locking trend is abnormally accumulated. At this time, the tool change operation is suspended and the environmental adaptation adjustment is triggered.

[0042] Preferably, the environmental adaptation and adjustment module includes an environmental impact analysis unit and a sensing and adjustment unit;

[0043] The environmental impact analysis unit extracts the current environmental perception dataset after triggering environmental adaptation adjustments through preliminary comparative assessment, and calculates and outputs the environmental impact coefficient N.

[0044] The environmental impact coefficient N is calculated and output using the following algorithm formula:

[0045] N(t)=(a1·Wd(t))+(a2·Sd(t))+(a3·Fs(t));

[0046] In the formula, N(t) represents the environmental impact coefficient at time t, a1, a2 and a3 represent the preset weight values ​​of temperature Wd, humidity Sd and wind speed Fs respectively, and a1+a2+a3=1, the specific values ​​of which are set by the user.

[0047] The sensing and adjustment unit calculates and outputs the micro-vibration frequency adjustment amount △Fm at the end of the robotic arm based on the environmental influence coefficient N(t) at time t, and dynamically adjusts the operating frequency of the robotic arm.

[0048] The micro-vibration frequency adjustment ΔFm at the end of the robotic arm is calculated and output using the following algorithm formula;

[0049]

[0050] In the formula, △Fm(t) represents the adjustment amount of the micro-vibration frequency at the end of the robotic arm at time t, and Fwind represents the threshold of wind speed frequency influence, which is dimensionless.

[0051] Preferably, the tool change operation risk analysis module includes a comprehensive tool change operation risk analysis unit and a tool change operation risk assessment unit;

[0052] The comprehensive tool change operation risk analysis unit, after environmental adaptation and adjustment, combines real-time feedback from the digital twin model and operational adjustments to extract and integrate the adjusted resonance risk factor R' and the tool locking trend index S', and performs comprehensive calculation to output the comprehensive risk index Palarm.

[0053] The comprehensive risk index Palarm is calculated and output using the following algorithm formula;

[0054]

[0055] In the formula, Palarm(t) represents the comprehensive risk index at time t, Tmax represents the upper limit of the tool change cycle, Rmax represents the upper limit of the resonance risk, dS(t)' represents the differential variable of the tool locking trend index at time t, dR(t) represents the differential variable of the resonance risk factor at time t, d represents the differential variable, R(t)' represents the resonance risk factor at time t, and S(t)' represents the tool locking trend index at time t.

[0056] Preferably, the tool changing operation risk assessment unit determines the tool changing warning threshold Pth when the tool changing operation is abnormal based on the historical comprehensive risk index Palarm. The tool changing warning threshold Pth is then compared and assessed with the comprehensive risk index Palarm(t) at time t to analyze the tool changing risk after environmental adaptation and adjustment. Based on the assessment results, relevant warning prompts are generated. The specific assessment content is as follows.

[0057] When the comprehensive risk index Palarm(t) at time t is greater than the tool change warning threshold Pth, it indicates that the tool change operation of the robotic arm is in an abnormal state after environmental adaptation and adjustment. At this time, the emergency warning mechanism is activated, and the warning information is sent to the operator through the digital twin model. At the same time, the tool change operation is suspended.

[0058] When the comprehensive risk index Palarm(t) at time t is less than or equal to the tool change warning threshold Pth, it indicates that the robotic arm tool change operation is in normal condition after environmental adaptation and adjustment. At this time, there is no need to generate a warning and start the tool change operation.

[0059] Beneficial effects

[0060] This invention provides a digital twin-based monitoring system for tool changing operations of industrial robotic arms. It offers the following advantages:

[0061] (1) This system integrates a floating platform model, a robotic arm model, and a tool changer model to create a complete digital twin model. A sensor array is then used to collect real-time physical world data from the platform, robotic arm, and tool changer. This data is transmitted in real-time to the digital twin model for processing and simulation, enabling bidirectional synchronization between the virtual and physical world states. This highly integrated monitoring method can promptly identify resonance risks between the platform and the robotic arm, as well as changes in tool locking trends. Through effective risk warning and adjustment mechanisms, the system ensures the accuracy and safety of tool changing operations, preventing tool changing failures or equipment damage due to resonance or locking anomalies, thereby improving production efficiency and reducing downtime.

[0062] (2) This system triggers an environmental adaptation adjustment module to collect real-time environmental sensing data such as temperature, humidity, and wind speed during tool changing operations. It then calculates the environmental influence coefficient N(t) and dynamically adjusts the robotic arm's operating frequency ΔFm(t) to avoid resonance or malfunctions caused by environmental changes. Especially in environments with high wind speeds or drastic temperature and humidity fluctuations, the system automatically adjusts the robotic arm's micro-vibration frequency, thereby reducing instability or resonance caused by wind speed or other environmental factors. This environmental adaptive adjustment not only improves the system's adaptability to various environmental conditions but also enhances its stability and operational accuracy, ensuring that tool changing operations can be successfully completed under different environmental conditions.

[0063] (3) Through the tool change operation risk analysis module, the system integrates the adjusted resonance risk factor R' and the adjusted tool locking trend index S', and calculates the comprehensive risk index Palarm. This index comprehensively considers the changes in tool locking trend, resonance risk, and environmental adaptation. When the comprehensive risk index Palarm exceeds the preset tool change warning threshold Pth, the system will automatically trigger an emergency warning mechanism, suspend the tool change operation, and send alarm information through a digital twin model. Through this mechanism, the system can identify potential risks in real time during tool change tasks, ensuring that operators can take necessary measures in a timely manner to avoid equipment damage or production stoppage caused by abnormal risks, thereby improving the safety and reliability of tool change operations. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of the industrial robotic arm tool changing operation monitoring system module based on digital twin of the present invention;

[0065] Figure 2 This is a flowchart for evaluating the digital twin-based industrial robotic arm tool changing operation monitoring system of the present invention.

[0066] Figure 3 This is a trend chart of the tool locking trend index S. Detailed Implementation

[0067] 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.

[0068] Example 1

[0069] This invention provides a digital twin-based industrial robotic arm tool changing operation monitoring system. Please refer to [link / reference]. Figure 1 , Figure 2 and Figure 3 ,include

[0070] Model integration module: By integrating the floating platform model, robotic arm model, and tool changer model, a digital twin model is created, and a group of sensing sensors is set up to collect real data in real time and transmit it to the digital twin model;

[0071] Digital twin synchronization module: Based on real data, it executes a two-way synchronization mechanism in the digital twin model to synchronize the state of the real physical world and the state of the virtual physical world, extracts the perception data, and then performs preprocessing to obtain a standard perception data set.

[0072] Resonance Coupling Simulation Module: By extracting standard resonance data sets, the module calculates and outputs the resonance risk factor R, and based on the resonance risk factor R, it calculates and outputs the tool locking trend index S. At the same time, it sets a risk threshold Sth for preliminary comparative evaluation.

[0073] Environmental Adaptation Adjustment Module: When environmental adaptation adjustment is triggered, the module extracts a standard environmental perception data set, calculates and outputs an environmental impact coefficient N, and then calculates and outputs the micro-vibration frequency adjustment amount △Fm at the end of the robotic arm based on the environmental impact coefficient N.

[0074] Tool changing operation risk analysis module: By integrating the adjusted resonance risk factor R' and the tool locking trend index S', a comprehensive risk index Palarm is calculated and output, and a tool changing early warning threshold Pth is set for secondary comparative evaluation.

[0075] In this embodiment, the system integrates the floating platform model, robotic arm model, and tool changer model through a model integration module. A high-precision digital twin model is constructed using finite element method, Lagrange modeling, and DEM methods. Real-world data is collected via a sensor array, achieving a fusion of model-driven and data-driven approaches. Secondly, the digital twin synchronization module employs timestamp synchronization, multimodal fusion, and the DDS / MQTT communication protocol to construct a bidirectional synchronization mechanism between virtual and real states, and extracts standard sensing data sets for subsequent analysis. Next, through a resonance coupling simulation module, the system calculates and outputs a resonance risk factor R based on the resonance sensing data set. It further combines this with an environmental temperature assessment of the tool locking trend index S, comparing it with the locking risk threshold Sth to identify potential locking risks in advance. When the initial resonance risk assessment exceeds the threshold, the system automatically triggers an environmental adaptation adjustment module. This module extracts parameters such as temperature, humidity, and wind speed from the standard environmental sensing data set, calculates the environmental influence coefficient N, and outputs the adjustment amount ΔFm for the micro-vibration frequency at the robotic arm end, dynamically optimizing operating parameters and enhancing the system's adaptability to environmental disturbances. Finally, the tool change operation risk analysis module integrates the adjusted resonance risk factor R' and the adjusted tool locking trend index S' to calculate and output the comprehensive risk index Palarm. This Palarm is then used for a secondary evaluation against the tool change warning threshold Pth to determine whether the current tool change task is in an abnormal state, thus achieving full-process risk control and intelligent decision-making. Compared to the passive monitoring mode of traditional tool change systems that ignores the platform's frequency domain characteristics and dynamic environmental changes, relying solely on static judgments based on the robotic arm itself, this system introduces "floating platform structure and frequency domain resonance coupling parameters." This enables dynamic simulation prediction of frequency coupling and phase synchronization between the floating platform and the robotic arm. Furthermore, by integrating environmental perception and frequency self-tuning capabilities, it achieves a more accurate and robust tool self-locking anomaly identification and risk avoidance mechanism.

[0076] Example 2

[0077] Please see Figure 1 Specifically: the model integration module includes an integration unit and a real data acquisition unit;

[0078] The integration unit creates a digital twin model by integrating the floating platform model, robotic arm model, and tool changer model.

[0079] The floating platform model is created by modeling the floating platform structure using finite element modeling tools. The floating platform structure includes the platform size and connection relationships. Structural dynamics analysis is then used to extract the simulation frequency, damping coefficient, and vibration transmission path. The platform structure and structural dynamics are then combined to obtain the floating platform model, which is then input into the digital twin engine as technical modal response characteristics.

[0080] The robotic arm model was constructed using ROS and GAzebo / Webots to build the physical model of the robotic arm, which includes a multi-joint structure and degrees of freedom. Lagrange was used for dynamic modeling to establish the angular velocity, angular acceleration and end effector trajectory of each key component of the robotic arm.

[0081] The tool changing device model is constructed by using the discrete element modeling method (DEM) to simulate the three-stage motion trajectory of tool installation, alignment and locking, and using a hybrid time-domain and frequency-domain model to describe the behavior deviation under re-vibration.

[0082] The real data acquisition unit collects real data from the floating platform, robotic arm, and tool changing device in real time by setting up a sensor group, and transmits the real data to the constructed digital twin model via wireless communication.

[0083] The sensor suite includes an IMU (Inertial Measurement Unit), a laser interferometer, a joint position encoder, a temperature sensor, a humidity sensor, a wind speed sensor, an industrial camera, and a vision recognition system.

[0084] Real-world data includes the three-axis acceleration and angular velocity of the floating platform and robotic arm, high-frequency micro-vibration, real-time posture of the robotic arm, temperature, humidity, wind speed, and tool changing status.

[0085] In this embodiment, the system, through the collaborative work of the integration unit and the real data acquisition unit, completes high-fidelity modeling and multi-dimensional data-driven integration of the three core systems: the floating platform, the industrial robotic arm, and the tool changer. The integration unit uses the finite element method to construct a structural model of the floating platform and extracts its modal response characteristics, such as simulated frequency, damping coefficient, and vibration transmission path, through structural dynamics. Simultaneously, it constructs a multi-joint physical model of the robotic arm based on ROS and GAzebo / Webots, and uses Lagrange dynamics modeling to accurately describe the angular velocity, angular acceleration, and end effector trajectory of each joint. Furthermore, it uses the DEM discrete element method to construct a model of the tool changer, covering the entire process of tool installation, alignment, and locking, and integrates time-domain and frequency-domain hybrid modeling to realistically simulate motion deviation behavior under vibration. The real data acquisition unit is equipped with multi-source sensing devices such as IMU, laser interferometer, encoder, temperature and humidity sensor, wind speed sensor, and industrial camera to realize real-time sensing and wireless transmission of the status parameters of the entire chain of robotic arm, platform and tool changer, such as three-axis acceleration, angular velocity, high-frequency micro-vibration, attitude and changing status, and build a data foundation for virtual and real two-way linkage.

[0086] Example 3

[0087] Please see Figure 1Specifically: the digital twin synchronization module includes a data processing input unit, a synchronization update and communication hub unit, and a sensory data extraction unit;

[0088] The data processing input unit receives real data in real time in the digital twin model and standardizes the data format of the real data. The data format standardization includes timestamp synchronization and multi-source heterogeneous data fusion. The digital twin model is driven based on the real data after data format standardization. The digital twin model driving method includes data driving and model prediction.

[0089] Data-driven approaches use real-time, real-world data to drive the state of the virtual model within a digital twin model.

[0090] The model prediction uses the virtual model state to predict the physical reality state 2 seconds from now.

[0091] Timestamp synchronization uses ROS time to synchronize the timestamps of all parameters in the real data to the same time.

[0092] Multi-source heterogeneous data fusion combines real data from different sources by employing a multimodal fusion algorithm.

[0093] The synchronization update and communication hub unit uses the DDS / MQTT protocol as the main communication hub skeleton to achieve low-latency, high-frequency data stream synchronization and execute a two-way synchronization mechanism to synchronize the state of the real physical world and the state of the virtual physical world.

[0094] The two-way synchronization mechanism obtains real data from the physical world in real time to simulate the state of the virtual model, and then predicts the state of the physical world 2 seconds later by simulating the state of the virtual model, and feeds back the control commands to the robotic arm control system.

[0095] The sensing data extraction unit performs virtual-real synchronization and predictive feedback based on the created digital twin model, extracts sensing data in real time, and preprocesses the sensing data to obtain a standard sensing dataset. The preprocessing includes dimensionless standardization to eliminate the dimensional influence of all parameters in the sensing data. The standard sensing dataset includes an environmental sensing dataset and a resonance sensing dataset.

[0096] The environmental perception dataset includes temperature Wd(t) at time t, humidity Sd(t) at time t, wind speed Fs(t) at time t, and wind speed frequency Fw(t) at time t;

[0097] The resonance sensing dataset includes the resonant frequency Fp(t) of the floating platform at time t, the micro-vibration frequency Fm(t) of the end effector of the robotic arm at time t, and the vibration amplitude Ap(t) of the floating platform at time t.

[0098] In this embodiment, the system's digital twin synchronization module achieves high-precision, low-latency bidirectional synchronization and closed-loop drive control between the virtual model and the real physical system through the coordinated operation of the data processing input unit, the synchronization update and communication hub unit, and the sensing data extraction unit. The module first performs timestamp synchronization and multimodal fusion processing on the received multi-source real data through the data processing input unit, unifying data timing, fusing heterogeneous data sources, and forming a standardized input format. This format is used to drive the digital twin model in real time and achieve a dual-mechanism linkage of data-driven and model prediction, ensuring high consistency of the current state while providing predictive feedback on the physical state within the next two seconds, providing a priori basis for early warning and control. The DDS / MQTT low-latency communication framework constructed by the synchronization update and communication hub unit ensures stable transmission of high-frequency data streams and real-time updates of the physical-virtual bidirectional state, ensuring the closed-loop response efficiency of system commands. The sensing data extraction unit extracts two core sensing data types—environment and resonance—from the synchronization model in real time and eliminates differences in physical dimensions through dimensionless standardization processing, forming a standard sensing dataset that can be used for risk calculation and dynamic adjustment. The deployment of this module completed key tasks such as dynamic driving, predictive feedback, data standardization, and perception synchronization of the digital twin model, constructing the core central mechanism of the digital twin system from "data acquisition" to "model-driven" and then to "feedback control." This is in contrast to the traditional robotic arm tool changing system, which suffers from scattered sensor data processing, asynchronous timing, and lagging feedback.

[0099] Example 4

[0100] Please see Figure 1 , Figure 2 and Figure 3 Specifically: the resonance coupling simulation module includes a resonance risk analysis unit, a tool locking trend analysis unit, and a locking evaluation unit;

[0101] The resonance risk analysis unit constructs a resonance risk factor calculation formula in the digital twin model, extracts the resonance perception dataset and inputs it into the resonance risk factor calculation formula, calculates and outputs the resonance risk factor R, and analyzes the resonance risk between the floating platform and the robotic arm.

[0102] The resonance risk factor R is calculated and output using the following resonance risk factor calculation formula;

[0103]

[0104] In the formula, R(t) represents the resonance risk factor at time t, cos represents the cosine function, φ(t) represents the phase difference of frequency coupling at time t, that is, the difference between the angular displacement of the end of the robotic arm changing with time and the angle between the angular displacement of the floating platform changing with time, γ represents the resonance sensitivity coefficient, e represents the exponential function, β represents the adjustment factor used to control the sensitivity of the floating platform vibration, and Ath represents the floating platform vibration threshold, which is dimensionless. That is, when the platform vibration exceeds this floating platform vibration threshold, the resonance risk increases sharply.

[0105] This represents the ratio of the frequency difference between the platform and the robotic arm. This item quantifies the frequency difference between the platform and the robotic arm. When the frequencies of the platform and the robotic arm are close;

[0106] (cos(φ(t))·γ) represents the effect of phase difference on resonance risk. This term represents the effect of phase difference between the platform and the robotic arm. If the phase difference φ(t) is large, that is, the vibration of the two is basically synchronized, then the resonance risk is relatively large.

[0107] This indicates the nonlinear enhancement of resonance risk when the platform vibration amplitude exceeds the threshold. It is used to adjust the influence of platform vibration on resonance risk. When the platform vibration amplitude Ap(t) approaches or exceeds Ath, the output of the function increases sharply, which indicates a significant increase in resonance risk.

[0108] The tool locking trend analysis unit extracts the resonance risk factor R(t) at the current time t and combines it with the ambient temperature to calculate and output the tool locking trend index S. It analyzes the cumulative risk of tool locking and performs dynamic simulation feedback based on the resonance risk assessment of the floating platform and the robotic arm and environmental changes.

[0109] The tool locking tendency index S is calculated and output using the following algorithm formula;

[0110]

[0111] In the formula, S(t) represents the tool locking tendency index at time t, Wdmax represents the upper limit of safe temperature, d represents the differential variable, and dt represents the time differential variable.

[0112] The tool locking assessment unit extracts the time points of tool locking failure under different combinations of frequency, temperature and floating platform acceleration, and obtains the risk threshold Sth through a reverse calculation formula, which is Sth = max(S(t)|1), where 1 indicates that the tool change process is completed safely. The tool locking trend index S(t) at time t is compared with the risk threshold Sth for preliminary assessment, and the cumulative risk status of the current tool is analyzed. The specific assessment content is as follows:

[0113] When the tool locking trend index S(t) at time t is less than the risk threshold Sth, it means that the locking trend is within a controllable range, the current tool change is safe, and no adjustment is required.

[0114] When the tool locking trend index S(t) at time t is greater than or equal to the risk threshold Sth, it indicates that the locking trend is abnormally accumulated and there is a risk that the tool will be clamped by resonance, locked, or misaligned. At this time, the tool change operation is suspended and environmental adaptation adjustment is triggered.

[0115] In this embodiment, the system's resonance coupling simulation module achieves precise modeling of the vibration coupling mechanism between the robotic arm and the floating platform and dynamic monitoring of the risk evolution process through hierarchical collaboration of the resonance risk analysis unit, the tool locking trend analysis unit, and the locking evaluation unit. First, the module uses the resonance risk analysis unit to call the resonance sensing dataset in the digital twin model and constructs a resonance risk factor calculation formula R based on key parameters such as coupling frequency difference, phase difference, and platform vibration amplitude, thereby achieving quantitative analysis of the potential resonance state between the floating platform and the robotic arm. Further, the tool locking trend analysis unit jointly models the real-time resonance risk factor R with external disturbance factors such as ambient temperature to calculate the tool locking trend index S, reflecting the cumulative change trend of tool locking risk under specific vibration and temperature environments. Finally, the locking evaluation unit uses successful sample data from historical tool changing processes to inversely deduce the safe locking threshold Sth for the tool changing task and compares the current locking trend index S with it, achieving dynamic evaluation of the current tool locking state and automatically triggering an environmental adaptation adjustment process when the threshold is exceeded. The core value of this module lies in its pioneering establishment of a physical interaction modeling path among the robotic arm, platform, and tool, primarily driven by resonant coupling. It accomplishes three major tasks: resonance risk identification, locking trend evolution tracking, and tool change safety assessment, truly achieving calculable, predictable, and intervention-oriented potential tool locking anomalies. Compared to traditional systems that rely solely on static judgments based on a single vibration threshold, which struggle to handle multi-source disturbances and complex working conditions, this module, supported by digital twins, integrates frequency domain resonance analysis, nonlinear coupling modeling, and a dynamic temperature and vibration factor coupling feedback mechanism, significantly improving the safety of the tool change process and the system's adaptability.

[0116] Example 5

[0117] Please see Figure 1 Specifically: the environmental adaptation and adjustment module includes an environmental impact analysis unit and a perception and adjustment unit;

[0118] The environmental impact analysis unit extracts the current environmental perception dataset after triggering environmental adaptation adjustments through preliminary comparative assessment, and calculates and outputs the environmental impact coefficient N.

[0119] The environmental impact factor N is calculated and output using the following algorithm formula:

[0120] N(t)=(a1·Wd(t))+(a2·Sd(t))+(a3·Fs(t));

[0121] In the formula, N(t) represents the environmental impact coefficient at time t, a1, a2 and a3 represent the preset weight values ​​of temperature Wd, humidity Sd and wind speed Fs respectively, and a1+a2+a3=1, the specific values ​​of which are set by the user.

[0122] The sensing and adjustment unit calculates and outputs the micro-vibration frequency adjustment amount △Fm at the end of the robotic arm based on the environmental influence coefficient N(t) at time t, and dynamically adjusts the operating frequency of the robotic arm to avoid misoperation or resonance caused by environmental changes.

[0123] The micro-vibration frequency adjustment ΔFm at the end of the robotic arm is calculated and output using the following algorithm formula;

[0124]

[0125] In the formula, △Fm(t) represents the adjustment amount of the micro-vibration frequency at the end of the robotic arm at time t, Fwind represents the threshold of wind speed frequency influence, and represents the critical value of the influence of wind speed on the robotic arm, with a dimensionless value.

[0126] This ratio quantifies the frequency difference between the platform and the robotic arm. When the frequencies of the platform and the robotic arm are close, resonance is likely to occur, resulting in violent vibrations and unstable operation. Therefore, by adjusting the operating frequency of the robotic arm to reduce the frequency difference, the risk of resonance can be avoided.

[0127] This indicates that the impact of wind speed on the operation of the robotic arm increases with the increase of wind speed. This item compares the current wind speed Fw(t) with the set wind speed threshold Fwind. When the wind speed exceeds the critical value, the frequency adjustment amount is increased to ensure that the robotic arm can adapt to the strong wind environment.

[0128] This formula adjusts the micro-vibration frequency of the robotic arm in real time based on environmental factors such as wind speed and humidity, avoiding resonance or misoperation caused by environmental changes.

[0129] In this embodiment, the system's environmental adaptation adjustment module consists of an environmental impact analysis unit and a sensing and adjustment unit. Its aim is to perceive, assess, and adaptively control external environmental disturbances, ensuring the stable operation of the robotic arm and the safety of tool changing operations under complex environmental conditions. First, the module, through the environmental impact analysis unit, extracts the environmental perception dataset in real time after resonance risk or abnormal tool locking trend is triggered, constructs an environmental impact coefficient calculation formula, and outputs the current environmental impact coefficient N, quantifying the degree of interference of the external environment on the system's dynamic performance. Subsequently, the sensing and adjustment unit uses the environmental impact coefficient N as the core adjustment basis, combining the current frequency difference ratio between the robotic arm and the platform with the wind speed influence ratio, to calculate and output the adjustment amount ΔFm(t) of the micro-vibration frequency at the robotic arm's end, achieving dynamic adjustment of the robotic arm's operating frequency. This avoids the risks of system resonance and misoperation induced by frequency coupling or environmental disturbances. The deployment of this module completes three key tasks: quantitative perception of environmental factor changes, dynamic response, and feedback control, compensating for the shortcomings of traditional tool changing systems that lack real-time environmental interference suppression mechanisms in complex operating environments. Traditional systems often rely on statically set parameters, which cannot respond in real time to drastic fluctuations in external conditions such as wind speed, humidity, and temperature. This can easily lead to problems such as unstable operation of the robotic arm, misalignment of the cutting tool, or abnormal resonance.

[0130] Example 6

[0131] Please see Figure 1 and Figure 3 Specifically: the tool change operation risk analysis module includes a comprehensive tool change operation risk analysis unit and a tool change operation risk assessment unit;

[0132] The comprehensive tool change operation risk analysis unit, after environmental adaptation and adjustment, combines real-time feedback from the digital twin model and operational adjustments to extract and integrate the adjusted resonance risk factor R' and the tool locking trend index S', and performs comprehensive calculation to output the comprehensive risk index Palarm.

[0133] The comprehensive risk index Palarm is calculated and output using the following algorithm formula;

[0134]

[0135] In the formula, Palarm(t) represents the comprehensive risk index at time t, Tmax represents the upper limit of the tool change cycle, Rmax represents the upper limit of the resonance risk, dS(t)' represents the differential variable of the tool locking trend index at time t, dR(t) represents the differential variable of the resonance risk factor at time t, d represents the differential variable, R(t)' represents the resonance risk factor at time t, and S(t)' represents the tool locking trend index at time t.

[0136] The tool change operation risk assessment unit detects anomalies in tool change operations based on historical comprehensive risk indicators (Palarm). For example, if the system finds in historical data that when the comprehensive risk indicator Palarm exceeds 0.8, the probability of tool change failure is high, which may lead to tool lock-up or operational malfunction. Therefore, a tool change warning threshold Pth = 0.8 can be set. The tool change warning threshold Pth is then compared with the comprehensive risk indicator Palarm(t) at time t for a second assessment. The risk of tool change after environmental adaptation is analyzed, and relevant warning prompts are generated based on the assessment results. The specific assessment content is as follows:

[0137] When the comprehensive risk index Palarm(t) at time t is greater than the tool change warning threshold Pth, it indicates that the tool change operation of the robotic arm is in an abnormal state after environmental adaptation and adjustment. At this time, the emergency warning mechanism is activated, and the warning information is sent to the operator through the digital twin model. At the same time, the tool change operation is suspended.

[0138] When the comprehensive risk index Palarm(t) at time t is less than or equal to the tool change warning threshold Pth, it indicates that the robotic arm tool change operation is in normal condition after environmental adaptation and adjustment. At this time, there is no need to generate a warning and start the tool change operation.

[0139] In this embodiment, the tool change operation risk analysis module of the system consists of a comprehensive tool change operation risk analysis unit and a tool change operation risk assessment unit. It is the core module for realizing system-level tool change safety judgment and operation decision control. After completing environmental adaptation adjustments, this module extracts and integrates the adjusted adjustment resonance risk factor R' and the adjusted tool locking trend index S' based on real-time feedback from the digital twin model. By introducing the risk change rate, i.e., the differential term, it constructs a comprehensive risk index calculation formula and outputs a dynamic tool change risk comprehensive judgment criterion Palarm. This index not only considers the current risk status of the system but also incorporates its changing trend, realizing a leap from static assessment to dynamic prediction. Subsequently, the risk assessment unit sets a tool change warning threshold Pth by comparing with historical data and performs a real-time secondary comparison of the tool change risk comprehensive judgment criterion Palarm to accurately determine whether the current tool change task is in an abnormal state. Once the threshold is exceeded, the system will immediately trigger the emergency warning mechanism within the digital twin model, simultaneously sending warning information to the control system and operators, and automatically suspending the tool change operation to prevent further risk spread. This module effectively accomplishes key tasks such as comprehensive multi-factor assessment of tool change risk, dynamic trend perception, historical experience threshold comparison, and automated response triggering. Compared with traditional methods that rely on a single static indicator for extensive risk assessment, this module integrates three types of risk factors—resonance, locking, and environment—to construct a more robust and forward-looking comprehensive judgment mechanism. Furthermore, by introducing risk differential terms to capture sudden risk trends and combining historical data with customized threshold strategies, the early warning system becomes more targeted and adaptable.

[0140] Specific example: Suppose at a certain time t:

[0141] The current tool locking trend index S(t) = 0.75 (the current tool locking risk is relatively high);

[0142] The current resonance risk factor R(t) = 0.8 (the resonance risk between the platform and the robotic arm is relatively high);

[0143] The current upper limit of resonance risk, Rmax, is 1.0.

[0144] The current maximum tool change cycle time, Tmax, is 10 seconds.

[0145] The current differential value of resonance risk is dR(t) = 0.02, and the rate of risk change is...

[0146] The current tool locking tendency differential value dS(t) = 0.03, and the locking tendency change rate;

[0147] Calculation steps

[0148] The Palarm composite risk index is calculated using the following formula:

[0149]

[0150] Assuming the weighting coefficient α = 1, then substituting the above data:

[0151] Palarm(t)=0.75 / 10+0.8 / 1.0+1·(0.03+0.02)=0.925

[0152] Determine if an alert is triggered

[0153] Assume the tool change warning threshold Pth = 0.8;

[0154] Based on the calculation result Palarm = 0.925, since Palarm(t) > Pth, the system judges that the current tool change task has a high risk and has exceeded the preset warning threshold. Therefore, the system will trigger the emergency warning mechanism, suspend the tool change operation, and issue an alarm to the operator.

[0155] 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. A digital twin-based industrial robotic arm tool changing operation monitoring system, characterized in that: include Model integration module: By integrating the floating platform model, robotic arm model, and tool changer model, a digital twin model is created, and a group of sensing sensors is set up to collect real data in real time and transmit it to the digital twin model; Digital twin synchronization module: Based on real data, it executes a two-way synchronization mechanism in the digital twin model to synchronize the state of the real physical world and the state of the virtual physical world, extracts the perception data, and then performs preprocessing to obtain a standard perception data set. Resonance Coupling Simulation Module: By extracting standard resonance data sets, the module calculates and outputs the resonance risk factor R, and based on the resonance risk factor R, it calculates and outputs the tool locking trend index S. At the same time, it sets a risk threshold Sth for preliminary comparative evaluation. Environmental Adaptation Adjustment Module: When environmental adaptation adjustment is triggered, the module extracts a standard environmental perception data set, calculates and outputs an environmental impact coefficient N, and then calculates and outputs the micro-vibration frequency adjustment amount △Fm at the end of the robotic arm based on the environmental impact coefficient N. Tool changing operation risk analysis module: By integrating the adjusted resonance risk factor R' and the tool locking trend index S', a comprehensive risk index Palarm is calculated and output, and a tool changing early warning threshold Pth is set for secondary comparative evaluation.

2. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 1, characterized in that: The model integration module includes an integration unit and a real data acquisition unit; The integration unit acquires the floating platform model, the robotic arm model, and the tool changing device model, and integrates them to create a digital twin model. The floating platform model is modeled using a finite element modeling tool. The floating platform structure includes platform dimensions and connection relationships. Structural dynamics analysis is used to extract simulation frequencies, damping coefficients, and vibration transmission paths. The platform structure and structural dynamics are then combined to obtain the floating platform model, which is input into a digital twin engine as technical modal response features. The robotic arm model is constructed using ROS and GAzebo / Webots to create a physical model of the robotic arm, which includes a multi-joint structure and degrees of freedom. Lagrange is used for dynamic modeling to establish the angular velocity, angular acceleration, and end effector trajectory of each key component of the robotic arm. The tool changing device model is constructed by using the discrete element modeling method (DEM) to simulate the three-stage motion trajectory of tool installation, alignment and locking, and using a hybrid time-domain and frequency-domain model to describe the behavior deviation under re-vibration. The real data acquisition unit collects real data from the floating platform, robotic arm, and tool changing device in real time by setting up a sensor group, and transmits the real data to the constructed digital twin model via wireless communication. The sensor group includes an IMU inertial detection unit, a laser interferometer, a joint position encoder, a temperature sensor, a humidity sensor, a wind speed sensor, an industrial camera, and a vision recognition system. The real data includes the three-axis acceleration and angular velocity of the floating platform and the robotic arm, high-frequency micro-vibration, real-time attitude of the robotic arm, temperature, humidity, wind speed, and tool changing status.

3. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 2, characterized in that: The digital twin synchronization module includes a data processing input unit, a synchronization update and communication hub unit, and a sensing data extraction unit; The data processing input unit receives real data in real time in the digital twin model and standardizes the data format of the real data. The data format standardization includes timestamp synchronization and multi-source heterogeneous data fusion. The digital twin model is driven based on the real data after data format standardization. The digital twin model driving method includes data driving and model prediction. The data-driven approach uses real-time, real-world data to drive the state of the virtual model in the digital twin model. The model prediction uses the virtual model state to predict the physical reality state 2 seconds from now. The timestamp synchronization uses ROS time to synchronize the timestamps of all parameters in the real data to the same time. The multi-source heterogeneous data fusion uses a multimodal fusion algorithm to fuse real data from different sources; The synchronization update and communication hub unit uses the DDS / MQTT protocol as the main communication hub skeleton and executes a two-way synchronization mechanism to synchronize the real physical world state and the virtual physical world state. The bidirectional synchronization mechanism obtains real data from the physical world in real time to simulate the state of the virtual model, and then predicts the state of the physical world 2 seconds later through the virtual model state simulation, and feeds back the control commands to the robotic arm control system.

4. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 3, characterized in that: The perception data extraction unit performs virtual-real synchronization and prediction feedback based on the created digital twin model, extracts perception data in real time, and preprocesses the perception data to obtain a standard perception dataset. The preprocessing includes dimensionless standardization to eliminate the dimensional influence of all parameters in the perception data. The standard perception dataset includes an environmental perception dataset and a resonance perception dataset. The environmental perception dataset includes temperature Wd(t) at time t, humidity Sd(t) at time t, wind speed Fs(t) at time t, and wind speed frequency Fw(t) at time t. The resonance sensing dataset includes the resonance frequency Fp(t) of the floating platform at time t, the micro-vibration frequency Fm(t) of the end effector of the robotic arm at time t, and the vibration amplitude Ap(t) of the floating platform at time t.

5. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 4, characterized in that: The resonance coupling simulation module includes a resonance risk analysis unit, a tool locking trend analysis unit, and a locking evaluation unit. The resonance risk analysis unit constructs a resonance risk factor calculation formula in the digital twin model, extracts the resonance perception dataset and inputs it into the resonance risk factor calculation formula, calculates and outputs the resonance risk factor R, and analyzes the resonance risk between the floating platform and the robotic arm. The resonance risk factor R is calculated and output using the following resonance risk factor calculation formula; In the formula, R(t) represents the resonance risk factor at time t, cos represents the cosine function, φ(t) represents the phase difference of frequency coupling at time t, γ represents the resonance sensitivity coefficient, e represents the exponential function, β represents the adjustment factor, and Ath represents the vibration threshold of the floating platform, which is dimensionless.

6. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 5, characterized in that: The tool locking trend analysis unit extracts the resonance risk factor R(t) at the current time t and combines it with the ambient temperature to calculate and output the tool locking trend index S, analyzes the cumulative risk of tool locking, and performs dynamic simulation feedback based on the resonance risk assessment of the floating platform and the robotic arm and environmental changes. The tool locking tendency index S is calculated and output using the following algorithm formula; In the formula, S(t) represents the tool locking tendency index at time t, Wdmax represents the upper limit of safe temperature, d represents the differential variable, and dt represents the time differential variable.

7. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 6, characterized in that: The locking assessment unit extracts the time points of tool locking failure under different combinations of frequency, temperature and floating platform acceleration, obtains the risk threshold Sth through a reverse calculation formula, and conducts a preliminary comparison and assessment between the tool locking trend index S(t) at time t and the risk threshold Sth to analyze the current cumulative risk status of the tool. The specific assessment content is as follows: When the tool locking trend index S(t) at time t is less than the risk threshold Sth, it means that the locking trend is within a controllable range, the current tool change is safe, and no adjustment is required. When the tool locking trend index S(t) at time t is greater than or equal to the risk threshold Sth, it indicates that the locking trend is abnormally accumulated. At this time, the tool change operation is suspended and the environmental adaptation adjustment is triggered.

8. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 7, characterized in that: The environmental adaptation and adjustment module includes an environmental impact analysis unit and a perception and adjustment unit; The environmental impact analysis unit extracts the current environmental perception dataset after triggering environmental adaptation adjustments through preliminary comparative assessment, and calculates and outputs the environmental impact coefficient N. The environmental impact coefficient N is calculated and output using the following algorithm formula: N(t)=(a1·Wd(t))+(a2·Sd(t))+(a3·Fs(t)); In the formula, N(t) represents the environmental impact coefficient at time t, a1, a2 and a3 represent the preset weight values ​​of temperature Wd, humidity Sd and wind speed Fs respectively, and a1+a2+a3=1, the specific values ​​of which are set by the user. The sensing and adjustment unit calculates and outputs the micro-vibration frequency adjustment amount △Fm at the end of the robotic arm based on the environmental influence coefficient N(t) at time t, and dynamically adjusts the operating frequency of the robotic arm. The micro-vibration frequency adjustment ΔFm at the end of the robotic arm is calculated and output using the following algorithm formula; In the formula, △Fm(t) represents the adjustment amount of the micro-vibration frequency at the end of the robotic arm at time t, and Fwind represents the threshold of wind speed frequency influence, which is dimensionless.

9. The industrial robotic arm tool changing operation monitoring system based on digital twin as described in claim 8, characterized in that: The tool changing operation risk analysis module includes a comprehensive tool changing operation risk analysis unit and a tool changing operation risk assessment unit; The comprehensive tool change operation risk analysis unit, after environmental adaptation and adjustment, combines real-time feedback from the digital twin model and operational adjustments to extract and integrate the adjusted resonance risk factor R' and the tool locking trend index S', and performs comprehensive calculation to output the comprehensive risk index Palarm. The comprehensive risk index Palarm is calculated and output using the following algorithm formula; In the formula, Palarm(t) represents the comprehensive risk index at time t, Tmax represents the upper limit of the tool change cycle, Rmax represents the upper limit of the resonance risk, dS(t)' represents the differential variable of the tool locking trend index at time t, dR(t) represents the differential variable of the resonance risk factor at time t, d represents the differential variable, R(t)' represents the resonance risk factor at time t, and S(t)' represents the tool locking trend index at time t.

10. The industrial robotic arm tool changing operation monitoring system based on digital twin according to claim 9, characterized in that: The tool change operation risk assessment unit determines the tool change operation abnormality based on the user's judgment of the historical comprehensive risk index Palarm. When the user determines that the tool change operation is abnormal, a tool change warning threshold Pth is set. The tool change warning threshold Pth is then compared and evaluated with the comprehensive risk index Palarm(t) at time t. The analysis of the tool change risk after environmental adaptation and adjustment is carried out, and relevant warning prompts are generated based on the evaluation results. The specific evaluation content is as follows. When the comprehensive risk index Palarm(t) at time t is greater than the tool change warning threshold Pth, it indicates that the tool change operation of the robotic arm is in an abnormal state after environmental adaptation and adjustment. At this time, the emergency warning mechanism is activated, and the warning information is sent to the operator through the digital twin model. At the same time, the tool change operation is suspended. When the comprehensive risk index Palarm(t) at time t is less than or equal to the tool change warning threshold Pth, it indicates that the robotic arm tool change operation is in normal condition after environmental adaptation and adjustment. At this time, there is no need to generate a warning and start the tool change operation.

Images