A robot collaborative programming and simulation method and system supporting cross-device state linkage

Abstract

The present invention provides a robot collaborative programming and simulation method and system supporting cross-device state linkage. It belongs to the cross-technical field of industrial robots, multi-agent systems and virtual simulation. The method comprises: constructing a unified state perception middleware, parsing private communication protocols of robots of different brands, extracting key parameters, and generating standardized device state data; deploying a multi-robot real-time interaction network based on the standardized device state data. By constructing the unified state perception middleware and dynamic collaborative logic scripts, the method realizes real-time sharing and linkage response of key state data of multi-brand robots, reduces the collision risk in collaborative operations, compresses on-site debugging time, and significantly improves the deployment efficiency of complex production lines.

Description

Technical Field

[0001] This invention proposes a robot collaborative programming and simulation method and system that supports cross-device state linkage, belonging to the cross-technical fields of industrial robots, multi-agent systems and virtual simulation. Background Technology

[0002] In the current era of rapid industrial automation development, multi-robot collaborative operation has become a key means to improve production efficiency and cope with complex manufacturing tasks. However, current robot collaborative programming and simulation technologies still have significant limitations: On the one hand, different brands of robots use proprietary communication protocols, making it difficult to share key state data such as torque and safety zones in real time, forming "state awareness silos." Collisions or inefficiencies often occur in collaborative operations due to information lag. On the other hand, traditional collaborative logic relies on preset timing scripts and cannot dynamically respond to sudden state changes such as emergency stops and load shifts, resulting in long debugging cycles and poor on-site adaptability. In addition, existing simulation technologies often ignore physical disturbances such as friction and inertia, causing "simulation-feasible" solutions to be frequently reworked in the field due to safety risks. More importantly, black-box AI strategies lack interpretability, making it difficult for engineers to trust their decision-making logic, still requiring a large amount of human intervention, which restricts the large-scale application of collaborative operations.

[0003] Although the published patents contain descriptions of technologies such as "multi-robot communication" and "offline programming," none of them have broken through the framework of isolated function superposition. They have failed to deeply integrate unified state middleware, state-behavior coupling graph, inverse reinforcement learning and high-fidelity perturbation simulation, and cannot build a cognitive-level collaborative architecture with the capabilities of "state linkage - policy traceability - virtual and real credibility - online evolution". Summary of the Invention

[0004] This invention provides a robot collaborative programming and simulation method and system that supports cross-device state linkage, in order to solve the problems mentioned in the background art above: This invention proposes a robot collaborative programming and simulation method that supports cross-device state linkage, the method comprising: S1. Construct a unified state perception middleware to parse the private communication protocols of robots from different brands, extract key parameters, and generate standardized equipment status data; deploy a multi-robot real-time interactive network based on the standardized equipment status data. S2. Based on the standardized device status data in the multi-robot real-time interaction network, construct a state and behavior coupling diagram to describe the state associations and collaborative constraints between robots; generate a dynamic collaborative logic script through the state and behavior coupling diagram to replace the preset timing script. S3. Build a high-fidelity disturbance simulation environment based on dynamic collaborative logic scripts, integrate physical disturbance models, and simulate complex working conditions in robot collaborative operations; run dynamic collaborative logic scripts in the high-fidelity disturbance simulation environment to generate collaborative operation simulation data; S4. Through inverse reinforcement learning, policy mining is performed on the collaborative operation simulation data to generate an interpretable collaborative decision-making model, revealing the causal relationship between robot behavior selection and state changes; the interpretable collaborative decision-making model is then transformed into a visualized policy graph. S5. Generate online evolution instructions based on the optimized collaborative decision-making model, and synchronously update the dynamic collaborative logic script in the multi-robot real-time interaction network; drive multiple robots to perform actual collaborative operations through online evolution instructions, and continuously collect on-site equipment status data to feed back to the unified status perception middleware to form a closed-loop cognitive architecture, and finally generate a collaborative operation evaluation report.

[0005] This invention proposes a robot collaborative programming and simulation system that supports cross-device state linkage, comprising: One or more processors; Memory, used to store one or more programs; Wherein, when the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method described in any one of the above.

[0006] The beneficial effects of this invention are as follows: By constructing a unified state-aware middleware and dynamic collaborative logic scripts, this method achieves real-time sharing and coordinated response of key state data of multi-brand robots, reducing the risk of collisions in collaborative operations, while also compressing on-site debugging time and significantly improving the deployment efficiency of complex production lines. The high-fidelity disturbance simulation environment, by integrating a physical disturbance model, avoids the rework problem of traditional simulations failing on-site, ensuring the on-site reliability of collaborative strategies. The policy graph generated by inverse reinforcement learning presents the decision logic in a visual form, enhancing engineers' trust in collaborative strategies and supporting rapid manual optimization of key parameters, shortening the strategy iteration cycle from weeks to hours. Furthermore, the closed-loop cognitive architecture continuously collects on-site state data to feed back into the simulation model, enabling online evolution of collaborative strategies. This allows for adaptation to dynamically changing operating environments while maintaining long-term stable collaborative efficiency, providing both safety and flexibility for large-scale collaborative applications of industrial robots. Attached Figure Description

[0007] Figure 1 This is a diagram illustrating the steps of the method described in this invention. Detailed Implementation

[0008] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0009] One embodiment of the present invention, such as Figure 1 As shown, a robot collaborative programming and simulation method supporting cross-device state linkage is disclosed, the method comprising: S1. Construct a unified state perception middleware to parse the private communication protocols of robots from different brands, extract key parameters, including torque, safe zone, and load status, and generate standardized equipment state data; deploy a multi-robot real-time interaction network based on the standardized equipment state data to achieve synchronous sharing and dynamic updating of cross-device state data. S2. Based on the standardized equipment status data in the multi-robot real-time interaction network, construct a state and behavior coupling diagram to describe the state association and cooperative constraints between robots; generate dynamic cooperative logic scripts through the state and behavior coupling diagram to replace the preset timing scripts, and support real-time adjustment of cooperative strategies according to state changes (such as emergency stop, load offset). S3. A high-fidelity disturbance simulation environment is built based on dynamic collaborative logic scripts, and a physical disturbance model is integrated. The physical disturbance model includes friction, inertia and elastic deformation to simulate complex working conditions in robot collaborative operations. The dynamic collaborative logic scripts are run in the high-fidelity disturbance simulation environment to generate collaborative operation simulation data. The collaborative operation simulation data includes indicators such as collision risk, load balance and motion trajectory smoothness. S4. Through inverse reinforcement learning, policy mining is performed on the collaborative operation simulation data to generate an interpretable collaborative decision model, revealing the causal relationship between robot behavior selection and state changes; the interpretable collaborative decision model is transformed into a visual policy graph, supporting engineers to quickly understand the collaborative logic and manually optimize key parameters. S5. Generate online evolution instructions based on the optimized collaborative decision-making model, and synchronously update the dynamic collaborative logic script in the multi-robot real-time interaction network; drive multiple robots to perform actual collaborative operations through online evolution instructions, and continuously collect on-site equipment status data and feed it back to the unified status perception middleware to form a closed-loop cognitive architecture, and finally generate a collaborative operation evaluation report, which includes risk level, efficiency improvement value and debugging cost savings.

[0010] The working principle and effects of the above technical solution are as follows: By constructing a unified state-aware middleware to parse the private protocols of robots from different brands, the compatibility of cross-device data interaction is improved, linkage failures caused by protocol differences are reduced, and the problem of robots from different brands being unable to collaborate is avoided. By generating dynamic collaborative logic scripts to replace preset timing scripts, the flexibility of collaborative strategies is enhanced, and the inability to adjust in time when state changes occur is avoided, resulting in job interruptions and improved stability of collaborative operations. By building a high-fidelity disturbance simulation environment, the debugging cost before actual operation is reduced, and equipment damage caused by on-site trial and error is reduced. By using reverse reinforcement learning to discover collaborative patterns and optimize the decision model, the efficiency of collaborative operations is improved, and time wasted due to blind debugging is avoided. It can achieve real-time state synchronization of multiple robots and continuously optimize the work process through a closed-loop architecture, taking into account both collaborative accuracy and operational reliability, further reducing the operational difficulty for engineers.

[0011] In one embodiment of the present invention, S1 includes: S11. Build a unified state perception middleware to adapt and connect the underlying communication interfaces of robots from different brands, and parse the message structure and data rules of various private communication protocols layer by layer to eliminate communication barriers between different devices. S12. Traverse the parsed protocol data stream, extract key parameters from the robot's operation feedback information, including torque values, safe area boundaries, and load operating status, and remove redundant and invalid protocol auxiliary information. S13. The extracted parameters are formatted and dimension unified, and the data encoding method and numerical measurement unit are unified to generate standardized equipment status data with universal interactive capabilities. S14. Based on standardized equipment status data, build a real-time interactive network for multiple robots, plan data transmission paths and node allocation rules, and open up data interaction channels between various devices. S15. In the completed interactive network, complete the synchronization, sharing and dynamic updating of cross-device status data, continuously refresh the device status information of each terminal, and ensure the real-time consistency of device status data across the entire network.

[0012] The working principle and effects of the above technical solution are as follows: By building a unified state-aware middleware and adapting it to the communication interfaces of different robot brands, and parsing private communication protocols layer by layer, communication barriers between different devices are effectively eliminated, avoiding the problem of devices failing to link together due to protocol incompatibility. By traversing the protocol data stream to extract key parameters and remove redundant information, the accuracy of parameter extraction is improved, and interference from invalid data to subsequent processing is reduced. By standardizing the format and unifying the dimensions of the extracted parameters, standardized device status data is generated, enhancing the universal interaction capability of data and reducing the difficulty of data interaction between different devices. By building a multi-robot real-time interaction network and planning transmission paths, data interaction channels are opened up, improving the transmission efficiency of cross-device status data, ensuring the real-time consistency of device status data across the entire network, enabling smooth linkage of multi-brand robots, providing reliable data support for subsequent collaborative operations, and reducing collaborative failures caused by data inconsistency.

[0013] In one embodiment of the present invention, S2 includes: S21. Retrieve standardized equipment status data continuously transmitted in the multi-robot real-time interaction network, statistically analyze the operating posture and operation progress of each robot, and sort out the operational relationships between robots. S22. Combining the relationships obtained from the analysis with the constraints of the actual operation scenario, construct a state and behavior coupling diagram to fully depict the state relationships and constraint logic in robot collaborative operation. S23. Generate dynamic collaborative logic scripts through the association rules of the state and behavior coupling diagram, abandoning the traditional fixed-time preset scripts, and allowing the script content to have adjustable space as the device state changes; S24. Monitor various state change information during robot operation in real time, focusing on capturing abnormal states, including emergency stop triggering, load offset and position deviation, to ensure complete collection of change information; S25. Based on the collected state change information, adjust the collaborative execution strategy in the dynamic collaborative logic script in real time to adapt to sudden changes in on-site operations and maintain the stability of multi-robot collaborative operations.

[0014] The working principle and effects of the above technical solution are as follows: By retrieving standardized equipment status data from a multi-robot real-time interactive network, the system statistically analyzes robot operating postures and work progress, clearly understanding the operational relationships between robots and reducing coordination errors caused by unclear relationships. By constructing a state and behavior coupling graph, the system fully depicts the collaborative constraint logic, improving the rationality of collaborative strategies and avoiding operational errors caused by chaotic collaborative logic. By generating dynamic collaborative logic scripts, the system abandons traditional fixed-sequence scripts, enhancing the flexibility of collaborative strategies and reducing the difficulty of adapting scripts to changes in the field. By monitoring state mutations in real time and fully collecting abnormal information, the system promptly adjusts collaborative execution strategies, avoiding work interruptions caused by inability to adapt to state mutations and maintaining the stability of multi-robot collaborative operations. This approach can accurately match the constraints of actual work scenarios and quickly respond to emergencies, reducing collaborative failures and further improving the smoothness and reliability of multi-robot collaborative operations.

[0015] In one embodiment of the present invention, S25 includes: S251. Receive the acquired state mutation content, complete the matching and association between the mutation information and the collaborative execution logic, and generate the collaborative strategy pointing information to be adjusted. S252. Based on the information pointing to the collaborative strategy to be adjusted, locate the collaborative execution nodes that need to be changed in the dynamic collaborative logic script, and filter out the execution logic of the corresponding nodes. S253. Perform parameter reallocation and process rearrangement on the selected execution logic to generate new collaborative execution content that adapts to the current mutation scenario; S254. Write the new collaborative execution content into the dynamic collaborative logic script and complete the replacement, thus completing the real-time update of the collaborative execution strategy; S255. Apply the updated collaborative execution strategy to guide multiple robots to carry out operations, adapt to sudden changes in on-site operations, and maintain the stable operation of multi-robot collaborative operations.

[0016] The working principle and effects of the above technical solution are as follows: By receiving and collecting state mutation content, the system matches and associates mutation information with collaborative execution logic, generating clear strategy guidance information and avoiding the problem of inaccurate adjustments due to the disconnect between mutation information and execution logic. By locating the collaborative execution nodes in the script that need to be changed and filtering the corresponding logic, the system reduces invalid adjustment operations, improves the targeting of strategy adjustments, and reduces redundant time consumption during the adjustment process. By reallocating parameters and re-arranging processes for the filtered execution logic, the system generates new execution content adapted to the mutation scenario, enhancing the adaptability of the collaborative strategy and preventing the inability to cope with emergencies due to inadequate adjustments. By replacing the script content in real time and applying the updated strategy, the system achieves rapid updates to the collaborative execution strategy, maintains stable operation of multi-robot collaborative work, and reduces job interruptions or failures caused by state mutations. It can quickly respond to various emergencies and accurately adjust collaborative logic, balancing adjustment efficiency and operational stability, further reducing the impact of emergencies on collaborative operations.

[0017] In one embodiment of the present invention, S253 includes: Extract all the running parameters within the filtered execution logic, combine them with the characteristics of state mutation to redistribute the parameter values, and generate a parameter configuration set adapted to the mutation scenario. Based on the updated parameter configuration set, the execution steps of the original collaborative operation are broken down, and the execution order and connection relationship of each step are adjusted. The adjusted execution steps are arranged in a time sequence, and the action intervals and connection times of each robot in the corresponding steps are planned; the rearranged execution steps and time sequence results are integrated to improve the interactive constraints of collaborative operations. Integrate parameter configuration and process orchestration results to form new collaborative execution content adapted to the current mutation scenario.

[0018] The working principle and effects of the above technical solution are as follows: By extracting the operating parameters within the execution logic and reallocating values ​​based on state change characteristics, a parameter configuration set adapted to the scenario is generated, improving the fit between parameters and on-site working conditions and avoiding coordination deviations caused by parameters not matching the actual state. By splitting the original execution links and adjusting the connection relationships, the workflow layout is optimized, reducing process bottlenecks and poor connections, and lowering the probability of robot action conflicts. By arranging the execution links in a time sequence and planning action intervals, multi-robot cooperation becomes smoother, enhancing the continuity of the workflow. By integrating the link arrangement and time sequence results to improve interaction constraints, the final results are integrated to form new collaborative execution content, which can accurately adapt to sudden state changes while maintaining the overall continuity of collaborative operations, avoiding disruption of the overall work rhythm due to local adjustments, while improving the accuracy of collaborative actions and reducing operational risks and downtime losses caused by process chaos.

[0019] In one embodiment of the present invention, S3 includes: S31. Using dynamic collaborative logic scripts as the core execution framework, build a high-fidelity disturbance simulation environment adapted to multi-robot collaborative operations, and plan the simulation scene size and operation area layout. S32. Integrate a physical disturbance model in the simulation environment. The physical disturbance model includes friction, inertia and elastic deformation to restore the physical laws of real-world scenarios and simulate the complex working conditions in robot collaborative operations. S33. Load the dynamic collaborative logic script into the completed high-fidelity disturbance simulation environment and run it continuously to fully reproduce the entire process of multi-robot collaborative operation. S34. During the simulation process, multi-dimensional simulation indicators are collected in all aspects, including collision risk, load balance and motion trajectory smoothness, etc., and the trend of indicator changes at different time periods is recorded. S35. Integrate the various simulation indicators and deduction process information collected, classify and summarize them to generate complete collaborative operation simulation data.

[0020] The working principle and effects of the above technical solution are as follows: By using dynamic collaborative logic scripts as the core, a high-fidelity disturbance simulation environment adapted to multi-robot operations is built and the scene layout is planned, making the simulation scene more closely match the actual operation requirements and avoiding debugging failures caused by the disconnect between simulation and on-site operations. By integrating physical disturbance models into the simulation environment, the real physical laws are restored, complex working conditions are simulated, the realism of the simulation is enhanced, and the situation where potential risks cannot be identified due to overly simple simulation scenes is reduced. By loading scripts for full-process simulation, the collaborative operation process is fully reproduced, improving the control over the operation process and avoiding action connection errors in actual operations. By comprehensively collecting multi-dimensional simulation indicators and recording their changing trends, complete simulation data is generated, which can not only identify risks such as collisions and load imbalances in advance, but also provide accurate basis for subsequent collaborative strategy optimization, reducing on-site debugging costs and equipment wear and tear, and avoiding failures and losses caused by blindly deploying them into actual operations.

[0021] In one embodiment of the present invention, step S4 includes: S41. Use inverse reinforcement learning to perform deep policy mining on collaborative operation simulation data, analyze the intrinsic relationship between robot behavior and state changes during the simulation process, and extract potential collaborative behavior patterns. S42. Based on the behavioral patterns obtained from mining, construct an interpretable collaborative decision-making model to clearly reveal the causal relationship between robot behavior choices and state changes, thereby improving the transparency of decision-making logic; S43. Transform the internal logic and decision-making path of the collaborative decision-making model into a visual strategy map to present the collaborative rules in an intuitive form, reducing the difficulty for engineers to understand the collaborative logic. S44. Engineers use visual strategy maps to locate weak links in the collaborative logic, and manually adjust and optimize key operating parameters in the decision-making model to improve the model's rationality. S45. Save and solidify the optimized collaborative decision-making model, and eliminate unreasonable logic in the model.

[0022] The working principle and effects of the above technical solution are as follows: By employing inverse reinforcement learning to perform deep policy mining on collaborative operation simulation data, the intrinsic relationship between robot behavior and state changes is analyzed, potential collaborative patterns are extracted, the scientific nature of collaborative strategies is improved, and the problem of inefficient collaboration caused by blindly formulating strategies is avoided. By constructing an interpretable collaborative decision-making model based on behavioral patterns, the causal relationship between behavior and state is clearly revealed, enhancing the transparency of decision-making logic and reducing the obstacles to engineers' understanding of collaborative logic. By transforming the decision model into a visual policy graph, collaborative rules are presented intuitively, reducing the operational difficulty for engineers and the time spent understanding the logic. By having engineers locate weak links and optimize key parameters, unreasonable logic is eliminated, the rationality of the model is improved, and operational failures caused by model bias are avoided. This not only provides a reliable basis for the generation of subsequent actual operation instructions but also shortens the debugging cycle, improves the accuracy of collaborative decision-making, and further reduces ineffective operations and failure losses in collaborative operations.

[0023] In one embodiment of the present invention, S41 includes: S411. Import collaborative operation simulation data, and break down the robot behavior trajectory and state change information in the data into segments to generate discrete data segments corresponding to behavior and state. S412. Perform correlation comparison on discrete data segments, analyze the transmission relationship between robot action triggering and state fluctuation, and form behavioral state correlation features; S413. Use inverse reinforcement learning to iteratively deduce associated features and track the origin and evolution of robot cooperative actions. S414. Extract the robot's action selection tendency under different working conditions from the iterative deduction results and summarize them to form stable cooperative behavior characteristics; S415. Integrate multiple sets of collaborative behavior characteristics and summarize them into potential collaborative behavior patterns applicable to multi-machine collaboration.

[0024] The working principle and effects of the above technical solution are as follows: By importing collaborative operation simulation data, the robot's behavior trajectory and state change information are broken down into segments to generate discrete data fragments, making the data presentation clearer, avoiding deviations in pattern extraction caused by data clutter, and reducing interference from invalid data. By comparing and correlating the discrete data fragments, the transmission relationship between action triggering and state fluctuations is analyzed to form correlation features, improving the accuracy of behavior and state correlation analysis and avoiding the problem of failing to capture the core correlation due to the disconnect between the two. By using inverse reinforcement learning to iteratively deduce the correlation features, the source and evolution of collaborative actions are traced, enhancing the depth of pattern mining and reducing the omission of potential collaborative logic. By extracting action selection tendencies under different working conditions and integrating collaborative behavior features, potential patterns of multi-machine collaboration are summarized, which not only provides solid support for the subsequent construction of collaborative decision-making models, but also improves the pertinence of collaborative strategies, avoids inefficiency caused by blindly formulating strategies, and further reduces potential faults in collaborative operations.

[0025] In one embodiment of the present invention, S412 includes: The generated discrete data fragments are classified and organized, and the data are divided into groups according to the robot's behavior type and state change dimension to generate a classified data set; Select the corresponding groups in the classification dataset, conduct a two-way comparison of behavioral trajectories and state changes for each group, and mark the nodes of synchronous change between the two; Analyze the temporal sequence of synchronous change nodes to track the transmission path and magnitude of state fluctuations after the robot's actions are triggered; By summarizing the transmission paths and change characteristics of each data group, the core correlation rules between action triggering and state fluctuations are extracted; the core correlation rules are integrated to form behavioral state correlation features that can accurately reflect the relationship between robot behavior and state.

[0026] The working principle and effects of the above technical solution are as follows: By classifying and organizing discrete data fragments, grouping them according to robot behavior type and state change dimension, a categorized data set is generated, making the data more organized, avoiding bias in correlation analysis caused by mixed data, and reducing the time spent on ineffective analysis. By selecting corresponding data groups, bidirectional comparison of behavior and state is carried out group by group, and synchronous change nodes are marked, improving the pertinence of correlation analysis, avoiding the omission of key correlation information, and ensuring that the core linkage points of behavior and state are captured. By analyzing the time sequence of synchronous change nodes, the transmission path and amplitude of state fluctuations after action triggering are tracked, and the transmission logic of the two is clearly understood, avoiding the problem of not being able to clarify the causal relationship. By summarizing the transmission path and change characteristics, extracting the core correlation rules, and integrating them to form behavior-state correlation features, it can not only accurately reflect the correlation between robot behavior and state, but also provide a reliable basis for subsequent iterative deduction, reduce the blindness of rule mining, and further improve the accuracy of subsequent collaborative strategy mining.

[0027] In one embodiment of the present invention, step S5 includes: S51. Based on the optimized collaborative decision-making model, generate online evolutionary instructions that are adapted to the operation of multiple robots on site, plan the instruction execution order and transmission cycle, and ensure the executability of the instructions; S52. The online evolution instructions are synchronously transmitted to the multi-robot real-time interaction network to cover and update the original dynamic collaborative logic scripts in the network, and complete the online iteration of the logic rules. S53. Drive multiple robots to carry out actual collaborative operations on site through the updated dynamic collaborative logic script, guide each robot to execute actions according to the optimized logic, and cooperate to complete the task. S54. Continuously collect on-site equipment status data during collaborative operations, including real-time torque, load changes and location information, and transmit the collected data back to the unified status perception middleware. S55. Based on the feedback data, a closed-loop cognitive architecture is formed, and the work process is continuously iterated and optimized. Finally, the entire process data is summarized to generate a collaborative work evaluation report that includes risk level, efficiency improvement value and debugging cost savings.

[0028] The working principle and effects of the above technical solution are as follows: Online evolutionary instructions are generated based on an optimized collaborative decision-making model, planning the execution sequence and transmission cycle to ensure smooth instruction execution, avoiding job interruptions caused by unreasonable instructions, and reducing instruction transmission delays or congestion. By synchronously transmitting online evolutionary instructions to the interactive network, the original dynamic collaborative logic script is updated, completing logic iteration, enhancing the adaptability of the collaborative logic, and avoiding the problem of old scripts being unable to adapt to the optimized strategy. The updated script drives multiple robots to carry out on-site operations, guiding each robot to cooperate precisely, improving collaborative operation efficiency, and reducing malfunctions caused by action deviations. By continuously collecting and transmitting on-site status data, a closed-loop cognitive architecture is formed to continuously optimize the process. Data is summarized to generate evaluation reports, enabling real-time monitoring of the operation status, continuous improvement of operation performance, clear presentation of operation results, reduced subsequent debugging costs, and avoidance of resource waste caused by blind optimization, further ensuring the stability and reliability of multi-robot collaborative operations.

[0029] In one embodiment of the present invention, S51 includes: Import the optimized collaborative decision-making model, extract the core decision-making logic and operating parameters from the model, and generate instructions to generate basic data. Based on the operating status and operational requirements of multiple robots on site, the basic data is converted into instructions to generate initial online evolution instructions; The execution order of the initial online evolution commands is planned, and the triggering conditions and sequential relationships of each command are clarified. Set the instruction transmission cycle to match the transmission rate of the multi-robot real-time interaction network to avoid instruction transmission delays or congestion; perform integrity verification on the planned online evolution instructions, correct logical loopholes in the instructions, and ensure that the instructions can be executed smoothly.

[0030] The working principle and effects of the above technical solution are as follows: By importing the optimized collaborative decision-making model, core decision-making logic and operating parameters are extracted to generate basic data, ensuring the scientific nature of instruction generation, avoiding instruction failure due to deviations in basic data, and reducing the time spent on subsequent instruction adjustments. By combining the on-site robot operating status and operational requirements, the basic data is transformed into initial online evolutionary instructions, making the instructions more aligned with the actual operational scenario and avoiding the problem of instructions being out of sync with on-site conditions and unable to be executed. By planning the instruction execution sequence, the triggering conditions and connection relationships of each instruction are clarified, enhancing the systematic nature of instruction execution and reducing robot action conflicts caused by instruction confusion. By setting an instruction transmission cycle adapted to the network transmission rate, instruction transmission delays or congestion are avoided. By performing integrity verification and correcting logical loopholes in the instructions, smooth execution of the instructions is guaranteed. This not only ensures the adaptability and executability of online evolutionary instructions but also provides reliable support for subsequent logic script updates and on-site operations, further reducing the probability of operational failures.

[0031] One embodiment of the present invention provides a robot collaborative programming and simulation system that supports cross-device state linkage, comprising: One or more processors; Memory, used to store one or more programs; Wherein, when the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method described in any one of the above.

[0032] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A robot collaborative programming and simulation method supporting cross-device state linkage, characterized in that, The method includes: S1. Construct a unified state perception middleware to parse the private communication protocols of robots from different brands, extract key parameters, and generate standardized equipment status data; deploy a multi-robot real-time interactive network based on the standardized equipment status data. S2. Based on the standardized device status data in the multi-robot real-time interaction network, construct a state and behavior coupling diagram to describe the state associations and collaborative constraints between robots; generate a dynamic collaborative logic script through the state and behavior coupling diagram to replace the preset timing script. S3. Build a high-fidelity disturbance simulation environment based on dynamic collaborative logic scripts, integrate physical disturbance models, and simulate complex working conditions in robot collaborative operations; run dynamic collaborative logic scripts in the high-fidelity disturbance simulation environment to generate collaborative operation simulation data; S4. Through inverse reinforcement learning, policy mining is performed on the collaborative operation simulation data to generate an interpretable collaborative decision-making model, revealing the causal relationship between robot behavior selection and state changes; the interpretable collaborative decision-making model is then transformed into a visualized policy graph. S5. Generate online evolution instructions based on the optimized collaborative decision-making model, and synchronously update the dynamic collaborative logic script in the multi-robot real-time interaction network; drive multiple robots to perform actual collaborative operations through online evolution instructions, and continuously collect on-site equipment status data to feed back to the unified status perception middleware to form a closed-loop cognitive architecture, and finally generate a collaborative operation evaluation report.

2. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 1, characterized in that, S1 includes: S11. Build a unified state perception middleware to adapt and connect the underlying communication interfaces of robots from different brands, and parse the message structure and data rules of various private communication protocols layer by layer. S12. Traverse the parsed protocol data stream, extract key parameters from the robot's operation feedback information, and remove redundant and invalid protocol auxiliary information; S13. The extracted parameters are formatted and dimension unified, and the data encoding method and numerical measurement unit are unified to generate standardized equipment status data with universal interactive capabilities. S14. Based on standardized equipment status data, build a real-time interactive network for multiple robots, plan data transmission paths and node allocation rules, and open up data interaction channels between various devices. S15. In the completed interactive network, complete the synchronization, sharing and dynamic updating of cross-device status data, and continuously refresh the device status information of each terminal.

3. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 1, characterized in that, The S2 includes: S21. Retrieve standardized equipment status data continuously transmitted in the multi-robot real-time interaction network, statistically analyze the operating posture and operation progress of each robot, and sort out the operational relationships between robots. S22. Combining the relationships obtained from the analysis with the constraints of the actual operation scenario, construct a state and behavior coupling diagram to fully depict the state relationships and constraint logic in robot collaborative operation. S23. Generate dynamic collaborative logic scripts through the association rules of the state and behavior coupling diagram, abandoning the traditional fixed-time preset scripts, and allowing the script content to have adjustable space as the device state changes; S24. Monitor various state change information during robot operation in real time, focusing on capturing abnormal states and ensuring complete collection of change information; S25. Based on the collected state change information, adjust the collaborative execution strategy in the dynamic collaborative logic script in real time to adapt to sudden changes in on-site operations and maintain the stability of multi-robot collaborative operations.

4. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 3, characterized in that, The S25 includes: S251. Receive the acquired state mutation content, complete the matching and association between the mutation information and the collaborative execution logic, and generate the collaborative strategy pointing information to be adjusted. S252. Based on the information pointing to the collaborative strategy to be adjusted, locate the collaborative execution nodes that need to be changed in the dynamic collaborative logic script, and filter out the execution logic of the corresponding nodes. S253. Perform parameter reallocation and process rearrangement on the selected execution logic to generate new collaborative execution content that adapts to the current mutation scenario; S254. Write the new collaborative execution content into the dynamic collaborative logic script and complete the replacement, thus completing the real-time update of the collaborative execution strategy; S255. Apply the updated collaborative execution strategy to guide multiple robots to carry out operations, adapt to sudden changes in on-site operations, and maintain the stable operation of multi-robot collaborative operations.

5. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 4, characterized in that, S253 includes: Extract all the running parameters within the filtered execution logic, combine them with the characteristics of state mutation to redistribute the parameter values, and generate a parameter configuration set adapted to the mutation scenario. Based on the updated parameter configuration set, the execution steps of the original collaborative operation are broken down, and the execution order and connection relationship of each step are adjusted. The adjusted execution steps are arranged in a time sequence, and the action intervals and connection times of each robot in the corresponding steps are planned; the rearranged execution steps and time sequence results are integrated to improve the interactive constraints of collaborative operations. Integrate parameter configuration and process orchestration results to form new collaborative execution content adapted to the current mutation scenario.

6. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 1, characterized in that, The S3 includes: S31. Using dynamic collaborative logic scripts as the core execution framework, build a high-fidelity disturbance simulation environment adapted to multi-robot collaborative operations, and plan the simulation scene size and operation area layout. S32. Integrate physical disturbance models in the simulation environment to restore the physical laws of real-world scenarios and simulate complex working conditions in robot collaborative operations. S33. Load the dynamic collaborative logic script into the completed high-fidelity disturbance simulation environment and run it continuously to fully reproduce the entire process of multi-robot collaborative operation. S34. During the simulation process, collect multi-dimensional simulation indicators in all aspects and record the trend of indicator changes at different time periods. S35. Integrate the various simulation indicators and deduction process information collected, classify and summarize them to generate complete collaborative operation simulation data.

7. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 1, characterized in that, The S4 includes: S41. Use inverse reinforcement learning to perform deep policy mining on collaborative operation simulation data, analyze the intrinsic relationship between robot behavior and state changes during the simulation process, and extract potential collaborative behavior patterns. S42. Based on the behavioral patterns obtained from mining, construct an interpretable collaborative decision-making model to clearly reveal the causal relationship between robot behavior choices and state changes; S43. Transform the internal logic and decision-making path of the collaborative decision-making model into a visual strategy map to present the collaborative rules in an intuitive form; S44. Locate the weak links in the collaborative logic through a visualized strategy map, and manually adjust and optimize the key operating parameters in the decision-making model. S45. Save and solidify the optimized collaborative decision-making model, and eliminate unreasonable logic in the model.

8. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 1, characterized in that, The S5 includes: S51. Based on the optimized collaborative decision-making model, generate online evolutionary instructions that are adapted to the operation of multiple robots on site, plan the instruction execution order and transmission cycle, and ensure the executability of the instructions; S52. The online evolution instructions are synchronously transmitted to the multi-robot real-time interaction network to cover and update the original dynamic collaborative logic scripts in the network, and complete the online iteration of the logic rules. S53. Drive multiple robots to carry out actual collaborative operations on site through the updated dynamic collaborative logic script, guide each robot to execute actions according to the optimized logic, and cooperate to complete the task. S54. Continuously collect on-site equipment status data during collaborative operations and transmit the collected data back to the unified status perception middleware. S55. Based on the feedback data, a closed-loop cognitive architecture is formed, the work process is continuously iterated and optimized, and finally the entire process data is summarized to generate a collaborative work evaluation report.

9. The robot collaborative programming and simulation method supporting cross-device state linkage according to claim 8, characterized in that, S51 includes: Import the optimized collaborative decision-making model, extract the core decision-making logic and operating parameters from the model, and generate instructions to generate basic data. Based on the operating status and operational requirements of multiple robots on site, the basic data is converted into instructions to generate initial online evolution instructions; The execution order of the initial online evolution commands is planned, and the triggering conditions and sequential relationships of each command are clarified. Set the instruction transmission cycle to match the transmission rate of the multi-robot real-time interaction network to avoid instruction transmission delays or congestion; perform integrity verification on the planned online evolution instructions, correct logical loopholes in the instructions, and ensure that the instructions can be executed smoothly.

10. A robot collaborative programming and simulation system supporting cross-device state linkage, comprising: One or more processors; Memory, used to store one or more programs; Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 9.

Images