Fault self-diagnosis system and method for robot

Abstract

The invention discloses a robot fault self-diagnosis system and method, relates to the technical field of robot fault diagnosis, and solves the technical problem of low automation degree of a robot caused by incapability of accurate self-diagnosis and self-repair of the robot in the prior art. The method comprises the following steps: collecting real-time operation data of a robot, and extracting real-time error logs and historical fault data from a database; the fault state of the robot is monitored according to the real-time operation data of the robot, a historical fault library is constructed based on historical fault data, the fault state is input into the historical fault library for matching, and mismatched faults, matched faults and corresponding matching repair methods are obtained. Different self-repairing methods are adopted for different faults, the historical fault library is updated, the repairing effect is monitored, and the matched repairing method is updated according to the repairing effect; and when the robot cannot complete self-repair, planning an optimal maintenance path and controlling to go to repair.

Description

Technical Field

[0001] The invention belongs to the field of robot fault diagnosis, relates to database technology, and specifically provides a robot fault self-diagnosis system and method. Background Art

[0002] In the field of modern industrial automation, robots perform various precise and repetitive tasks on the production line, greatly improving production efficiency and product quality. As an indispensable member of the production line, robots undertake a variety of tasks such as precision assembly, welding, spraying, and handling. Their efficiency and consistency are difficult to match with traditional manual labor; however, it is precisely because the robot system integrates multiple complex subsystems such as mechanics, electronics, control, and sensing that the failure of any single component may cause abnormal operation of the entire system, thereby affecting the smoothness of the production process and the quality of the product.

[0003] The invention patent with application number CN2012102930843 discloses a robot vacuum cleaner and its self-diagnosis method. The invention performs self-diagnosis according to the needs of the user before use, and detects its own status and whether it can work according to the results of the self-diagnosis. The robot in the invention can only realize self-status diagnosis. It is difficult to locate the cause of the fault when a fault occurs, and there is no built-in fault repair program, which makes the user unable to do anything after the robot fails, affecting the overall work efficiency of the robot, and the robot's degree of automation also needs to be improved.

[0004] The present invention provides a robot fault self-diagnosis system and method to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a robot fault self-diagnosis system and method, which are used to solve the technical problem in the prior art that the robot is unable to accurately self-diagnose and self-repair, resulting in a low degree of automation of the robot.

[0006] To achieve the above-mentioned object, the first aspect of the present invention provides a robot fault self-diagnosis system and method, comprising: a database, a data collection module, a fault diagnosis module, a self-repair module and a path planning module;

[0007] Data collection module: used to collect real-time operation data, real-time error logs and historical fault data of the robot during operation;

[0008] Fault diagnosis module: used to monitor the fault status of the robot based on real-time operation data; build a historical fault library based on historical fault data; trigger fault diagnosis when a fault status occurs; input the fault status into the historical fault library for matching to obtain matching faults, corresponding matching repair methods and mismatching faults; analyze mismatching faults based on real-time error logs to obtain mismatching fault data;

[0009] Self-repair module: used to implement the corresponding matching repair method for the matching fault and detect the matching repair effect; perform repair attempts on the mismatching fault according to the mismatching fault data and detect the mismatching repair effect, and enter the repair attempt method into the historical fault library for update based on the mismatching repair effect;

[0010] Path planning module: When the fault cannot be repaired automatically, it is used to automatically plan the maintenance path to the maintenance site based on the real-time position in the robot's real-time operation data and the workshop road map in the database.

[0011] Preferably, monitoring the fault status of the robot according to the real-time operation data comprises:

[0012] Extracting real-time operation data; extracting preset data from the control system database; the real-time operation data includes the current position, current action and current electrical data of the robot; the preset data includes a preset path, a preset action and a preset electrical range;

[0013] Determine whether the current position of the robot meets the preset position on the preset path; if yes, mark the position state of the robot as a normal position; if no, mark the position data of the robot as an abnormal position;

[0014] Determine whether the current action of the robot is consistent with the preset action; if yes, mark the action state of the robot as normal action; if no, mark the action state of the robot as abnormal action;

[0015] Determine whether the current electrical data of the robot meets the preset electrical range; if yes, mark the electrical state of the robot as normal electrical; if no, mark the electrical state of the robot as abnormal electrical;

[0016] Mark abnormal positions, abnormal actions, and abnormal electrical conditions as fault conditions.

[0017] Preferably, the building of a historical fault library based on historical fault data includes:

[0018] Extracting historical fault data; the historical fault data includes historical fault states that have occurred in the robot, corresponding historical repair methods, and corresponding error repair methods;

[0019] An initial database is constructed using a database management system, and historical fault data is input into the initial database for recording to obtain a historical fault library.

[0020] Preferably, the step of inputting the fault status into a historical fault library for matching to obtain matching faults, corresponding matching repair methods and mismatching faults includes:

[0021] Extracting fault status and historical fault library; the historical fault library includes historical fault status, historical fault type and corresponding historical repair method in historical fault data;

[0022] Determine whether the historical fault state contains a fault state; if yes, mark the fault state of the corresponding robot as a matching fault; if no, mark the fault state of the corresponding robot as a non-matching fault;

[0023] The historical repair method corresponding to the matching fault is marked as the matching repair method.

[0024] Preferably, the step of analyzing the mismatch fault according to the real-time error log to obtain mismatch fault data includes:

[0025] Extract mismatch failures and real-time error logs;

[0026] Check the real-time error log to get the error code;

[0027] Determine whether the error code matches the mismatch fault; if yes, mark the corresponding error code as mismatch fault data; if no, mark the electrical data corresponding to the mismatch fault as mismatch fault data.

[0028] Preferably, the step of implementing a corresponding matching repair method for the matching failure and detecting the matching repair effect includes:

[0029] Extract matching repair methods and implement them;

[0030] Control the robot to re-complete the work task when the failure occurred, and continue to monitor whether the robot can complete the work task normally; if the robot fails again, the matching repair effect of the corresponding matching repair method will be marked as error repair, the error repair will be updated and corrected, and the historical fault library will be updated according to the correction result; otherwise, the matching repair effect of the corresponding matching repair method will be marked as correct repair.

[0031] Preferably, updating and correcting the error repair and updating the historical fault library according to the correction result includes:

[0032] Extraction bug fixes;

[0033] According to the fault status corresponding to the error repair, detect the error repair and mark the corresponding fault status as fault data, detect the error code in the error log, if it contains an error code, try to repair the error code, mark the repaired error code as the correct code, implement the correct code and monitor the repair effect, if the robot can complete the task, record the correct code in the historical fault library and replace the matching repair method corresponding to the error repair;

[0034] If no error code is included, the current electrical data of the robot is detected, the current electrical data is corrected, and the corrected electrical data is recorded in the historical fault library to replace the matching repair method.

[0035] Preferably, the step of attempting to repair the mismatch fault according to the mismatch fault data and detecting the mismatch repair effect includes:

[0036] Extract mismatch fault data;

[0037] When the mismatch fault data contains an error code, try to retrieve the error code and make a repair attempt on the software logic according to the error code, control the robot to complete the work task again, monitor the robot's task completion status, and if a fault occurs, mark the mismatch repair effect of the corresponding repair attempt method as abnormal repair; otherwise, mark the mismatch repair effect of the corresponding repair attempt method as normal repair;

[0038] When the mismatch fault data contains the current electrical data, adjust the current electrical data to meet the preset electrical range, control the robot to complete the work task again, monitor the robot's task completion status, and if a fault occurs, mark the mismatch repair effect of the corresponding repair attempt method as an abnormal repair; otherwise, mark the mismatch repair effect of the corresponding repair attempt method as a normal repair.

[0039] Preferably, the method of automatically planning a maintenance route to a maintenance location according to the real-time position in the real-time operation data of the robot and a workshop road map in a database includes:

[0040] Extract the real-time position of the robot from the real-time operation data, and extract the route map of the work shop and the location of the maintenance place from the control system database;

[0041] The route map and the position of the robot are input into MATLAB to obtain several driving paths for the robot to reach the maintenance site. The driving paths are sorted according to their lengths, and the driving path with the shortest length is marked as the maintenance path.

[0042] A robot fault self-diagnosis method, applied to a robot fault self-diagnosis system according to the first aspect, is characterized by comprising:

[0043] Step 1: Collect the robot's real-time operation data, real-time error logs, and historical fault data;

[0044] Step 2: Build a historical fault library based on historical fault data; monitor the fault status of the robot according to the real-time operation data, input the fault status into the historical fault library for matching to obtain matching faults, corresponding matching repair methods and mismatching faults; analyze mismatching faults according to the real-time error log to obtain mismatching fault data;

[0045] Step 3: Detect the matching repair effect of the matching repair method; perform a repair attempt on the mismatching fault according to the mismatching fault data and detect the mismatching repair effect, and enter the repair attempt method into the historical fault database for updating based on the matching repair effect and the mismatching repair effect;

[0046] Step 4: When the robot cannot automatically repair itself, a maintenance path is generated based on the robot's real-time position in the real-time operating status and the route map in the database, and the robot goes for maintenance along the maintenance path.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. The present invention collects the real-time operation data of the robot, and extracts the real-time error log and historical fault data from the database; monitors the fault state of the robot according to the real-time operation data of the robot, can timely discover the fault of the robot, and make the robot enter the self-check state; builds a historical fault library according to the historical fault data, and inputs the fault state of the robot into the historical fault library for matching, obtains matching faults, corresponding matching repair methods and non-matching faults, can quickly match the faults that have occurred, reduces the time of fault diagnosis, speeds up the maintenance process, and reduces downtime; for matching faults, implements corresponding matching repair methods, can reduce the time of fault diagnosis and repair, and improve production efficiency; for non-matching faults, the robot performs different repair attempts for different types of non-matching fault data, records the successful repair attempt methods into the historical fault library, can learn and accumulate new fault handling knowledge, expand its fault handling capability range, so that the robot can have more reference repair strategies when encountering similar but not completely the same faults in the future, and enhances the self-learning and adaptability of the system.

[0049] 2. When implementing the matching repair method corresponding to the matching fault, the present invention monitors the repair effect and updates the matching repair method according to the repair effect. The dynamically updated fault library can be continuously improved to improve the system's self-repair capability. The system can also gradually eliminate known unstable factors, improve the overall operational reliability, and reduce the frequency and severity of faults. When attempting to repair mismatched faults, different repair attempts are made for different types of mismatched fault data, and the repair attempt methods that can eliminate the faults are recorded in the historical fault library, which can enrich the types in the historical fault library and dynamically update the historical fault library, so that the fault library can be continuously optimized to maintain its relevance and practicality, and more fault handling skills can be learned and mastered to improve one's own problem-solving capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 The following is a flowchart of an embodiment of the present invention.

[0052] Figure 2 It is a system composition diagram of the present invention.

[0053] Figure 3 A complete flowchart for obtaining the robot fault status in one embodiment of the present invention. DETAILED DESCRIPTION

[0054] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] See also Figure 1-Figure 3 , the first aspect of the present invention provides a robot fault self-diagnosis system and method, including: a data collection module, a fault diagnosis module and a self-repair module;

[0056] Data collection module: used to collect real-time operation data, real-time error logs and historical fault data of the robot while it is working.

[0057] Exemplarily, the current position, current action, current current, voltage and power of the transport robot in the workshop are collected through the sensor network and the control system panel; the real-time updated error log is extracted from the background program panel; the fault status, historical fault type and corresponding historical repair method that occurred in the historical work of the robot in the workshop are extracted from the control system database.

[0058] Fault diagnosis module: used to monitor the fault status of the robot based on real-time operation data; build a historical fault library based on historical fault data; trigger fault diagnosis when a fault status occurs; input the fault status into the historical fault library for matching to obtain matching faults, corresponding matching repair methods and mismatching faults; analyze mismatching faults according to real-time error logs to obtain mismatching fault data.

[0059] Exemplarily, the robot preset path, preset action and preset electrical range are extracted from the control system database; the current position and current time of the robot are detected by the positioning system, the preset position expected to be reached at the current time of the preset path is matched, and it is judged whether the current position of the robot meets the preset position on the preset path, and the current position of the robot is judged to be the same as the preset position, and the position state of the robot is marked as a normal position; it is judged whether the current action of the robot at the current position is consistent with the preset action of the preset position, and the current action of the robot is judged to be consistent with the preset action, and the action state of the robot is marked as a normal action; the preset electrical range in this embodiment includes a preset current range, a preset voltage range and a preset power range; the preset current range is 50A to 200A, the preset voltage range is 380V to 480V, and the preset power range is 5kW to 20kW; the current electrical data of the robot is 100A current, 400V voltage, and 22kW power; it is judged whether the current electrical data of the robot meets the preset electrical range, and if it is judged that the current electrical data of the robot does not meet the preset electrical range, the electrical state of the robot is marked as abnormal electrical, and the fault state of the robot is abnormal electrical.

[0060] A historical fault library is constructed based on historical fault data, which includes historical fault states, historical fault types and corresponding historical repair methods in the historical fault data. The historical fault library stores faults that have occurred in robots in historical workshops, corresponding solutions, and error detection methods. The fault state of the robot is input into the historical fault library to determine whether the historical fault state contains the fault state. If it is determined that the historical fault library does not contain the current fault state of the robot, the current fault state of the robot is marked as an unmatched fault.

[0061] Self-repair module: used to implement the corresponding matching repair method for the matching fault and detect the matching repair effect; attempt to repair the mismatching fault according to the mismatching fault data and detect the mismatching repair effect, and enter the repair attempt method into the historical fault library for updating based on the mismatching repair effect.

[0062] Exemplarily, check the error code in the real-time error log in the control system program to determine whether the error code is consistent with the mismatch fault. If it is determined that the error code is inconsistent with the matching fault, the electrical data corresponding to the mismatch fault is marked as mismatch fault data; if the mismatch fault data is electrical data, and specifically the power does not meet the preset electrical range, then reduce the power supplied to the robot to make it meet the preset electrical range; control the robot to re-perform the node task of the current work, monitor the real-time operation data of the robot reaching the fault position, re-analyze the fault state of the robot, and mark the real-time operation data of the robot reaching the fault position again as secondary operation data, the secondary operation data including secondary position, secondary action and secondary electrical data; determine the secondary operation of the robot Whether the secondary position is consistent with the preset position on the preset path, it is judged that the secondary position of the robot is consistent with the preset position on the preset path, and the position state of the robot is marked as normal position; whether the secondary action of the robot is consistent with the preset action, it is judged that the secondary action is consistent with the preset action, and the action state of the robot is marked as normal action; the electrical data of the robot after adjustment is 90 amps, 390 volts, and 18 kilowatts. It is judged whether the current electrical data of the robot is consistent with the preset electrical range, and it is judged that the adjusted electrical data of the robot is consistent with the preset electrical range, then the electrical state of the robot is marked as normal electrical; the adjustment of electrical data is used as a repair attempt method, and after judgment, the mismatch repair effect of the adjusted electrical data is marked as normal repair.

[0063] In some other embodiments, when a robot fails and the failure can be matched in a historical fault library, the robot can directly use the corresponding matching repair method in the historical fault library to attempt a repair and monitor the repair effect; if the repair effect can eliminate the robot's failure and enable it to complete the work task intact, the matching repair effect of the corresponding matching repair method will be marked as a correct repair; if the repair effect cannot eliminate the robot's failure, it may mean that the previously recorded method is no longer applicable to the current working situation due to robot updates or program updates, then the matching repair effect of the corresponding matching repair method will be marked as an error repair, the matching repair method corresponding to the error repair will be updated and the updated repair effect will be monitored, and when the updated repair method can eliminate the robot's failure, the updated repair method will be recorded in the historical fault library to replace the matching repair method corresponding to the error repair.

[0064] In other embodiments, when the unmatched fault data contains an error code, the robot program retrieves the error code and attempts to repair the error code, so that the robot can automatically correct the current fault according to a preset path and preset action, so that it can eliminate the fault and complete the task.

[0065] Path planning module: When the fault cannot be repaired automatically, it is used to automatically plan the maintenance path to the maintenance site based on the real-time position in the robot's real-time operation data and the workshop road map in the database.

[0066] For example, when the robot cannot repair a fault by itself, the current position of the robot, the working workshop road and the location of the maintenance place are input into the MATLAB software for path planning, and several paths are obtained. The path with the shortest path length is selected as the maintenance path for the robot to go to the maintenance place for maintenance.

[0067] Working principle of the present invention:

[0068] The present invention collects the real-time operation data of the robot, the real-time error log of the system program and the historical fault data in the database; analyzes the real-time operation data, monitors the fault state of the robot, builds a historical fault library based on the historical fault data, inputs the fault state into the historical fault library for matching, and obtains the mismatch fault, the matching fault and the corresponding matching repair method; if the robot corresponds to a matching fault, implements the matching repair method and monitors the repair effect, and updates the matching repair method according to the repair effect; if the robot corresponds to a mismatch fault, performs different repair attempts according to the type of the mismatch fault, and records the repair attempt method that successfully eliminates the fault in the historical fault library; when the robot cannot complete self-repair, inputs the circuit diagram and the position of the robot into MATLAB to obtain the maintenance path of the robot to the maintenance site.

[0069] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A robot fault self-diagnosis system, characterized in that: include: Database, data collection module, fault diagnosis module, self-repair module and path planning module; Data collection module: used to collect real-time operation data, real-time error logs and historical fault data of the robot during operation; Fault diagnosis module: used to monitor the fault status of the robot based on real-time operation data; Build a historical fault database based on historical fault data; trigger fault diagnosis when a fault state occurs; Input the fault status into the historical fault library for matching to obtain matching faults, corresponding matching repair methods and non-matching faults; Analyze the mismatch fault according to the real-time error log to obtain mismatch fault data; Self-repair module: used to implement the corresponding matching repair method for matching failures and detect the matching repair effect; Attempt to repair the mismatch fault and detect the mismatch repair effect, and enter the repair attempt method into the historical fault database for updating based on the matching repair effect and the mismatch repair effect; Path planning module: When the fault cannot be repaired automatically, it is used to automatically plan the maintenance path to the maintenance site based on the real-time position in the robot's real-time operation data and the workshop road map in the database.

2. A robot fault self-diagnosis system according to claim 1, characterized in that: The monitoring of the fault status of the robot according to the real-time operation data includes: Extracting real-time operation data; extracting preset data from the control system database; the real-time operation data includes the current position, current action and current electrical data of the robot; the preset data includes a preset path, a preset action and a preset electrical range; Determine whether the current position of the robot meets the preset position on the preset path; if yes, mark the position state of the robot as a normal position; if no, mark the position data of the robot as an abnormal position; Determine whether the current action of the robot is consistent with the preset action; if yes, mark the action state of the robot as normal action; if no, mark the action state of the robot as abnormal action; Determine whether the current electrical data of the robot meets the preset electrical range; if yes, mark the electrical state of the robot as normal electrical; if no, mark the electrical state of the robot as abnormal electrical; Mark abnormal positions, abnormal actions, and abnormal electrical conditions as fault conditions.

3. A robot fault self-diagnosis system according to claim 1, characterized in that: The building of a historical fault database based on historical fault data includes: Extracting historical fault data; the historical fault data includes historical fault states that have occurred in the robot, corresponding historical repair methods, and corresponding error repair methods; An initial database is constructed using a database management system, and historical fault data is input into the initial database for recording to obtain a historical fault library.

4. A robot fault self-diagnosis system according to claim 1, characterized in that: The step of inputting the fault status into the historical fault library for matching to obtain matching faults, corresponding matching repair methods and mismatching faults includes: Extracting fault status and historical fault library; the historical fault library includes historical fault status, historical fault type and corresponding historical repair method in historical fault data; Determine whether the historical fault state contains a fault state; if yes, mark the fault state of the corresponding robot as a matching fault; if no, mark the fault state of the corresponding robot as a non-matching fault; The historical repair method corresponding to the matching fault is marked as the matching repair method.

5. A robot fault self-diagnosis system according to claim 1, characterized in that: The step of analyzing the mismatch fault according to the real-time error log to obtain mismatch fault data includes: Extract mismatch failures and real-time error logs; Check the real-time error log to get the error code; Determine whether the error code matches the mismatch fault; if yes, mark the corresponding error code as mismatch fault data; if no, mark the electrical data corresponding to the mismatch fault as mismatch fault data.

6. A robot fault self-diagnosis system according to claim 1, characterized in that: The implementing a corresponding matching repair method for the matching failure and detecting the matching repair effect includes: Extract matching repair methods and implement them; Control the robot to re-complete the work task when the failure occurred, and continue to monitor whether the robot can complete the work task normally; if the robot fails again, the matching repair effect of the corresponding matching repair method will be marked as error repair, the error repair will be updated and corrected, and the historical fault library will be updated according to the correction result; otherwise, the matching repair effect of the corresponding matching repair method will be marked as correct repair.

7. A robot fault self-diagnosis system according to claim 1, characterized in that: The updating and correcting of the error repair and updating of the historical fault library according to the correction result include: Extraction bug fixes; According to the fault status corresponding to the error repair, detect the error repair and mark the corresponding fault status as fault data, detect the error code in the error log, if it contains an error code, try to repair the error code, mark the repaired error code as the correct code, implement the correct code and monitor the repair effect, if the robot can complete the task, record the correct code in the historical fault library and replace the matching repair method corresponding to the error repair; If no error code is included, the current electrical data of the robot is detected, the current electrical data is corrected, and the corrected electrical data is recorded in the historical fault library to replace the matching repair method.

8. A robot fault self-diagnosis system according to claim 1, characterized in that: The step of attempting to repair the mismatch fault and detecting the mismatch repair effect includes: Extract mismatch fault data; When the mismatch fault data contains an error code, try to retrieve the error code and make a repair attempt on the software logic according to the error code, control the robot to complete the work task again, monitor the robot's task completion status, and if a fault occurs, mark the mismatch repair effect of the corresponding repair attempt method as abnormal repair; otherwise, mark the mismatch repair effect of the corresponding repair attempt method as normal repair; When the mismatch fault data contains the current electrical data, adjust the current electrical data to meet the preset electrical range, control the robot to complete the work task again, monitor the robot's task completion status, and if a fault occurs, mark the mismatch repair effect of the corresponding repair attempt method as an abnormal repair; otherwise, mark the mismatch repair effect of the corresponding repair attempt method as a normal repair.

9. A robot fault self-diagnosis system according to claim 1, characterized in that: The method of automatically planning a maintenance route to a maintenance location based on the real-time position of the robot in the real-time operation data and the workshop road map in the database includes: Extract the real-time position of the robot from the real-time operation data, and extract the route map of the work shop and the location of the maintenance place from the control system database; The route map and the position of the robot are input into MATLAB to obtain several driving paths for the robot to reach the maintenance site. The driving paths are sorted according to their lengths, and the driving path with the shortest length is marked as the maintenance path.

10. A robot fault self-diagnosis method, used for running a robot fault self-diagnosis system according to any one of claims 1 to 9, characterized in that: include: Step 1: Collect the robot's real-time operation data, real-time error logs, and historical fault data; Step 2: Build a historical fault database based on historical fault data; Monitor the fault status of the robot according to the real-time operation data, input the fault status into the historical fault library for matching to obtain matching faults, corresponding matching repair methods and mismatching faults; analyze the mismatching faults according to the real-time error log to obtain mismatching fault data; Step 3: Detect the matching repair effect of the matching repair method; Perform repair attempts on the mismatch faults according to the mismatch fault data and detect the mismatch repair effect, and enter the repair attempt method into the historical fault database for updating based on the match repair effect and the mismatch repair effect; Step 4: When the robot cannot automatically repair itself, a maintenance path is generated based on the robot's real-time position in the real-time operating status and the route map in the database, and the robot goes for maintenance along the maintenance path.

Images