CN115056228A - Robot abnormity monitoring and processing system and method - Google Patents

Abstract

The invention provides a system and a method for monitoring and processing the abnormity of a robot, wherein the system comprises: the system comprises a construction module, a task execution module and a task execution module, wherein the construction module is used for constructing an abnormity monitoring library corresponding to a first task when the first robot executes the first task; the monitoring module is used for monitoring the abnormality of the first robot based on the abnormality monitoring library; the processing module is used for carrying out corresponding processing based on the result of the abnormity monitoring; and the guiding module is used for acquiring the processing condition when corresponding processing is carried out, and carrying out corresponding processing guidance based on the processing condition. According to the system and the method for monitoring and processing the robot abnormity, a monitoring person does not need to analyze the operation parameters of the robot, the labor cost is reduced, when the processing person processes basic abnormity problems or complexity abnormity problems, processing guidance can be performed, especially some complexity abnormity problems can be performed, a professional does not need to wait for processing, and the abnormity solving efficiency is improved.

Description

Robot abnormity monitoring and processing system and method

Technical Field

The invention relates to the technical field of robots, in particular to a system and a method for monitoring and processing robot abnormity.

Background

At present, the abnormity monitoring of the robot during operation is mostly completed manually, and monitoring personnel analyze the operation parameters of the robot to realize abnormity monitoring, so that the labor cost is high. In addition, when the abnormity is monitored, some basic abnormity problems can be solved by abnormity processing personnel, but when some complicated abnormity problems exist, the abnormity processing personnel need to wait for professional personnel (such as after-sales engineers of robot manufacturers) to process the abnormity problems, so that the convenience is low, and the solution efficiency is influenced.

Therefore, a solution is needed.

Disclosure of Invention

The invention provides an abnormity monitoring and processing system and method of a robot, which do not need to analyze the operation parameters of the robot by monitoring personnel, reduce the labor cost, can process and guide when the processing personnel processes basic abnormity problems or complexity abnormity problems, particularly certain complexity abnormity problems, do not need to wait for professional personnel to process, and improve the abnormity solving efficiency.

The invention provides an abnormality monitoring and processing system of a robot, comprising:

the system comprises a construction module, a task execution module and a task execution module, wherein the construction module is used for constructing an abnormity monitoring library corresponding to a first task when the first robot executes the first task;

the monitoring module is used for monitoring the abnormality of the first robot based on the abnormality monitoring library;

the processing module is used for carrying out corresponding processing based on the result of the abnormity monitoring;

and the guiding module is used for acquiring the processing condition when corresponding processing is carried out, and carrying out corresponding processing guidance based on the processing condition.

Preferably, the building module builds an anomaly monitoring library corresponding to the first task, including:

acquiring a plurality of robot operation abnormity record items from a local and/or big data platform, wherein the robot operation abnormity record items comprise: at least one first abnormal item generated when the second robot executes the second task and the current first attribute information of the second robot;

performing a first difference analysis on the first task and the second task;

performing second difference analysis on the current first attribute information of the second robot and the current second attribute information of the first robot;

performing feature extraction on results of the first difference analysis and the second difference analysis based on a preset first feature extraction template to obtain a plurality of first feature values;

constructing a difference description vector based on the first characteristic value;

determining a value degree based on the difference description vector and a preset value recognition library;

if the value degree is larger than or equal to a preset value degree threshold value, at least one first abnormal item generated when the corresponding second robot executes the corresponding second task is used as a target to be warehoused;

and integrating and warehousing the targets to be warehoused to obtain an abnormal monitoring library corresponding to the first task, and completing construction.

Preferably, the monitoring module monitors the abnormality of the first robot based on an abnormality monitoring library, and includes:

acquiring operation parameters of a first robot;

acquiring a preset abnormal monitoring strategy corresponding to a first abnormal item in an abnormal monitoring library, wherein the abnormal monitoring strategy comprises the following steps: a parameter extraction strategy and an anomaly analysis strategy;

extracting target parameters from the operating parameters based on a parameter extraction strategy;

and performing anomaly analysis on the target parameters based on an anomaly analysis strategy to complete monitoring.

Preferably, the processing module performs corresponding processing based on the result of the anomaly monitoring, including:

when the result of the anomaly monitoring contains at least one second abnormal item, counting the total number of the second abnormal items;

if the total number is 1, acquiring a preset first processing strategy corresponding to the second abnormal item;

based on the first processing strategy, corresponding processing is carried out;

if the total number is not 1, performing feature extraction on the second abnormal item based on a preset second feature extraction template to obtain a plurality of second feature values;

constructing an anomaly description vector based on the second characteristic value;

determining an abnormal association relation between the at least two randomly selected second abnormal items based on the abnormal description vectors of the at least two randomly selected second abnormal items and a preset abnormal association recognition library;

acquiring an abnormal combination strategy corresponding to the abnormal association relation;

based on an abnormal combination strategy, performing abnormal combination on at least two second abnormal items which are selected correspondingly and randomly to obtain combined abnormal items;

acquiring a preset second processing strategy corresponding to the abnormal item combination and a preset third processing strategy corresponding to the second abnormal item which is not subjected to abnormal combination;

and performing corresponding processing based on the second processing strategy and the third processing strategy.

Preferably, the guidance module acquires the processing condition, and includes:

acquiring the processing condition of a processing person performing corresponding processing based on the timed reply of the person terminal;

and/or the presence of a gas in the gas,

acquiring a processing condition through a maintenance recorder worn by a processing person who performs corresponding processing;

and/or the presence of a gas in the gas,

and acquiring the processing condition recorded by the activated maintenance recording unit after the first robot enters the maintenance mode.

Preferably, the guidance module performs corresponding processing guidance based on the processing situation, and includes:

inputting the result of the abnormal processing and the processing condition into a preset processing defect identification model, and determining at least one processing defect item;

acquiring preset guide information corresponding to the defect item;

delivering the instruction information to a personnel terminal of a processing personnel and/or a worn maintenance recorder;

and/or the presence of a gas in the atmosphere,

acquiring a field image of a processing field for corresponding processing;

determining a first position and a first orientation of a face of a processing person based on the live image;

acquiring second positions and second orientations of a plurality of display devices in a preset range around the first position in the processing site;

constructing a first direction vector based on the first position and the first orientation;

constructing a second direction vector based on the second position and the second orientation;

and controlling the display equipment corresponding to the maximum included angle in the included angles of the second direction vector and the first direction vector to temporarily display the guide information.

Preferably, the abnormality monitoring and processing system for a robot further includes:

the docking module is used for docking the processing personnel with the expert personnel when the processing personnel performing corresponding processing inputs the expert personnel needing docking;

wherein, the butt joint module will handle personnel and expert's butt joint, include:

constructing an online meeting room;

processing personnel and expert personnel are accessed into an online meeting room;

classifying and grouping the processing conditions to obtain a plurality of grouped data of a first condition type;

constructing a total number of display partitions of a first case type in a common display area in an online conference room;

randomly mapping a plurality of grouped data of a first case type to each display subarea;

continuously acquiring a plurality of communication records generated by communication between a processing person and an expert in an online conference room;

establishing a time axis;

correspondingly setting the alternating current records on a time axis based on the generation time points of the alternating current records;

performing semantic extraction on the communication record in the latest preset time on the time axis to obtain at least one first semantic;

acquiring a preset trigger semantic library corresponding to a first condition type;

matching the first semantic with a second semantic in a trigger semantic library;

if the matching is in accordance with the preset trigger value corresponding to the matched second semantic;

if the sum of the trigger values of the accumulated calculation trigger values is greater than or equal to the preset trigger value and the threshold value, taking the corresponding first condition type as a second condition type;

acquiring preset trigger values and an amplification strategy library corresponding to the total number of the second condition types, and determining the trigger values and the amplification strategies corresponding to the second condition types;

and based on the amplification strategy, carrying out amplification processing on the display partition to which the grouped data corresponding to the second case type is mapped.

The invention provides an abnormity monitoring and processing method of a robot, which comprises the following steps:

step 1: when a first robot executes a first task, constructing an abnormality monitoring library corresponding to the first task;

step 2: monitoring the abnormality of the first robot based on the abnormality monitoring library;

and 3, step 3: based on the result of the abnormal monitoring, corresponding processing is carried out;

and 4, step 4: and when corresponding processing is carried out, acquiring the processing situation, and carrying out corresponding processing guidance based on the processing situation.

Preferably, in step 1, constructing an anomaly monitoring library corresponding to the first task includes:

acquiring a plurality of robot operation abnormity record items from a local and/or big data platform, wherein the robot operation abnormity record items comprise: at least one first abnormal item generated when the second robot executes the second task and the current first attribute information of the second robot;

performing a first difference analysis on the first task and the second task;

performing second difference analysis on the current first attribute information of the second robot and the current second attribute information of the first robot;

performing feature extraction on results of the first difference analysis and the second difference analysis based on a preset first feature extraction template to obtain a plurality of first feature values;

constructing a difference description vector based on the first characteristic value;

determining a value degree based on the difference description vector and a preset value recognition library;

if the value degree is larger than or equal to a preset value degree threshold value, at least one first abnormal item generated when the corresponding second robot executes the corresponding second task is used as a target to be warehoused;

and integrating and warehousing the targets to be warehoused to obtain an abnormal monitoring library corresponding to the first task, and completing construction.

Preferably, step 2: based on the abnormal monitoring library, the abnormal monitoring is carried out on the first robot, and the abnormal monitoring comprises the following steps:

acquiring operation parameters of a first robot;

acquiring a preset abnormal monitoring strategy corresponding to a first abnormal item in an abnormal monitoring library, wherein the abnormal monitoring strategy comprises the following steps: a parameter extraction strategy and an anomaly analysis strategy;

extracting target parameters from the operating parameters based on a parameter extraction strategy;

and performing anomaly analysis on the target parameters based on an anomaly analysis strategy to complete monitoring.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

FIG. 1 is a schematic diagram of an anomaly monitoring and handling system for a robot in an embodiment of the present invention;

fig. 2 is a flowchart of an abnormality monitoring and processing method for a robot according to an embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

The present invention provides an abnormality monitoring and processing system for a robot, as shown in fig. 1, including:

the system comprises a construction module 1, a task management module and a task management module, wherein the construction module is used for constructing an abnormity monitoring library corresponding to a first task when the first robot executes the first task;

the monitoring module 2 is used for monitoring the abnormality of the first robot based on the abnormality monitoring library;

the processing module 3 is used for carrying out corresponding processing based on the result of the abnormity monitoring;

and the guiding module 4 is used for acquiring the processing situation when corresponding processing is carried out, and carrying out corresponding processing guidance based on the processing situation.

The working principle and the beneficial effects of the technical scheme are as follows:

the first robot is a manipulator. The first task may be, for example: and a manipulator loading and unloading task and the like. And introducing an abnormity monitoring library, wherein a large number of possible abnormity of the first robot when executing the first task is stored in the abnormity monitoring library. And monitoring the abnormality of the first robot based on the abnormality monitoring library. The operation parameters of the robot are not required to be analyzed by monitoring personnel, and the labor cost is reduced. Based on the results of the anomaly monitoring, exception handling is performed, for example: and the manipulator joint of the first robot does not move accurately, and maintenance personnel related to the manipulator joint of the robot are dispatched to go to the site for exception handling. When performing exception handling, a handling condition is obtained, and the handling condition may be, for example: on-site process scheduling and maintenance projects, etc. Based on the processing situation, processing guidance is performed, for example: instructing the service personnel how to perform the service project. When the processing personnel process basic abnormal problems or complex abnormal problems, processing guidance can be carried out, especially on some complex abnormal problems, professional personnel do not need to wait for processing, and the abnormal solving efficiency is improved.

The invention provides an abnormity monitoring and processing system of a robot, a construction module 1 constructs an abnormity monitoring library corresponding to a first task, and the abnormity monitoring library comprises:

acquiring a plurality of robot operation abnormity record items from a local and/or big data platform, wherein the robot operation abnormity record items comprise: at least one first abnormal item generated when the second robot executes the second task and the current first attribute information of the second robot;

performing a first difference analysis on the first task and the second task;

performing second difference analysis on the current first attribute information of the second robot and the current second attribute information of the first robot;

performing feature extraction on results of the first difference analysis and the second difference analysis based on a preset first feature extraction template to obtain a plurality of first feature values;

constructing a difference description vector based on the first characteristic value;

determining a value degree based on the difference description vector and a preset value recognition library;

if the value degree is larger than or equal to a preset value degree threshold value, at least one first abnormal item generated when the corresponding second robot executes the corresponding second task is used as a target to be warehoused;

and integrating and warehousing the targets to be warehoused to obtain an abnormal monitoring library corresponding to the first task, and completing construction.

The working principle and the beneficial effects of the technical scheme are as follows:

when constructing the anomaly monitoring library corresponding to the first task, it is necessary to predict an anomaly that may be generated when the first robot executes the first task. And acquiring an exception record, wherein the exception record comprises a first exception item generated when other second robots execute other second tasks and first attribute information of the second robots (when the second tasks are executed). And predicting the possible abnormality of the first robot when executing the first task based on the abnormality record, namely judging whether the first robot generates a first abnormal item when executing the first task.

The judgment of whether the first abnormal item is generated when the first robot executes the first task is divided into two types: first, whether the tasks have commonality or not refers to that when the robot respectively executes two tasks, the working states of each joint of the robot are similar enough, and the working states are specifically as follows: the joint moving speed, moving angle, moving distance, working temperature and the like, and the working state of the joint is common knowledge in the field of robots and is not described in detail. For example: the joint movement speed, movement distance, movement angle, and the like when the robot performs the first task and the second task are similar, which indicates that a first abnormal item related to the joint, which is generated when the second robot performs the second task, may occur when the first robot performs the first task. Secondly, whether the attributes of the robots have commonality or not refers to that the attributes of the two robots are similar enough, and the attributes of the robots are specifically as follows: the model, the historical maintenance record, the historical working record, the historical abnormal record and the like, and the attributes of the robot are also common knowledge in the field of robots and are not described in detail. For example: the model, the historical abnormal record, the historical use record and the like of the robot indicate that the first abnormal item generated when the second robot executes the second task may occur when the first robot executes the first task. In addition, the commonality between tasks and the commonality between the attributes of the robots are taken as a basis for judging whether the first abnormal item is generated when the first robot executes the first task, and only a principle description is made, and the results of the first difference analysis and the second difference analysis respectively describe the degree of the commonality between the tasks and the degree of the commonality between the attributes of the robots.

Therefore, a first difference analysis is performed on the first task and the second task, and a second difference analysis is performed on the current first attribute information of the second robot and the current second attribute information of the first robot. Introducing a preset first feature extraction template, and performing feature extraction on the difference analysis result to obtain a plurality of first feature values; the preset first feature extraction template may be, for example: the robot model is the same, the eigenvalue 1, again for example: the robot task execution joint movement angles are the same, and the characteristic value is 1. Constructing a difference description vector based on the first characteristic value; can be realized based on a vector construction technology, belongs to the field of the prior art and is not described in detail. The method includes the steps that a preset value recognition base is introduced, value degrees corresponding to different difference description vectors are stored in the value recognition base, the higher the value degree is, the higher the possibility that a first abnormal item generated when a corresponding second robot executes a corresponding second task possibly occurs when a first robot executes a first task is, the higher the possibility that a worker conducts statistics on experimental results in a probability degree judgment experiment that historical abnormalities of other robots represented by the different difference description vectors occur on the first robot in advance is conducted in the value recognition base, generally, the smaller the difference of characteristic values is, the larger the difference between the two robots is, the smaller the possibility that the historical abnormalities of the other robots occur on the first robot is, and the lower the value degree is. And determining a value degree based on the value recognition library by contrast, and determining a warehousing target based on the value degree. And constructing an abnormity monitoring library corresponding to the first task based on the warehousing target.

Generally, when an abnormality that may occur when a first robot executes a first task is predicted, there is a limitation that the first abnormality item that may occur when a large number of other second robots execute a second task cannot be used, because the robot has different attributes and the diversity of tasks executed, only based on the abnormality that may occur when the first robot has executed the first task historically, and the like. According to the method and the device, whether the first abnormal item generated when the second robot executes other second tasks can be used as the warehousing target or not is judged according to two dimensions of the fact whether the tasks have commonality and whether the attributes of the robots have commonality, so that the limitation that only the first robot can execute the abnormality and the like generated when the first robot executes the first tasks historically is broken through, and the first abnormal item generated when other second robots execute the second tasks is utilized. The applicability is improved to a great extent, and the development trend of big data sharing is met.

The invention provides an abnormity monitoring and processing system of a robot, a monitoring module 2 monitors abnormity of a first robot based on an abnormity monitoring library, and the abnormity monitoring and processing system comprises:

acquiring operation parameters of a first robot;

acquiring a preset abnormal monitoring strategy corresponding to a first abnormal item in an abnormal monitoring library, wherein the abnormal monitoring strategy comprises the following steps: parameter extraction strategy and abnormal analysis strategy;

extracting target parameters from the operating parameters based on a parameter extraction strategy;

and performing anomaly analysis on the target parameters based on an anomaly analysis strategy to complete monitoring.

The working principle and the beneficial effects of the technical scheme are as follows:

the operating parameters may be, for example: parameters of various sensors provided on the robot, and the like. Introducing a preset abnormal monitoring strategy corresponding to the first abnormal item, wherein the abnormal monitoring strategy comprises a parameter extraction strategy and an abnormal analysis strategy; the parameter extraction policy may be, for example: when detecting whether the motor temperature is abnormal, extracting a motor temperature strategy; the anomaly analysis strategy may be, for example: and analyzing whether the temperature curve of the motor is in an ascending trend or not. And the abnormity monitoring of the first robot is realized based on the parameter extraction strategy and the abnormity analysis strategy, so that the abnormity monitoring efficiency is improved.

The invention provides an abnormity monitoring and processing system of a robot, a processing module 3 carries out corresponding processing based on the result of abnormity monitoring, comprising:

when the result of the abnormality monitoring contains at least one second abnormal item, counting the total number of the second abnormal items;

if the total number is 1, acquiring a preset first processing strategy corresponding to the second abnormal item;

based on the first processing strategy, corresponding processing is carried out;

if the total number is not 1, performing feature extraction on the second abnormal item based on a preset second feature extraction template to obtain a plurality of second feature values;

constructing an anomaly description vector based on the second characteristic value;

determining an abnormal association relation between the at least two randomly selected second abnormal items based on the abnormal description vectors of the at least two randomly selected second abnormal items and a preset abnormal association recognition library;

acquiring an abnormal combination strategy corresponding to the abnormal association relation;

based on an abnormal combination strategy, performing abnormal combination on at least two second abnormal items which are selected correspondingly and randomly to obtain combined abnormal items;

acquiring a preset second processing strategy corresponding to the abnormal item combination and a preset third processing strategy corresponding to the second abnormal item which is not subjected to abnormal combination;

and performing corresponding processing based on the second processing strategy and the third processing strategy.

The working principle and the beneficial effects of the technical scheme are as follows:

when the second outlier is monitored, the total number is counted. If the total number is 1, only one second abnormal item is shown, a corresponding preset first processing strategy is obtained, and corresponding processing is carried out based on the first processing strategy; for example: the second abnormal item is that the robot joint has low movement precision, and the first processing strategy is to schedule maintenance personnel related to the robot joint to perform movement calibration on the joint again. If the total number is not 1, it indicates that there are a plurality of second abnormal items. When an anomaly is identified, some later-occurring anomalies may be the result of a previously-occurring anomaly, i.e., "a chain reaction," such as: when the motor that the robot drives the joint to move is abnormal, the joint movement is abnormal, if every second abnormal item is processed, the scheduling of personnel can be repeated, and after the personnel who are repeatedly scheduled carry out maintenance and inspection on the site, the chain reaction is discovered, so that the processing scheduling resource is increased, and the processing efficiency is reduced. Therefore, a preset second feature extraction template is introduced to perform feature extraction on the second abnormal item to obtain a plurality of second feature values; the preset second feature extraction template may be, for example: the motor temperature is higher than the preset value, and the characteristic value is 1. Constructing an anomaly description vector based on the second characteristic value; can be realized based on a vector construction technology, belongs to the field of the prior art and is not described in detail. Introducing a preset abnormal association recognition library, wherein abnormal description vectors with 'chain reaction' abnormal association relation are stored in the abnormal association recognition library, historical abnormal items of the 'chain reaction' of the robot are collected in advance by workers in the abnormal association recognition library, feature extraction is carried out on the historical abnormal items (a second feature extraction template is also used in the feature extraction method), the abnormal description vectors are respectively constructed, the constructed abnormal description vectors are subjected to association pairing to obtain pairing items, and the abnormal association recognition library is constructed on the basis of the pairing items. And determining an abnormal association relation between at least two second abnormal items selected randomly based on the abnormal association recognition library. Introducing a preset abnormal combination strategy corresponding to the abnormal association relationship, and performing abnormal combination on at least two second abnormal items which are selected correspondingly and randomly to obtain combined abnormal items; the exception merge policy may be, for example: the source abnormity causing the abnormity of the 'chain reaction' is taken as a merging abnormal item. And acquiring a preset second processing strategy corresponding to the abnormal item combination and a preset third processing strategy corresponding to the second abnormal item which is not subjected to abnormal combination, and performing corresponding abnormal processing. And the resources for processing and scheduling are reduced to a great extent, and the processing efficiency is improved.

The invention provides an abnormity monitoring and processing system of a robot, which guides a module 4 to acquire processing conditions and comprises the following steps:

acquiring the processing condition of a processing person performing corresponding processing based on the timed reply of the person terminal;

and/or the presence of a gas in the gas,

acquiring a processing condition through a maintenance recorder worn by a processing person who performs corresponding processing;

and/or the presence of a gas in the gas,

and acquiring the processing condition recorded by the maintenance recording unit activated after the first robot enters the maintenance mode.

The working principle and the beneficial effects of the technical scheme are as follows:

there are three ways to obtain the handling situation: firstly, a processing person replies a processing condition based on a person terminal, and the person terminal can be a smart phone and the like; secondly, a treating person wears a maintenance recorder, the maintenance recorder is provided with a camera which is generally worn on the head, and the treatment field recording is carried out at a first visual angle; and thirdly, when a processing person maintains the robot, the robot can enter a maintenance mode, after the robot enters the maintenance mode, the maintenance recording unit of the robot can be activated to record the processing condition, the parameter change of the robot is recorded, and the maintenance recording unit can be a cpu (central processing unit) and the like.

The invention provides an abnormity monitoring and processing system of a robot, a guidance module 4 carries out corresponding processing guidance based on processing conditions, and the system comprises:

inputting the result of the abnormal processing and the processing condition into a preset processing defect identification model, and determining at least one processing defect item;

acquiring preset guide information corresponding to the defect item;

delivering the instruction information to a personnel terminal of a processing personnel and/or a worn maintenance recorder;

and/or the presence of a gas in the gas,

acquiring a field image of a processing field for corresponding processing;

determining a first position and a first orientation of a face of a processing person based on the live image;

acquiring second positions and second orientations of a plurality of display devices in a preset range around the first position in the processing site;

constructing a first direction vector based on the first position and the first orientation;

constructing a second direction vector based on the second position and the second orientation;

and controlling the display equipment corresponding to the maximum included angle in the included angles of the second direction vector and the first direction vector to temporarily display the guide information.

The working principle and the beneficial effects of the technical scheme are as follows:

the method includes the steps that a preset processing defect recognition model is introduced, when a worker utilizes a large amount of abnormal robots to be processed in advance, records such as defects and insufficiency in processing conditions serve as training samples to train the neural network model to the converged neural network model, and training convergence indicates that training of the neural network model is completed. And determining the processing defect item existing in the processing based on the processing defect identification model, the abnormal processing result and the processing condition. Introducing preset guide information corresponding to the processing defect items, for example: the robot is standard in disassembly and assembly, and the guidance information is a demonstration picture for guiding a processor how to disassemble and assemble the robot. When the guidance information is viewed by a processing person, there are three corresponding ways: firstly, sending the information to a personnel terminal; the second type is sent to a maintenance recorder; third, the processor is prompted by a display device, which may be, for example: a display screen, etc. The accessibility of the guidance information is fully ensured, the processing personnel can be timely reminded, and the efficiency and the quality of exception handling are improved.

In addition, generally, a large number of display devices are provided in a treatment site, but the display devices are selected and the most suitable display device for prompting a treatment person is selected. In the selection, the selection is performed from the visual distance and the visual angle. A preset range is introduced, which may be, for example: 2 m. Respectively constructing a first direction vector and a second direction vector, calculating an included angle between the first direction vector and the second direction vector, and selecting display equipment corresponding to the maximum included angle to temporarily display guide information; vector construction and vector angle calculation belong to the category of the prior art and are not described in detail. Typically, the included angle is a maximum of 180 ° when the treating person is looking at a display device. The accessibility of the guidance information is further improved.

The invention provides an abnormity monitoring and processing system of a robot, which also comprises:

the docking module is used for docking the processing personnel with the expert personnel when the processing personnel performing corresponding processing inputs the expert personnel needing docking;

wherein, the butt joint module will handle personnel and expert's butt joint, include:

constructing an online meeting room;

processing personnel and expert personnel are accessed into an online meeting room;

classifying and grouping the processing conditions to obtain a plurality of grouped data of a first condition type;

constructing a total number of display partitions of a first case type in a common display area in an online conference room;

randomly mapping a plurality of grouped data of a first case type to each display subarea;

continuously acquiring a plurality of communication records generated by communication between a processing person and an expert in an online conference room;

establishing a time axis;

correspondingly setting the alternating current records on a time axis based on the generation time points of the alternating current records;

performing semantic extraction on the communication record in the latest preset time on the time axis to obtain at least one first semantic;

acquiring a preset trigger semantic library corresponding to a first condition type;

matching the first semantic with a second semantic in a trigger semantic library;

if the matching is in accordance with the preset trigger value corresponding to the matched second semantic;

if the sum of the trigger values of the accumulated calculation trigger values is greater than or equal to the preset trigger value and the threshold value, taking the corresponding first condition type as a second condition type;

acquiring preset trigger values and an amplification strategy library corresponding to the total number of the second condition types, and determining the trigger values and the amplification strategies corresponding to the second condition types;

and based on the amplification strategy, performing amplification processing on the display partition to which the grouped data corresponding to the second case type is mapped.

The working principle and the beneficial effects of the technical scheme are as follows:

in the actual exception handling process, when a processing person finds some difficult and complicated problems, the processing person knows who may be skilled in solving and can communicate with a corresponding target person through a telephone, but when the telephone communication is carried out, the target person cannot know specific field conditions, and after the processing person needs to carefully introduce and the like, the target person can know the specific field conditions, so that the processing person is complicated, the exception handling efficiency is reduced, and particularly when the target person has something at hand, the time of the target person is delayed, and the experience is reduced. Therefore, a solution is needed.

And constructing an online meeting room, and accessing the processing personnel and the expert personnel which are requested to be docked by the processing personnel into the online meeting room. Classifying and grouping the processing conditions to obtain a plurality of grouped data of a first condition type; the first case type may be, for example: scene pictures and robot parameters, etc. And mapping the grouped data to different display partitions in sequence. The system is used for the processing personnel and the expert personnel to check and mainly used for the expert personnel to check. However, when the expert checks, the expert needs to find the corresponding grouped data according to the content dictated by the processing personnel, which is complicated, and generally, the display screen of the intelligent terminal used by the expert is limited, which is difficult to check, especially when the first situation type is more. Thus, a record of the communication between the processing staff and the expert is obtained, for example: and recording the voice call. And setting on a time axis, and extracting a first semantic meaning of the communication record in the latest preset time. The method comprises the steps of introducing a preset triggering semantic library corresponding to a first situation type, storing a plurality of second semantics related to an exchange theme and the first situation type in the triggering semantic library, collecting the semantics related to different exchange themes in advance by a worker in the triggering semantic library, and constructing the triggering semantic library based on the collected semantics. And matching the first semantics with the second semantics, and if the matching is accordant, introducing a preset trigger value corresponding to the matched second semantics, wherein the larger the trigger value is, the more relevant the matched second semantics and the first condition type is. If the sum of the trigger values of the accumulated and calculated trigger values is greater than or equal to the preset trigger value and the threshold value, the situation that the processing personnel and the professional are mainly communicating and corresponding to the first situation type is shown,the corresponding first case type is taken as a second case type. And introducing a preset trigger value and an amplification strategy library corresponding to the total number of the second case types, and determining the trigger value and the corresponding amplification strategy corresponding to the second case type, wherein the larger the trigger value sum is, the more the processing personnel and the professional are mainly communicating, and the larger the amplification strategy is, the larger the amplification degree of the corresponding display subarea is. The convenience and the communication efficiency of communication between the processing personnel and the professional are improved to a great extent, and the processing efficiency is improved. The time of delaying the professional is avoided as much as possible, and the experience is improved. Meanwhile, the method is more applicable. In addition, the calculation formula for cumulatively calculating the trigger value is as follows:

to trigger the value sum, Q _d For the cumulative calculated d-th trigger value, n is the total number of the cumulative calculated trigger values.

The invention provides an abnormality monitoring and processing method of a robot, as shown in fig. 2, comprising the following steps:

step 1: when a first robot executes a first task, constructing an abnormality monitoring library corresponding to the first task;

step 2: monitoring the abnormality of the first robot based on the abnormality monitoring library;

and 3, step 3: based on the result of the abnormal monitoring, corresponding processing is carried out;

and 4, step 4: and when corresponding processing is carried out, acquiring the processing situation, and carrying out corresponding processing guidance based on the processing situation.

The invention provides an abnormity monitoring and processing method of a robot, wherein in step 1, an abnormity monitoring library corresponding to a first task is constructed, and the method comprises the following steps:

acquiring a plurality of robot operation abnormity record items from a local and/or big data platform, wherein the robot operation abnormity record items comprise: at least one first abnormal item generated when the second robot executes the second task and the current first attribute information of the second robot;

performing a first difference analysis on the first task and the second task;

performing second difference analysis on the current first attribute information of the second robot and the current second attribute information of the first robot;

performing feature extraction on results of the first difference analysis and the second difference analysis based on a preset first feature extraction template to obtain a plurality of first feature values;

constructing a difference description vector based on the first characteristic value;

determining a value degree based on the difference description vector and a preset value recognition library;

if the value degree is larger than or equal to a preset value degree threshold value, at least one first abnormal item generated when the corresponding second robot executes the corresponding second task is used as a target to be warehoused;

and integrating and warehousing the targets to be warehoused to obtain an abnormal monitoring library corresponding to the first task, and completing construction.

The invention provides an abnormity monitoring and processing method of a robot, which comprises the following steps: based on the abnormal monitoring library, the abnormal monitoring is carried out on the first robot, and the abnormal monitoring comprises the following steps:

acquiring operation parameters of a first robot;

acquiring a preset abnormal monitoring strategy corresponding to a first abnormal item in an abnormal monitoring library, wherein the abnormal monitoring strategy comprises the following steps: a parameter extraction strategy and an anomaly analysis strategy;

extracting target parameters from the operating parameters based on a parameter extraction strategy;

and performing anomaly analysis on the target parameters based on an anomaly analysis strategy to complete monitoring.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (10)

1. An anomaly monitoring and handling system for a robot, comprising:

the system comprises a construction module, a task execution module and a task execution module, wherein the construction module is used for constructing an abnormity monitoring library corresponding to a first task when the first robot executes the first task;

the monitoring module is used for monitoring the abnormality of the first robot based on the abnormality monitoring library;

the processing module is used for carrying out corresponding processing based on the result of the abnormity monitoring;

and the guiding module is used for acquiring the processing condition when corresponding processing is carried out, and carrying out corresponding processing guidance based on the processing condition.

2. The system of claim 1, wherein the building module builds an anomaly monitoring library corresponding to the first task, comprising:

acquiring a plurality of robot operation abnormity record items from a local and/or big data platform, wherein the robot operation abnormity record items comprise: at least one first abnormal item generated when a second robot executes a second task and the current first attribute information of the second robot;

performing a first difference analysis on the first task and the second task;

performing second difference analysis on the current first attribute information of the second robot and the current second attribute information of the first robot;

performing feature extraction on results of the first difference analysis and the second difference analysis based on a preset first feature extraction template to obtain a plurality of first feature values;

constructing a difference description vector based on the first characteristic value;

determining a value degree based on the difference description vector and a preset value recognition library;

if the value degree is larger than or equal to a preset value degree threshold value, at least one first abnormal item generated when the corresponding second robot executes the corresponding second task is used as a target to be warehoused;

and integrating and warehousing the targets to be warehoused to obtain an abnormal monitoring library corresponding to the first task, and completing construction.

3. The robot anomaly monitoring and handling system according to claim 2, wherein said monitoring module monitors anomalies of said first robot based on said anomaly monitoring library, comprising:

acquiring operation parameters of the first robot;

acquiring a preset abnormal monitoring strategy corresponding to the first abnormal item in the abnormal monitoring library, wherein the abnormal monitoring strategy comprises the following steps: a parameter extraction strategy and an anomaly analysis strategy;

extracting target parameters from the operating parameters based on the parameter extraction strategy;

and performing anomaly analysis on the target parameters based on the anomaly analysis strategy to complete monitoring.

4. The system for monitoring and processing robot abnormality according to claim 1, wherein said processing module performs corresponding processing based on the result of abnormality monitoring, including:

when the result of the anomaly monitoring contains at least one second abnormal item, counting the total number of the second abnormal item;

if the total number is 1, acquiring a preset first processing strategy corresponding to the second abnormal item;

based on the first processing strategy, corresponding processing is carried out;

if the total number is not 1, performing feature extraction on the second abnormal item based on a preset second feature extraction template to obtain a plurality of second feature values;

constructing an anomaly description vector based on the second characteristic value;

determining an abnormal association relation between at least two randomly selected second abnormal items based on the abnormal description vectors of the at least two randomly selected second abnormal items and a preset abnormal association recognition library;

obtaining an abnormal combination strategy corresponding to the abnormal association relation;

based on the abnormal combination strategy, performing abnormal combination on at least two second abnormal items which are correspondingly and randomly selected to obtain combined abnormal items;

acquiring a preset second processing strategy corresponding to the abnormal merging item and a preset third processing strategy corresponding to the second abnormal item which is not subjected to abnormal merging;

and performing corresponding processing based on the second processing strategy and the third processing strategy.

5. The system of claim 1, wherein the guidance module obtains a handling condition, comprising:

acquiring the processing condition of a processing person performing corresponding processing based on the timed reply of the person terminal;

and/or the presence of a gas in the gas,

acquiring a processing condition through a maintenance recorder worn by a processing person who performs corresponding processing;

and/or the presence of a gas in the gas,

and acquiring the processing condition recorded by the maintenance recording unit activated after the first robot enters the maintenance mode.

6. The system for monitoring and processing robot abnormality according to claim 5, wherein said guidance module performs corresponding processing guidance based on said processing situation, comprising:

inputting the result of the abnormal processing and the processing condition into a preset processing defect identification model, and determining at least one processing defect item;

acquiring preset guide information corresponding to the processing defect item;

delivering the guidance information to a personnel terminal of the processing personnel and/or a worn maintenance recorder;

and/or the presence of a gas in the gas,

acquiring a field image of a processing field for corresponding processing;

determining a first position and a first orientation of the face of the treatment person based on the live image;

acquiring second positions and second orientations of a plurality of display devices within a preset range around the first position in the processing site;

constructing a first direction vector based on the first location and the first orientation;

constructing a second direction vector based on the second location and the second orientation;

and controlling the display equipment corresponding to the largest included angle in the included angles of the second direction vector and the first direction vector to temporarily display the guidance information.

7. A robot anomaly monitoring and handling system as recited in claim 5, further comprising:

the docking module is used for docking the processing personnel with the expert personnel when the processing personnel performing corresponding processing inputs the expert personnel needing docking;

wherein the docking module docks the processing personnel with the expert personnel, including:

constructing an online meeting room;

accessing the processing personnel and the expert personnel to the online meeting room;

classifying and grouping the processing conditions to obtain a plurality of grouped data of a first condition type;

constructing a total number of display partitions of the first case type within a common display area within the online conference room;

randomly mapping a plurality of packet data of a first case type to each display partition;

continuously acquiring a plurality of communication records generated by communication between the processing personnel and the expert in the online conference room;

establishing a time axis;

correspondingly setting the alternating current records on the time axis based on the generation time points of the alternating current records;

performing semantic extraction on the alternating current records in the latest preset time on the time axis to obtain at least one first semantic;

acquiring a preset trigger semantic library corresponding to the first condition type;

matching the first semantic with a second semantic in the trigger semantic library;

if the matching is in accordance with the preset trigger value corresponding to the second semantic meaning in accordance with the matching, obtaining a preset trigger value corresponding to the second semantic meaning in accordance with the matching;

if the sum of the trigger values is calculated in an accumulating manner and is larger than or equal to a preset trigger value and a threshold value, taking the corresponding first condition type as a second condition type;

acquiring a preset trigger value and an amplification strategy library corresponding to the total number of the second condition types, and determining the trigger value and the corresponding amplification strategy corresponding to the second condition type;

and based on the amplification strategy, performing amplification processing on the display partition to which the grouped data corresponding to the second case type is mapped.

8. An anomaly monitoring and handling method for a robot, comprising:

step 1: when a first robot executes a first task, constructing an abnormity monitoring library corresponding to the first task;

step 2: monitoring the abnormality of the first robot based on the abnormality monitoring library;

and step 3: based on the result of the abnormal monitoring, corresponding processing is carried out;

and 4, step 4: and when corresponding processing is carried out, acquiring the processing condition, and carrying out corresponding processing guidance based on the processing condition.

9. The method for monitoring and processing robot abnormality according to claim 8, wherein in said step 1, constructing an abnormality monitoring library corresponding to said first task includes:

acquiring a plurality of robot operation abnormity record items from a local and/or big data platform, wherein the robot operation abnormity record items comprise: at least one first abnormal item generated when a second robot executes a second task and the current first attribute information of the second robot;

performing a first difference analysis on the first task and the second task;

performing second difference analysis on the current first attribute information of the second robot and the current second attribute information of the first robot;

performing feature extraction on results of the first difference analysis and the second difference analysis based on a preset first feature extraction template to obtain a plurality of first feature values;

constructing a difference description vector based on the first characteristic value;

determining a value degree based on the difference description vector and a preset value recognition library;

if the value degree is larger than or equal to a preset value degree threshold value, at least one first abnormal item generated when the corresponding second robot executes the corresponding second task is used as a target to be warehoused;

and integrating and warehousing the targets to be warehoused to obtain an abnormal monitoring library corresponding to the first task, and completing construction.

10. A robot abnormality monitoring and processing method according to claim 9, wherein said step 2: based on the abnormity monitoring library, carrying out abnormity monitoring on the first robot, wherein the abnormity monitoring comprises the following steps:

acquiring operation parameters of the first robot;

acquiring a preset abnormal monitoring strategy corresponding to the first abnormal item in the abnormal monitoring library, wherein the abnormal monitoring strategy comprises the following steps: a parameter extraction strategy and an anomaly analysis strategy;

extracting target parameters from the operating parameters based on the parameter extraction strategy;

and performing anomaly analysis on the target parameters based on the anomaly analysis strategy to complete monitoring.

Images