A method and system for robot intelligent inspection

Abstract

The application provides a robot intelligent inspection method, which comprises the following steps: S1, sample data of an inspection point is acquired, an inspection task is executed for each inspection point one by one, and a new round of calculation is started; S2, whether an inspection point is added to the inspection task is judged based on a prior probability value P i‑1 obtained in the last round of calculation; S3, if the inspection point has been added to the inspection task, an abnormality occurrence probability of the inspection point is calculated and predicted based on Bayes, a prediction result, i.e. a posterior probability value P i is updated and stored in a database; S4, if the inspection point has not been added to the inspection task, the probability value of the inspection point remains unchanged and is P i‑1 ; S5, one round of calculation is ended, data is sorted and stored in the database, and a new round of calculation is prepared. The application improves the detection efficiency of abnormal inspection points, enables the robot to detect the inspection point most likely to fail in an accurate manner, and further avoids missed detection.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent inspection, and particularly relates to a robot intelligent inspection method and system. BACKGROUND

[0002] Currently, there are several commonly used modes of robot inspection, including full-station inspection, routine inspection and special inspection. Full-station inspection refers to periodic full-volume inspection of all valid measurement points recorded in the detection site, which is also a relatively common inspection method and does not require manual intervention, and is repeated daily at fixed points. Routine inspection refers to adding some specified measurement points to the inspection task by manual selection, and performing periodic routine inspection operation. Special inspection refers to selecting measurement points for the robot to perform a one-time inspection task for some measurement points that may have hidden problems.

[0003] From the three execution modes, it can be seen that full-station inspection has the simplest configuration, the highest coverage, but the lowest inspection efficiency, and a large number of invalid inspection actions are performed. Routine inspection has strong pertinence, but it needs to be configured manually, and there may be a large difference between manual judgment and actual fault points, which can easily cause missed inspection. Special inspection is a supplement, and a one-time task is temporarily issued to the robot for execution, which requires a large amount of manual configuration work.

[0004] Intelligent inspection refers to dynamically adjusting the inspection frequency of each device observation point according to the device detection condition and the alarm level of the abnormality, and increasing the inspection frequency of the device with detection abnormalities according to a reasonable strategy. Starting from the actual inspection scene requirements, it is necessary to improve the detection efficiency and let the robot detect the most likely fault measurement point in an accurate way, and to avoid missed inspection, which may occur in the original manual configuration. However, since the previous inspection mode cannot meet the current requirements, the inspection strategy has defects, and the inspection process of the device is prone to re-inspection and missed inspection, which cannot meet the requirements of intelligent inspection.

[0005] Therefore, it is very meaningful to propose a robot intelligent inspection method and system. SUMMARY

[0006] In order to solve the problems of easy re-inspection and missed inspection in the existing device inspection process, the application provides a robot intelligent inspection method and system to solve the above technical defects.

[0007] In a first aspect, the application provides a robot intelligent inspection method, which comprises the following steps:

[0008] S1, acquiring sample data of inspection points, performing an inspection task for each inspection point one by one, and starting a new round of calculation;

[0009] S2, obtaining the prior probability value P based on the result of the last round of calculation i-1 , determining whether to add the inspection point to the inspection task;

[0010] S3, if the inspection point has been added to the inspection task, calculating and predicting the probability of abnormal occurrence of the inspection point based on Bayes, and the prediction result is the posterior probability value P i updating and storing in the database;

[0011] S4, if the inspection point has not been added to the inspection task, the probability value of the inspection point remains P i-1 ;

[0012] S41, further calculating the real-time probability value of abnormal occurrence of the inspection point based on Bernoulli test;

[0013] S42, further determining whether the real-time probability value exceeds the preset threshold;

[0014] S43, if the preset threshold is exceeded, the inspection point is added to the inspection task, otherwise, the n Bernoulli tests are continued to be executed; and

[0015] S5, ending one round of calculation, organizing data and storing in the database, and preparing for a new round of calculation.

[0016] Preferably, in S2, further comprising: determining whether to add the inspection point to the inspection task according to the alarm level of the inspection point and the probability of abnormal occurrence, wherein the alarm level is A-prewarning, B-general warning, C-severe warning, D-critical warning, and E-normal.

[0017] Further preferably, in S3, the formula for calculating the probability of abnormal occurrence based on Bayes is:

[0018]

[0019] wherein, the hyperparameter λ is set to 100, the result of this inspection is r, r=1 represents abnormal inspection result, and r=0 represents normal inspection result.

[0020] Further preferably, further comprising: determining whether to inspect the measurement point according to whether the probability of occurrence of each said alarm level exceeds the preset safety threshold, i.e.

[0021] 1-(1-P)>W

[0022] wherein, the safety threshold represents the probability threshold of abnormal occurrence of the measurement point, denoted by W, W<=1, and P is the probability of occurrence of the alarm level.

[0023] Preferably, in S4, further comprising: in a patrol period, fixedly performing n Bernoulli trials in each unit time T, calculating a real-time probability value of abnormality, wherein the unit time T is less than the time of performing a round of patrol task; and a formula for calculating the real-time probability value of abnormality based on the Bernoulli trials is:

[0024] P=1-(1-P i ) n

[0025] wherein P is the real-time probability value, and n is a period counter.

[0026] Further preferably, further comprising: assigning P0=1, assuming that all patrol points are likely to have abnormality, so as to ensure that all patrol points can be patrolled.

[0027] In a second aspect, the present application further provides a system for intelligent patrol of robots, comprising:

[0028] an acquisition module, configured to acquire sample data of patrol points;

[0029] a calculation module, configured to complete calculation in a patrol task process, and perform Bayesian calculation and n Bernoulli trials;

[0030] a patrol module, configured to perform a patrol task;

[0031] a first judgment module, configured to judge whether a patrol point is added to the patrol task;

[0032] a second judgment module, configured to judge whether a real-time probability value exceeds a preset threshold value;

[0033] an update module, configured to update data in the patrol task process.

[0034] In a third aspect, an electronic device is provided, comprising: one or more processors; and a storage device configured to store one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the implementation manners of the first aspect.

[0035] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program, when the computer program is executed by a processor, the computer program implements the method described in any of the implementation manners of the first aspect.

[0036] Compared with the prior art, the present application has the following advantages:

[0037] The robot intelligent inspection scheme of the application predicts the accuracy rate of the abnormal inspection point based on Bayesian prediction using prior probability distribution, and the inspection point is included in the inspection task according to the abnormal probability cumulative value; on the other hand, the coverage of all inspection points in the inspection task is ensured, the n-fold Bernoulli test is adopted, the time cumulative effect is considered, the time cumulative effect is numerized, and the judgment logic is added to avoid the situation that the inspection point is missed. The efficiency of detecting abnormal inspection points is improved, the robot can accurately detect the most likely failure inspection point, and further avoids missing detection. BRIEF DESCRIPTION OF DRAWINGS

[0038] The accompanying drawings are included to provide a further understanding of embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and serve to explain principles of the application. Other embodiments and many of the intended advantages of the present application will be readily appreciated as the same becomes better understood by reference to the following detailed description. The elements of the drawings are not necessarily to scale relative to each other. Like reference numerals designate corresponding similar parts.

[0039] Figure 1 is an exemplary device architecture diagram to which an embodiment of the application can be applied;

[0040] Figure 2 is a schematic diagram of the overall flow of the method of robot intelligent inspection of an embodiment of the application;

[0041] Figure 3 is a schematic diagram of the flow of the method of robot intelligent inspection of an embodiment of the application;

[0042] Figure 4 is a graph of experimental data in the method of robot intelligent inspection of an embodiment of the application;

[0043] Figure 5 is a schematic diagram of the abnormal detection rate in the method of robot intelligent inspection of an embodiment of the application;

[0044] Figure 6 is a trend graph of the detected point in the method of robot intelligent inspection of an embodiment of the application;

[0045] Figure 7 is a schematic diagram of the system of robot intelligent inspection of an embodiment of the application;

[0046] Figure 8 is a schematic diagram of the computer device structure of an electronic device suitable for implementing an embodiment of the application. DETAILED DESCRIPTION

[0047] In the following detailed description, reference is made to the accompanying drawings, which form a part hereof, and in which are shown by way of illustration illustrative embodiments in which the application can be practiced. This description references directional terms, such as "top," "bottom," "front," "back," "left," "right," "up," and "down," which can be used to simplify the description of the illustrative embodiments. It is to be understood that the directional terms are used in this description for the purpose of illustration and are not limiting. It is to be understood that other embodiments or logical changes can be made without departing from the scope of the present application. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present application is defined by the appended claims.

[0048] It should be understood that Figure 1 The number of terminal devices, networks and servers in the system is merely illustrative. Any number of terminal devices, networks and servers can be provided as needed.

[0049] Figure 1 An exemplary system architecture 100 for a method for processing information or an apparatus for processing information to which embodiments of the present application can be applied is shown.

[0050] As Figure 1 The system architecture 100 can include terminal devices 101, 102, 103, a network 104 and a server 105, as shown. The network 104 is a medium for providing a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links or fiber optic cables, etc.

[0051] A user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0052] The terminal devices 101, 102, 103 can be various electronic devices with communication functions, including but not limited to smart phones, tablet computers, laptop computers and desktop computers, etc.

[0053] The server 105 can be a server providing various services, such as a background information processing server for processing verification request information sent by the terminal devices 101, 102, 103. The background information processing server can analyze and process the received verification request information, and obtain a processing result (such as verification success information for indicating that the verification request is a legal request).

[0054] It should be noted that the method for processing information provided by the embodiment of the present application is generally executed by the server 105, and accordingly, the device for processing information is generally arranged in the server 105. In addition, the method for sending information provided by the embodiment of the present application is generally executed by the terminal device 101, 102 or 103, and accordingly, the device for sending information is generally arranged in the terminal device 101, 102 or 103.

[0055] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers or as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, to provide distributed services) or as a single software or multiple software modules, which are not specifically limited here.

[0056] From the actual inspection scene requirements, two main problems need to be solved: improve detection efficiency, and let the robot detect the most likely fault measurement point in an accurate way; avoid missing inspection, and the original manual configuration may miss the inspection point. Since the previous inspection mode cannot meet the requirements, a smart inspection algorithm is needed to realize this function.

[0057] Smart inspection refers to dynamically adjusting the inspection frequency of each device observation point according to the device detection condition and combining the alarm level of the exception, and automatically increasing or decreasing the inspection frequency of the device with detection exception according to a reasonable strategy. The inspection strategy has two key factors: alarm level (derived from preset value); and alarm number (derived from historical statistical value).

[0058] Bayesian prediction is a method of predicting observable points by using prior information and sample information to dynamically estimate parameters based on the Bayesian formula. Prior distribution is a subjective understanding of unknown parameters, and the use of prior distribution is the greatest feature of Bayesian statistics.

[0059] Considering the influence of time accumulation effect on the probability of possible inspection points, n Bernoulli trials are introduced. The mathematical model of repeating the same experiment is called independent trial sequence pattern. If any result of the experiment is not affected by the occurrence of other experiment results, the n experiments are independent. In particular, when there are only two possible results for each experiment, it is called n Bernoulli trial.

[0060] The above two probability distribution calculation methods need to be modeled in actual application scenarios to form mathematical formulas based on various parameter values.

[0061] In a first aspect, Figure 2 An embodiment of the present application is shown to disclose a method for intelligent inspection of a robot, asFigure 2 and Figure 3 As shown in the method comprises the following steps:

[0062] S1, the sample data of the inspection point is obtained, and each inspection point is executed one by one to perform the inspection task, and a new round of calculation is started;

[0063] S2, based on the prior probability value P i-1 obtained in the last round of calculation, it is judged whether a certain inspection point is added to the inspection task;

[0064] In a specific embodiment, P0=1 is assigned, and it is assumed that all inspection points may have abnormality to ensure that all inspection points can be inspected; including judging whether the inspection point is added to the inspection task according to the alarm level of the inspection point and the abnormality occurrence probability, wherein the alarm level is A-prewarning, B-general warning, C-severe warning, D-critical warning and E-normal.

[0065] Further, it also includes: according to whether the probability of each said alarm level occurrence exceeds a preset safety threshold, it is determined whether to inspect the inspection point, that is,

[0066] 1-(1-P)>W

[0067] Wherein, the safety threshold represents the probability threshold of the abnormality of the measuring point, represented by W, W<=1, and P is the probability of the alarm level occurrence.

[0068] S3, if the inspection point has been added to the inspection task, the abnormality occurrence probability of the inspection point is calculated and predicted based on Bayes, and the prediction result, that is, the posterior probability value P i is updated and stored in the database;

[0069] Specifically, the formula for calculating the abnormality occurrence probability based on Bayes is:

[0070]

[0071] Wherein, the hyperparameter λ is set to 100, the result of this inspection is r, r=1 represents abnormal inspection result, and r=0 represents normal inspection result.

[0072] S4, if the inspection point is not added to the inspection task, the probability value of the inspection point remains unchanged and is P i-1 ;

[0073] S41, the real-time probability value of the abnormality occurrence of the inspection point is further calculated based on Bernoulli test;

[0074] S42, it is further judged whether the real-time probability value exceeds a preset threshold;

[0075] S43, if the threshold is exceeded, the inspection point is added to the inspection task, otherwise continue to perform S41, S42 to conduct n Bernoulli trials; and

[0076] Specifically, in the inspection period, fixed in each unit of time T continues to conduct n Bernoulli trials, calculate the real-time probability value of abnormal occurrence, the unit of time T is less than the time of performing a round of inspection task; wherein, based on the formula for calculating the real-time probability value of abnormal occurrence of Bernoulli test is:

[0077] P = 1-(1-P i ) n

[0078] Wherein, P is the real-time probability value, n is the period counter.

[0079] S5, end of a round of calculation, organize data and into the warehouse, ready for a new round of calculation.

[0080] In specific embodiments, according to the "abnormal level (degree), namely alarm level" and "abnormal occurrence probability" of the measurement point A (for example, one measurement point, the same for others) to determine whether to add the measurement point to the inspection task, wherein the alarm level is divided into A-early warning, B-general warning, C-severe warning, D-critical warning, and E-normal.

[0081] Wherein, based on the Bayesian estimation of abnormal occurrence probability iterative calculation method, the cumulative effect probability of historical data is quantified by using Bayesian estimation.

[0082] Set the initial probability of abnormal occurrence before iteration as P i , i = 0, 1, 2, 3, …, the hyperparameter is set as λ, the larger λ represents the more attention to historical values, generally λ = 100. The result of this inspection r (result), r = 1 indicates abnormal, r = 0 indicates normal.

[0083]

[0084] Specifically, if the current round of inspection, A is not inspected, P i = P i-1 , remains unchanged. However, in order to ensure that all measurement points are detected, P0 = 1 is assigned in actual projects, assuming that all measurement points may have abnormal occurrence.

[0085] Further, the random number exhausts P i , i = 0, 1, 2, 3, …, the value changes as follows:

[0086] ① P0 = 1, r = 0, P1 = (1 x 100 + 0) / (100 + 1) = 100 / 101 = 0.99

[0087] P2=0.98, r=0, P3=0.96

[0088] P2=0.98, r=0, P3=0.96

[0089] ……After simulation iteration data

[0090] If all the measurement points are not abnormal, P 62 =0.54, P 462 =0.01, P 952 =0.001 (approximate calculation), the probability will gradually decay to 0, and >0;

[0091] If all the measurement points are abnormal, P1=(1x100+1) / 101=1, P2=1,..., P n =1, n>3;

[0092] If abnormal and not abnormal can occur, 0<P n <=1.

[0093] Further, according to whether the probability of each alarm level exceeds the safety threshold, it is determined whether to patrol the measurement point.

[0094] The alarm level occurrence probability and the corresponding severity level are explained as A~D, and one branch reaching the threshold coefficient W (Weight Value, threshold or weight) triggers. This part is configured as configuration data.

[0095] 1-(1-P)>W constraint trigger condition formula

[0096] The A level coefficient W(a) is 1 / 2, which means that as long as there is a 1 / 2 possibility of the abnormality, the measurement point will be added to the next inspection task;

[0097] The B level coefficient W(b) is 1 / 10, which means that as long as there is a 1 / 10 possibility of the abnormality, the measurement point will be added to the next inspection task;

[0098] The C level coefficient W(c) is 1 / 20; which means that as long as there is a 1 / 20 possibility of the abnormality, the measurement point will be added to the next inspection task;

[0099] The D level coefficient W(d) is 1 / 40; which means that as long as there is a 1 / 40 possibility of the abnormality, the measurement point will be added to the next inspection task;

[0100] E refers to the normal situation, and the abnormality level is 0, which does not affect whether the measurement point is added to the inspection task.

[0101] Further, the time accumulation factor of probability is numerized by using n Bernoulli trials, and the probability of time accumulation effect is calculated by using Bernoulli trials.

[0102] Specifically, the dynamic calculation of probability change uses the probability of the occurred event and the sample value to predict the probability value of the next event, and the probability change value of the non-occurred event also needs to be considered. In short, even if an abnormality rarely occurs at a measurement point, the measurement point is rarely added to the inspection task, but as time goes by, the longer the time, the more likely the measurement point will have an abnormality.

[0103] There are only two cases for the occurrence of an abnormality (1-occurred / 0-not occurred), which meets the characteristics of binomial distribution. In this regard, n Bernoulli trials are introduced, that is, the time accumulation factor is numerized. Before numerization, two factors need to be considered:

[0104] 1. Construct a calculation formula between the occurrence probability of an abnormality and the estimated duration (period);

[0105] 2. Accurately set the granularity of the estimated period;

[0106] For the first point, the n Bernoulli calculation result is calculated in a fixed period based on binomial distribution. For the second point, since the probability is recalculated based on Bayes each time the inspection is performed, the dynamic calculation period must be less than the inspection period to be meaningful, and can be measured by test data. The following details these two factors.

[0107] The prerequisite of n Bernoulli trials is that each time period is fixed as T, the period counter is n, and the real-time probability P value is continuously calculated

[0108] P = 1-(1-P i ) n Iteration algorithm formula (II)

[0109] If the measurement point is not added to the inspection task, the real-time probability value P = 1-(1-P i ) n , and n = n + 1 (the P value does not need to be stored in the database, and it can be calculated in real time), when the P value triggers the constraint condition, the inspection task is added, the Bernoulli calculation is stopped, otherwise the Bernoulli calculation of the next T time period is continued;

[0110] If the measurement point has been added to the inspection task, the iteration algorithm formula (I) is executed to calculate the initial probability value P i = (P i × 100 + r) / (100 + 1), n = 0, and the database updates the data value of the initial probability P i .

[0111] Specifically, the initial abnormal probability P0(B) = 5% at a certain time point i = 0 is known, the safety threshold of B level abnormality is W(b) = 0.1, and the safety level is 1-W(b) = 0.9;

[0112] 0T-P = P0, 1-P0(B) = 95% indicates that 95% of the time no abnormality occurs, and 1-(1-P b )<W(b) indicates that the possibility of B abnormality in the initial period is lower than the safety threshold (or referred to as higher than the safety level) in theory;

[0113] 1T-P = 1-(1-P0) 2 = 0.0975 < 0.1, indicating that the possibility of B abnormality in the second round of inspection task is still lower than the safety threshold;

[0114] 2T-P = 1-(1-P0) 3 = 0.1426 > 0.1, indicating that the current abnormality has exceeded the safety threshold, and if B is not added in the third round of inspection task, risk may occur.

[0115] When the exhaustive method is not convenient to reflect the processing of multiple data, the logarithmic function log 0.9 0.95 = 2.05408 (rounded to 3), indicating that the safety level is lower than the safety level at the third round, and the measurement point needs to be added to the inspection task.

[0116] In this specific embodiment, when the real-time probability in the Bernoulli test is higher than the safety threshold, it is determined that the detection point has abnormal risk and needs to be added to the inspection task. According to the exhaustive or logarithmic statistical method, the calculation result will change in the nth round, and in actual application, n experiments will be performed within the inspection period, that is, T = t (inspection time) / n. If n is too large, the risk will be exposed too early, resulting in low inspection efficiency; if n is too small, the Bernoulli test will fail, reducing the switch to trigger the abnormality. The value of n can be finally determined based on test verification to meet the design expectation requirements.

[0117] Further, in this embodiment, 80 rounds of inspection tasks are tested, and the abnormal probability change of the measurement point and the inspection coverage rate are tested. The inspection period is periodically performed every hour, and the real-time probability is estimated every 3 periods in each interval period, that is, the Bernoulli test is performed every 20 minutes, and the following data are obtained, as shown in Figure 4 , and the specific abnormal detection rate and detection point trend chart is shown in Figure 4 and 5 .

[0118] According to the results of the above intelligent inspection test (simulation of 80 rounds), the following conclusions are drawn:

[0119] 1. Abnormal detection rate change

[0120] Figure 4 The initial value of the abnormality detection rate is 0.4419, and the detection rate is 90% in the 22th round of inspection. According to the cumulative probability effect under the joint action of the abnormality level factor and the time factor, the abnormality rate of the measuring point changes between 57.58% and 95.06%;

[0121] 2. Trend chart of detected point

[0122] Figure 5 The initial value of the detected point is the ratio of the total measuring point to the abnormal measuring point (43 / 19), and the closest detection result is (21 / 20), and the median value changes relatively stably.

[0123] In summary, on the one hand, the intelligent inspection based on Bayesian prediction uses the prior probability distribution to predict the accuracy of the abnormal measuring point, and the measuring point is included in the inspection task according to the cumulative value of the abnormality probability; on the other hand, the coverage of all measuring points in the inspection task is ensured, the n-fold Bernoulli test is adopted, the time cumulative effect is considered, the time cumulative effect is numerized, and the judgment logic is added to avoid the situation that the account measuring point is missed, and the function preliminarily meets the expected design requirements.

[0124] Secondly, the present application further provides a robot intelligent inspection system, as shown in the accompanying drawings, comprising: Figure 7

[0125] The acquisition module 71 is used for acquiring sample data of the inspection point;

[0126] The calculation module 72 is used for completing the calculation in the inspection task process, performing Bayesian calculation and n-fold Bernoulli test;

[0127] The inspection module 73 is used for performing the inspection task;

[0128] The first judgment module 74 is used for judging whether a certain inspection point is added to the inspection task;

[0129] The second judgment module 75 is used for judging whether the real-time probability value exceeds the preset threshold value;

[0130] The update module 76 is used for updating the data in the inspection task process.

[0131] ​The robot intelligent inspection scheme of the application predicts the accuracy of the abnormal inspection points based on Bayesian prediction using prior probability distribution, and the inspection points are included in the inspection task according to the abnormal probability cumulative value; on the other hand, the coverage of all inspection points in the inspection task is ensured, the n-fold Bernoulli experiment is adopted, the time cumulative effect is considered, the time cumulative effect is numerized, and the judgment logic is added to avoid the situation that the inspection points are missed. The efficiency of detecting abnormal inspection points is improved, the robot can accurately detect the most possible fault inspection points, and further avoids missing.

[0132] Reference will now be made to the following description Figure 8 which shows a schematic diagram of a computer device 600 of an electronic device (such as a server or terminal device) suitable for use in implementing embodiments of the application. Figure 1 The electronic device shown is merely an example and should not impose any limitation on the function and scope of use of embodiments of the application. Figure 8 The electronic device shown is merely an example and should not impose any limitation on the function and scope of use of embodiments of the application.

[0133] As shown in Figure 8 , the computer device 600 includes a central processing unit (CPU) 601 and a graphics processor (GPU) 602, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 603 or programs loaded from a storage portion 609 into a random access memory (RAM) 606. In the RAM 604, various programs and data required for the operation of the device 600 are also stored. The CPU 601, GPU 602, ROM 603, and RAM 604 are connected to each other by a bus 605. An input / output (I / O) interface 606 is also connected to the bus 605.

[0134] The following components are connected to the I / O interface 606: an input portion 607 including a keyboard, a mouse, and the like; an output portion 608 including a display such as a cathode ray tube (CRT) display, a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage portion 609 including a hard disk, and the like; and a communication portion 610 including a network interface card such as a LAN card, a modem, and the like. The communication portion 610 performs communication processing via a network such as the Internet. A drive 611 can also be connected to the I / O interface 606 as needed. A removable medium 612 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 611 as needed, so that a computer program read therefrom is installed in the storage portion 609 as needed.

[0135] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program in accordance with embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 610, and / or installed from the removable medium 612. When the computer program is executed by the central processing unit (CPU) 601 and the graphics processor (GPU) 602, the above-described functions defined in the methods of the present invention are executed.

[0136] It should be noted that the computer readable medium described in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable medium may, for example, be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor device, device or means, or any combination of the above. More specific examples of the computer readable medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer readable medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution device, device or means. In the present invention, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable medium that can send, propagate or transmit the program for use by or in conjunction with an instruction execution device, device or means. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0137] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0138] The flow diagrams and the block diagrams in the drawings are meant as possible implementations of apparatus, methods, and computer program products according to various embodiments of the application. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or the block diagrams, can be implemented by analogues of computer program code means or instructions. These analogues of computer program code means or instructions can be provided to a processor of a computer system, or other programmable data processing apparatus, to produce a machine, such that the computer system, other programmable data processing apparatus, or other computer system can implement the functions or acts specified in the flow diagrams and / or block diagrams.

[0139] The modules described in the embodiments of the present application can be implemented by software, or by hardware. The described modules can also be arranged in a processor.

[0140] As another aspect, the present application also provides a computer readable medium, which can be included in the electronic device described in the above embodiments, or can exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the following steps: S1, obtaining sample data of the inspection points, performing the inspection task for each inspection point one by one, and starting a new round of calculation; S2, based on the prior probability value P i-1, judge whether a certain inspection point is added to the inspection task; S3, if the inspection point is added to the inspection task, calculate and predict the abnormal occurrence probability of the inspection point based on Bayes, and the prediction result, i.e. the posterior probability value P i is not changed and is still P i-1 ; S41, further calculate the real-time probability value of the abnormal occurrence of the inspection point based on Bernoulli test; S42, further judge whether the real-time probability value exceeds the preset threshold value; S43, if the real-time probability value exceeds the preset threshold value, add the inspection point to the inspection task, otherwise continue to perform S41 and S42 for n times of Bernoulli test; and S5, end one round of calculation, organize data and store in the database, and prepare for a new round of calculation.

[0141] The above description is only the preferred embodiment of the present application and the explanation of the applied technical principles. Those skilled in the art should understand that the scope of the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combinations of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with the technical features disclosed in the present application (but not limited to) having similar functions to form the technical solutions.

Claims

1. A method for intelligent robot inspection, characterized in that, The method includes the following steps: S1. Obtain sample data of inspection points, perform inspection tasks for each inspection point one by one, and start a new round of calculations; S2, Based on the prior probability value P obtained in the previous round of calculation. i-1 To determine whether a certain inspection point has been added to the inspection task; S3. If this inspection point has been added to the inspection task, then calculate and predict the probability of an anomaly occurring at this inspection point based on Bayesian calculation, and set the prediction result, i.e., the posterior probability value P. i Update and add to the database; S4. If this inspection point is not included in the inspection task, the probability value of this inspection point remains unchanged at P. i-1 ; S41. Next, calculate the real-time probability of an anomaly occurring at this inspection point based on the Bernoulli trial. The formula for calculating the probability of an anomaly based on Bayesian methods is as follows: In this case, the hyperparameter λ = 100 is set, and the result of this inspection is r. r = 1 indicates that the inspection result is abnormal, and r = 0 indicates that the inspection result is normal. S42. Further determine whether the real-time probability value exceeds the preset threshold; S43. If the threshold is exceeded, add the inspection point to the inspection task; otherwise, continue with S41 and S42 to perform n Bernoulli trials. The formula for calculating the real-time probability of an anomaly based on the Bernoulli trial is: P = 1 - (1 - P i ) n Where P is the real-time probability value, and n is the period counter; and S5. End one round of calculation, organize the data and store it in the database, and prepare for the next round of calculation.

2. The method for intelligent robot inspection according to claim 1, characterized in that, S2 also includes: determining whether to add the inspection point to the inspection task based on the alarm level and the probability of abnormal occurrence of the inspection point, wherein the alarm levels are A-early warning, B-general warning, C-serious warning, D-critical warning, and E-normal.

3. The method for intelligent robot inspection according to claim 2, characterized in that, It also includes: determining whether to inspect the inspection point based on whether the probability of each alarm level occurring exceeds a preset safety threshold, i.e. 1-(1-P)>W The safety threshold represents the probability threshold of an anomaly occurring at the measuring point, denoted by W, where W <= 1, and P is the probability of an alarm level occurring.

4. The method for intelligent robot inspection according to claim 1, characterized in that, S4 also includes: within the inspection cycle, continuously performing n Bernoulli trials at a fixed unit time T to calculate the real-time probability value of an anomaly, wherein the unit time T is less than the time to perform one round of inspection tasks.

5. The method for intelligent robot inspection according to claim 3, characterized in that, Also includes: Assign P0 = 1, assuming that all inspection points may experience anomalies, to ensure that all inspection points can be inspected.

6. A robot-based intelligent inspection system, characterized in that, include: Acquisition module: Used to acquire sample data from inspection points; Calculation module: Used to perform calculations during the inspection process, executing Bayesian calculations and n-fold Bernoulli trials. The formula for the probability of an anomaly occurring based on Bayesian calculations is as follows: Here, the hyperparameter λ = 100 is set, and the result of this inspection is denoted as r, where r = 1 indicates an abnormal inspection result and r = 0 indicates a normal inspection result. The formula for calculating the real-time probability of an abnormality based on the Bernoulli trial is: P = 1 - (1 - P i ) n Where P is the real-time probability value and n is the period counter; Inspection module: Used to execute inspection tasks one by one; First judgment module: used to determine whether a certain inspection point has been added to the inspection task; The second judgment module is used to determine whether the real-time probability value exceeds the preset threshold. Update module: Used to update data during the inspection process.

7. An electronic device, comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images