Method and apparatus for health assessment of a transport apparatus
The method and device for health assessment of transfer devices in autonomous systems address the lack of effective fault prediction by mapping in-situ motions to base motions, enabling predictive diagnostics and proactive maintenance to reduce downtime.
Patent Information
- Application Number
- KR1020237037265
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2018-05-04
- Filing Date
- 2018-05-07
- Publication Date
- 2026-11-06
- Estimated Expiration
- 2038-05-07
AI Technical Summary
Existing fault diagnosis systems for robotic manipulators and mechatronics devices in autonomous manufacturing tools lack effective fault prediction capabilities, often relying on mathematical modeling that is prone to non-conservative factors like signal noise and modeling errors, leading to unpredictable and costly unscheduled downtime.
A method and device for health assessment of transfer devices using a registration system, motion resolver, and processor to map in-situ process motions to predetermined base motions, comparing dynamic performance variables to evaluate health and predict performance deterioration, providing predictive diagnostics.
Enables accurate health evaluation and predictive maintenance by correlating dynamic performance variables to baseline values, allowing for proactive identification of performance issues and reducing unscheduled downtime in autonomous systems.
Smart Images

Figure R1020237037265_ABST
Abstract
Description
Technology Field
[0001] Cross-reference regarding related applications
[0002] This regular patent application claims priority and interest to U.S. provisional patent application No. 62 / 502,292 filed on May 5, 2017, which is incorporated herein by reference in its entirety.
[0003] Technology field
[0004] These exemplary embodiments generally relate to autonomous processing systems, and more specifically, to autonomous processing systems for health assessment and predictive diagnosis. Background Technology
[0005] Material damage and unscheduled downtime resulting from the failure of robotic manipulators and other mechatronics devices used in autonomous manufacturing tools, such as robotic material-handling platforms for producing semiconductor devices, are common problems that often entail a significant cost burden for the end users of the manufacturing tools.
[0006] Numerous health-monitoring and fault-diagnostic (HMFD) methods have been developed for industrial, automotive, and aerospace applications. Existing systems typically implement fault detection to indicate that something is wrong in the monitored system, fault isolation to determine the exact location of the fault—specifically, the faulty component—and fault verification to determine the magnitude of the fault.
[0007] The aforementioned isolation and verification tasks are often referred to together as fault diagnosis. Many existing systems implement only the fault detection and isolation steps.
[0008] Although such fault diagnosis methods are useful for fault detection, fault isolation, and adaptive recovery, they nevertheless cause the aforementioned devices, tools, FABs (e.g., manufacturing facilities / plants), or other autonomous equipment to operate in a substantially responsive manner with a limited or substantially non-existent prediction range. Prediction methods, such as mathematical modeling of autonomous equipment, are known to seek to increase the prediction range to that of fault diagnosis systems, where sensor measurements of autonomous equipment variables are compared to analytically calculated values of each variable (e.g., generated from Newtonian dynamic models of said autonomous equipment), and mathematical models representing nominal conditions exist. Such methods suffer from non-conservative factors, such as signal noise and modeling errors, which unpredictably and adversely affect the resulting difference between the analytical (nominal) values and the values from the sensor measurements, and require additional investment by the fault diagnosis system in processing capacitive and / or dual / redundant sensor systems and data systems to analyze such non-conservative factors.
[0009] It would be advantageous to have a fault diagnosis system that provides fault prediction without non-conservative factors associated with mathematical modeling.
[0010] Further background technology for the present invention is described in more detail in published patent documents, such as US 2011 / 0173496, US 2014 / 0222352, and US 2014 / 0336791, cited in the international search report of the international patent application (PCT / US2018 / 031422, May 7, 2018), which is the basis of the priority claim of the present application. The problem to be solved
[0011] We intend to provide a method and device for health evaluation of a transfer device for a fault diagnosis system as described above. means of solving the problem
[0012] According to one or more aspects of the disclosed embodiments, a method for health assessment of a system including a transfer device is:
[0013] A step of registering predetermined operation data that implements at least one dynamic performance variable output by a transfer device executing a predetermined set of predetermined motion bases of predetermined base motions, using a registration system communicably coupled to a device controller;
[0014] A base value (C) defined by the probability density function of each of the dynamic performance variables output by the transfer device for each motion of the above-determined set of motion bases. pkBase A step of determining ) using a processor communically coupled to the above registration system;
[0015] A step of analyzing in situ process motion commands of the device controller from the transfer device using a motion resolver communicably coupled to the device controller, wherein in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and another predetermined motion set of the transfer device is limited using the mapped in situ process motions;
[0016] Predetermined operation data specifying the at least one dynamic performance variable output by the transfer device executing the other predetermined motion set is registered using the registration system, and other values (C) defined by the probability density function of each dynamic performance variable output by the transfer device executing the mapped original position process motions of the other predetermined motion set. pkOther A step of determining ) using the above processor; and
[0017] For each of the dynamic performance variables output by the transfer device, corresponding to the aforementioned predetermined motion base set and the aforementioned other predetermined motion set, the other value and the base value (C pkBase The method includes comparing the transfer device using the processor, wherein the transfer device is a unique transfer device common to both the predetermined motion base set and the other predetermined motion set, and evaluating the health of the transfer device based on the comparison.
[0018] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-place process motion substantially maps onto a corresponding template motion among the template motions.
[0019] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0020] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command defines a template motion in at least one degree of freedom of the transfer device motion.
[0021] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, in a register of the device controller, and the processor analyzes the mapped motions from periodic access to the motion histogram in the register.
[0022] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0023] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0024] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0025] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering trend data for each dynamic performance variable using the registration system, wherein the trend data defines the deterioration trend of each dynamic performance variable.
[0026] According to one or more aspects of the disclosed embodiments, the method further comprises the step of aggregating, using the processor, the dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device, and predicting the occurrence of the transfer device having performance below a predetermined performance state.
[0027] According to one or more aspects of the disclosed embodiments, the method further comprises the step of providing to an operator of the transfer device, using the processor, an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0028] According to one or more aspects of the disclosed embodiments, a method for health assessment of a system including a transfer device is:
[0029] A step of registering predetermined operation data that implements at least one dynamic performance variable output by a transfer device executing a predetermined set of motion bases arranged to limit the statistical definition of predetermined base motions, using a registration system communicably coupled to a device controller;
[0030] A step of determining, using a processor communicably coupled to the registration system, a normalized value that statistically defines the nominal performance of each of the dynamic performance variables output by the transfer device for each motion of the aforementioned predetermined set of motion bases;
[0031] A step of analyzing in situ process motion commands of the device controller from the transfer device using a motion resolver communicably coupled to the device controller, wherein in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and another predetermined motion set of the transfer device is limited using the mapped in situ process motions;
[0032] A step of registering, using the registration system, predetermined operation data that specifies the at least one dynamic performance variable output by the transfer device executing the other predetermined motion set, and determining, using the processor, another normalized value that statistically defines the in-place process performance of each dynamic performance variable output by the transfer device executing the mapped in-place process motion of the other predetermined motion set; and
[0033] A step of comparing the other normalized value and the normalized value for each of the dynamic performance variables of the transfer device, corresponding to the aforementioned predetermined base motion set and the aforementioned other predetermined motion set, using the processor, and determining the rate of performance deterioration from the nominal performance of the transfer device based on the comparison, wherein the device is unique, and each normalized value (C) for each predetermined base motion of the predetermined motion base set pkbase ) and each different value (C) for each mapped in-place process motion of the other predetermined motion sets mentioned above. pkOther ) is uniquely correlated only with the above-mentioned unique device, and the above-mentioned predetermined performance deterioration rate is uniquely correlated only with the above-mentioned unique device, the step includes.
[0034] According to one or more aspects of the disclosed embodiments, the method further comprises the step of providing the system with a plurality of different unique devices and the transfer device connected to each other, wherein each different unique device from the plurality of different unique devices (i) has different corresponding normalized values (C) for each base motion of the predetermined base motion set. pkBasei) and other normalized values (C) for each mapped in-situ process motion of the other predetermined motion set that corresponds uniquely only to the different corresponding unique device (i) from a plurality of different unique devices. pkOtheri has ).
[0035] According to one or more aspects of the disclosed embodiments, the method comprises, for each different unique device (i), the corresponding normalized values (C pkBasei ) and other normalized values (C) uniquely correlated to the different corresponding unique device (i). pkOtheri A step of registering ) using a controller respectively coupled to the different corresponding unique devices, and for each different unique device (i), the uniquely correlated normalized values (C pkBasei ) and other normalized values of that different unique device (i) (C pkOtheri The method further includes the step of determining the corresponding performance deterioration rate for the different unique device (i) from the comparison of ) on the device based on the device (i=1...n).
[0036] According to one or more aspects of the disclosed embodiments, each different unique device from the plurality of different unique devices is in common with the transfer device.
[0037] According to one or more aspects of the disclosed embodiments, each different unique device from the plurality of different unique devices is configured differently from the transfer device.
[0038] According to one or more aspects of the disclosed embodiments, the method further includes the step of registering trend data defining the performance deterioration trend of each of the plurality of different unique devices of the system and the transfer device in the register of the controller.
[0039] According to one or more aspects of the disclosed embodiments, the method further comprises the step of determining system performance deterioration trends that define performance deterioration of the system by combining performance deterioration trends corresponding to the transfer device and each of the plurality of different unique devices of the system using the processor.
[0040] According to one or more aspects of the disclosed embodiments, the method further comprises the steps of: using the processor to compare the performance deterioration trend of the transfer device with the performance deterioration trend of another of the plurality of different unique devices; and using the processor to determine whether the performance deterioration trend of the transfer device or the performance deterioration trend of another of the plurality of different unique devices is a control performance deterioration trend and whether the control performance deterioration trend is a determining cause of the performance deterioration trend of the system.
[0041] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-situ process motion substantially maps onto a corresponding template motion among the template motions.
[0042] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0043] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command is defined by a template motion within at least one degree of freedom of motion of the transfer device.
[0044] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, in a register of the device controller, and the processor analyzes the mapped motions from periodic access of the motion histogram in the register.
[0045] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0046] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0047] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0048] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering trend data for each dynamic performance variable using the registration system, wherein the trend data defines the deterioration trend of each dynamic performance variable.
[0049] According to one or more aspects of the disclosed embodiments, the method further comprises the step of aggregating, using the processor, the dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device, and predicting the occurrence of the transfer device having performance below a predetermined performance state.
[0050] According to one or more aspects of the disclosed embodiments, the method further comprises the step of providing to an operator of the transfer device, using the processor, an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0051] According to one or more aspects of the disclosed embodiments, a health assessment device for evaluating the health of a system including a transfer device comprises:
[0052] A registration system communically coupled to the transfer device controller of the above-mentioned transfer device, wherein the registration system is,
[0053] It is configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing a predetermined motion base set of predetermined base motions, and
[0054] A registration system configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing another predetermined set of motions;
[0055] As a motion resolver communicably coupled to the above-mentioned device controller, the motion resolver is,
[0056] The above device controller is configured to analyze in situ process motion commands from the transfer device, and the in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and
[0057] A motion resolver configured to limit another predetermined set of motions of the above transfer device using the mapped original position process motions; and
[0058] As a processor communicably coupled to the above-mentioned registration system, the processor,
[0059] A base value (C) defined by the probability density function of each of the dynamic performance variables output by the transfer device for each motion of the above-determined set of motion bases. pkBase It is configured to determine ),
[0060] Another value (C) defined by the probability density function of each dynamic performance variable output by the transfer device executing the mapped in-place process motions of the other predetermined motion set. pkOther It is configured to determine ),
[0061] For each of the dynamic performance variables output by the transfer device, corresponding respectively to the aforementioned predetermined motion base set and the aforementioned other predetermined motion set, the other value and the base value (C pkBase It is configured to compare ), and
[0062] It includes a processor configured to evaluate the health of the transfer device based on the above comparison,
[0063] The above transfer device is a common transfer device for both the above-determined motion base set and the other above-determined motion set.
[0064] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-place process motion substantially maps onto a corresponding template motion among the template motions.
[0065] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0066] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command defines a template motion in at least one degree of freedom of the transfer device motion.
[0067] According to one or more aspects of the disclosed embodiments, the transfer device controller comprises a register configured to register a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, and the processor is further configured to analyze the mapped motions from periodic access to the motion histogram in the register.
[0068] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0069] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0070] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0071] According to one or more aspects of the disclosed embodiments, the registration system is further configured to register trending data for each dynamic performance variable, said trending data defines the deterioration trend of each dynamic performance variable.
[0072] According to one or more aspects of the disclosed embodiments, the processor is further configured to aggregate dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device using the processor, and to predict the occurrence of the transfer device having performance below a predetermined performance state.
[0073] According to one or more aspects of the disclosed embodiments, the transfer device is further configured to provide the operator of the transfer device with an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0074] According to one or more aspects of the disclosed embodiments, a health assessment device for evaluating the health of a system including a transfer device comprises:
[0075] A registration system communically coupled to the transfer device controller of the above-mentioned transfer device, wherein the registration system is,
[0076] It is configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing a predetermined motion base set arranged to limit the statistical definition of predetermined base motions, and
[0077] A registration system configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing another predetermined set of motions;
[0078] As a motion resolver communicably coupled to the above-mentioned device controller, the motion resolver is,
[0079] The above device controller is configured to analyze in situ process motion commands from the transfer device, and the in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and
[0080] A motion resolver configured to limit another predetermined set of motions of the above transfer device using the mapped original position process motions; and
[0081] As a processor communicably coupled to the above-mentioned registration system, the processor,
[0082] It is configured to determine a normalized value that statistically defines the nominal performance of each of the dynamic performance variables output by the transfer device for each motion of the above-determined set of motion bases, and
[0083] It is configured to determine other normalized values that statistically define the in-situ process performance of each dynamic performance variable output by the transfer device executing the mapped in-situ process motion of the other predetermined motion set, and
[0084] It is configured to compare the other normalized value and the normalized value for each of the dynamic performance variables of the transfer device, corresponding to the above-determined base motion set and the other-determined motion set, respectively, and
[0085] It includes a processor configured to determine the performance deterioration rate of the transfer device from the nominal performance based on the above comparison,
[0086] The above transfer device is a common transfer device for both the above-determined base motion set and the other-determined motion set.
[0087] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-place process motion substantially maps onto a corresponding template motion among the template motions.
[0088] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0089] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command defines a template motion in at least one degree of freedom of the transfer device motion.
[0090] According to one or more aspects of the disclosed embodiments, the transfer device controller comprises a register configured to register a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, and the processor is further configured to analyze the mapped motions from periodic access to the motion histogram in the register.
[0091] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0092] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0093] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0094] According to one or more aspects of the disclosed embodiments, the registration system is further configured to register trending data for each dynamic performance variable, said trending data defines the deterioration trend of each dynamic performance variable.
[0095] According to one or more aspects of the disclosed embodiments, the processor is further configured to aggregate dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device using the processor, and to predict the occurrence of the transfer device having performance below a predetermined performance state.
[0096] According to one or more aspects of the disclosed embodiments, the transfer device is further configured to provide the operator of the transfer device with an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0097] It should be understood that the foregoing description is merely illustrative of the aspects of the disclosed embodiments. Various alternatives and modifications may be devised by those skilled in the art without departing from the aspects of the disclosed embodiments. Accordingly, the aspects of the disclosed embodiments are intended to include all such alternatives, modifications, and variations that fall within the scope of the appended claims. Furthermore, the mere fact that different features are described in mutually different dependent or independent claims does not indicate that a combination of these features cannot be advantageously used, and such combinations remain within the scope of the aspects of the invention. Effects of the invention
[0098] The effects of the present invention are individually specified in the relevant parts of this specification. Brief explanation of the drawing
[0099] The aforementioned aspects and other features of the disclosed embodiments are described in the following description taken together with the accompanying drawings. FIG. 1 is a schematic diagram of a controller for an autonomous device, such as an autonomous material handling platform, according to the embodiments disclosed. FIG. 2 is a schematic diagram of an autonomous material handling platform according to the appearances of the disclosed embodiments. FIG. 2a is a schematic diagram of a system comprising a plurality of different unique devices according to the appearances of the disclosed embodiments. FIG. 3 is a schematic diagram of a device such as a transfer robot of an autonomous material handling device according to the appearances of the disclosed embodiments. FIGS. 4a-4e are schematic diagrams of different arm configurations for the apparatus of FIG. 3 according to the appearances of the disclosed embodiments. FIG. 5a is a schematic drawing of a part of an autonomous material handling platform showing base movements and in-situ process movements according to the appearances of the disclosed embodiments. FIGS. 5b and FIGS. 5c are schematic drawings of simple and complex movements according to the embodiments disclosed. FIG. 6 is an exemplary chart illustrating the statistical convergence of a sample of movement performed by the device of FIG. 3 according to the appearances of the disclosed embodiments. FIG. 7 is an exemplary movement histogram according to the appearances of the disclosed embodiments. FIG. 8a is a schematic diagram of an exemplary process flow according to the appearances of the disclosed embodiments. FIG. 8b is a schematic diagram of part of the exemplary process flow of FIG. 8a according to the appearances of the disclosed embodiments. FIG. 9 is an exemplary Gaussian distribution of sample shifts showing upper and lower limits according to the appearances of the disclosed embodiments. FIG. 10 is a graphic diagram of a comparison between different values and a baseline value generated from in-situ process movements according to the appearances of the disclosed embodiments. FIG. 11 is an exemplary illustration of the application of the health assessment of the device of FIG. 3 regarding predictive diagnoses according to the embodiments disclosed. FIG. 12 is an exemplary drawing of a health evaluation indication according to the appearances of the disclosed embodiments. FIG. 13 is an exemplary flowchart according to the appearances of the disclosed embodiments. FIG. 14 is a flowchart according to the appearances of the disclosed embodiments. FIG. 15 is a flowchart according to the appearances of the disclosed embodiments. Specific details for implementing the invention
[0100] Although the embodiments disclosed above will be described with reference to the drawings, it should be understood that the embodiments disclosed above may be realized in many shapes. Additionally, elements or materials of any suitable size, shape, or type may be used.
[0101] The embodiments disclosed herein described herein provide a method and apparatus for quantifying health status and predictive diagnoses for an autonomous system using available variables monitored by any suitable controller of the autonomous system (such as those described herein in relation to FIGS. 1–4e) (wherein the controller comprises non-transient computer software code for implementing the embodiments disclosed herein). The health status metrics are achieved using the embodiments disclosed herein by unique statistical data handling for collected variables that uniquely correlate with the health status quantity, which defines a health status quantity that provides predictive diagnoses for a given device and / or system of multiple devices. The embodiments disclosed above may allow the controller of the autonomous system to determine the statistical signature of the monitored variables (the unique characteristic of the unique device) by using the concept of "baselines" (including base values and / or base motions as described herein) and by further comparing future performances with respect to such baselines. As a result, the method and apparatus of the disclosed embodiments may allow the controller of the autonomous system to perform predictions based on trending analysis and to make recommendations for preventive maintenance based on the unique data from the autonomous system being monitored. The embodiments disclosed above may also allow for the identification of expected limits of operation for variables for which it is difficult to determine acceptable and practical specifications.
[0102] Although aspects of the embodiments disclosed above will be described herein with respect to a semiconductor robot (also referred to herein as a robot manipulator) having three degrees of freedom (theta rotation, R extension, and Z lift motion); in other aspects, the semiconductor robot may have more or fewer than three degrees of freedom. In other aspects, the embodiments disclosed above may be applied to other components of a semiconductor processing system having a single degree of freedom motion (such as robot transfers, load ports, aligners, pumps, fans, valves, etc.). It should also be understood that aspects of the embodiments disclosed above may be used for autonomous and / or powered device(s) or systems (including combinations of the aforementioned devices and / or devices) capable of sampling similar or related performance monitoring data that uniquely defines each unique device, device, and / or system and is uniquely correlated to them.
[0103] The embodiments disclosed above provide types of normalized metrics based on statistical parameters that allow for a direct comparison of variables of different physical meanings, such as temperature versus peak torque. Such a comparison allows for the calculation of the influence of such unrelated variables on the overall health status of the autonomous system being monitored.
[0104] FIG. 1 shows an exemplary controller 100 for an autonomous device that integrates autonomous device health assessment and predictive diagnosis according to aspects of the disclosed embodiments. The aspects of the disclosed embodiments may operate in hardware or software. For example, the aspects of the disclosed embodiments may exist within a component controller, a controller that directs the operation of a plurality of components, a controller that controls a sub-system, or a system controller. The aspects of the disclosed embodiments may also be implemented in dedicated hardware or software.
[0105] The above controller 100 may be any suitable controller of an autonomous device (such as the autonomous material handling platform 300 shown in FIG. 3) and may generally include a processor 105, read-only memory 110, random access memory 115, program storage unit 120, user interface 125, and network interface 130. The processor 105 may include an onboard cache 135 and may normally operate to read information and programs from a computer-usable medium, such as the onboard cache 135, read-only memory 110, random access memory 115, and program storage unit 120, from a computer program product.
[0106] When power is supplied, the processor 105 starts operating programs found within the read-only memory 110 and, after initialization, loads instructions from the program storage unit 120 into the random access memory 115 so that it can operate under the control of those programs. Frequently used instructions may be temporarily stored on the onboard cache 135. Both the read-only memory 110 and the random access memory 115 may utilize semiconductor technology or any other suitable materials and technologies. The program storage unit 120 may include a diskette, a computer hard drive, a compact disc (CD), or any other device capable of storing programs in the form of code that a computer can use.
[0107] The onboard cache 135, read-only memory 110, random access memory 115, and program storage unit 120 may include operating system programs individually or in any combination. The operating system programs may include an optional real-time operating system to improve the quality of data provided by the functional controller 100 and to allow the functional controller 100 to provide a guaranteed response time.
[0108] In particular, the onboard cache 135, read-only memory 110, random access memory 115, and program storage unit 120 may include programs that enable the processor 105 to perform fault diagnosis and fault prediction according to the embodiments disclosed herein, either individually or in any combination thereof. The network interface 130 may typically be adapted to provide an interface between the controller 100 and other controllers or other systems. The network interface 130 may operate to receive data from one or more additional functional controllers and to transport data to the same or different functional controllers. The network interface 130 may also provide an interface to a global diagnostic system capable of providing remote monitoring and diagnostic services.
[0109] Communication network 190 may include PSTN (Public Switched Telephone Network), the Internet, wireless networks, wired networks, LAN (Local Area Network), WAN (Wide Area Network), VPN (virtual private network), etc., and may further include other types of networks including X.25, TCP / IP, ATM, etc.
[0110] The controller 100 may include a user interface 125 equipped with a display 140 and an input device such as a keyboard 155 or a mouse 145. The user interface may be operated by a user interface controller 150 under the control of a processor 105 and may provide a graphical user interface to the user to visualize the results of health monitoring and fault diagnosis. The user interface may also be used to guide service to employees through troubleshooting routines or repair processes. Additionally, the user interface controller may also provide a connection or interface 155 for communication with other functional controllers, an external network, another control system, or a host computer.
[0111] An exemplary material-handling platform for producing semiconductor devices in which the aspects of the disclosed embodiments can be implemented is schematically illustrated in FIG. 2, and descriptive notes for the major components are listed in Table 1. One or more controllers of the material-handling platform of FIG. 2 may include controllers such as those described herein with respect to FIG. 1.
[0112] number explanation 301 Waiting section 302 vacuum section 303 Process Module 304 Sealing part 305 Lordport 306 Waiting robot money player 307 PCB aligner 308 Fan-filter unit 309 vacuum chamber 310 load-lock 311 Vacuum robot manipulator 312 vacuum pump 313 slit valve 314 Tool controller 315 Standby section controller 316 Vacuum section controller 317 Process controller 318 Load port controller 319 Standby robot controller 320 Aligner controller 321 Fan-filter unit controller 322 Motor controller 323 Vacuum robot controller
[0113] Table 1: Annotations to the example for the autonomous material handling platform 300 (also referred to as a process tool) of FIG. 2. The autonomous material handling platform 300 comprises an atmospheric section 301, a vacuum section 302, and one or more process modules 303.
[0114] The above-mentioned atmosphere section 301 may include a sealing section 304, one or more load ports 305, one or more robot manipulators 306, one or more substrate aligners 307, and a fan-filter unit 308. It may also include one or more ionization units (not shown). The above-mentioned vacuum section may include a vacuum chamber (309), one or more load-locks 310, one or more robot manipulators 311, one or more vacuum pumps 312, and a plurality of slit valves 313, which are typically placed at the interface between the vacuum chamber 309 and the process module 303 with the load-locks 310 of the atmosphere section 301.
[0115] The operation of the above platform is coordinated by tool controller 314, which supervises standby section controller 315, vacuum section controller 316, and one or more process controllers 317. Standby section controller 315 manages one or more loadport controllers 318, one or more standby robot controllers 319, one or more aligner controllers 320, and fan-filter unit controller 321. Each of the loadport controllers 318, standby robot controllers 319, and aligner controllers 320 manages one or more motor controllers 322 in turn. The vacuum section controller 316 manages one or more vacuum robot controllers 323, controls the vacuum pump 312, and operates the slit valves 313. The role of process controller 317 is subordinate to the operations performed within the process modules 303.
[0116] In some cases, it may be practical to combine two or more layers of control into a single controller. For example, the standby robot controller 319 and the corresponding motor controllers 322 may be combined within a single centralized robot controller, or the standby section controller 315 may be combined with the standby robot controller 319 to eliminate the need for two separate controller units.
[0117] A 5-axis direct-drive robot manipulator 400 may be used in the autonomous material handling platform 300 of FIG. 2, wherein one or more of the atmospheric robot manipulator 306 and the vacuum robot manipulator 311 are substantially similar to the robot manipulator 400. A simplified drawing of such a robot manipulator 400 is provided in FIG. 3. Exemplary notes for the principal components are listed in Table 2. In one aspect, the appearances of the disclosed embodiments may be realized within the robot manipulator 400; However, it should be understood that while the embodiments disclosed are described with respect to a robot manipulator, the embodiments disclosed may be implemented in any suitable autonomous part of the autonomous material handling platform 300, including but not limited to transfer robots, load ports, aligners, pumps, fans, valves, etc., and that the controller 800 in FIG. 8a is a general representation of a controller for any one of the aforementioned autonomous equipment. It should be noted that the robot manipulator 400 is depicted as a 5-axis direct drive robot manipulator for exemplary purposes only, and that in other embodiments, the robot manipulator (or other autonomous part of the process tool including the embodiments disclosed) may have suitable degrees of freedom and a suitable number of drive axes while having a direct or indirect drive system.
[0118] number explanation 401 Robot frame 402 Mounting flange 403 vertical rail 404 linear bearings 405 carriage 406 Vertical drive motor 407 ball screw 408 Motor 1 (Drive Link 1) 409 Motor 2 (Drive Link 2) 410 Encoder 1 (coupled to Motor 1) 411 Encoder 2 (coupled to Motor 2) 412 External sharpener 413 Internal sharpener 414 Link 1 (Top Arm) 415 Belt drive link 2 416 Link 2 (foream) 417A Motor A (driving end-effector A) 417B Motor B (driving end-effector B) 418A First stage of belt drive A 418B First stage of belt drive B 419A The second stage of belt drive A 419B The second stage of belt drive B 420A End-Effector A (Top End-Effector) 420B End-Effector B (Bottom End-Effector) 421A, 421B Payload on end-effectors A and B 422 Master controller 423A, 423B, 423C Motor controllers 424A, 424B Electronic units for end-effects A and B 425 communication network 426 slip-ring 428A, 428B Mapper sensors 429 Power supply 430 vacuum pump 431A, 431B valves 432A, 432B Pressure sensors 433, 434A, 434B lip-seals 435 brake
[0119] Table 2: Referring to Note 3 of the example for the robot manipulator of FIG. 3, the robot manipulator 400 is constructed around an open cylindrical frame 301 suspended from a circular mounting flange 402. The frame 401 incorporates a vertical rail 403 with a linear bearing 404 to provide guidance to a carriage 405 driven by a brushless DC motor 406 via a ball-screw mechanism 407. The carriage 405 houses a pair of coaxial brushless DC motors 408 and 409 equipped with optical encoders 410 and 411. The upper motor 408 drives a hollow outer shaft 412 connected to a first link 414 of the robot arm. The lower motor 409 is connected to a coaxial inner shaft 413 coupled to a second link 416 via a belt drive 415. The first link 414 houses a brushless DC motor 417A that drives the upper end-effector 420A via two-stage belt sets 418A and 419A. Another DC brushless motor 417B and two-stage belt drives 418B and 419B are used to operate the lower end-effector 420B. Each of the stages 418A, 418B, 419A, and 419B is designed to have a 1:2 ratio between the input pulley and the output pulley. The substrates 421A and 421B are attached to and held to the end-effectors 420A and 420B, respectively, by means of vacuum-operated edge-contact grippers, surface-contact suction grippers, or manual grippers.
[0120] The first link 414, second link 416, upper end-effector 420A, and lower end-effector 420B mentioned above are also referred to throughout this document as the upper arm, forearm, end-effector A, and end-effector B, respectively. Points A, B, and C indicate revolute couplings, referred to as solder, elbow, and wrist joints, respectively. Point D indicates a reference point indicating a desired location of the center of the substrate on the corresponding end-effector.
[0121] The control system of the robot manipulator of the above example may be of a distributed type. The control system includes a power supply 429, a master controller 422, and motor controllers 423A, 423B, and 423C. The master controller 422 is responsible for management tasks and trajectory planning. Each of the motor controllers 423A, 423B, and 423C executes position and current feedback loops for one or more motors. In FIG. 3, controller 423A controls motors 408 and 409, controller 423B controls motors 417A and 417B, and controller 423C controls motor 406. In addition to executing the feedback loops, the motor controllers also collect data such as motor current, motor position, and motor speed, and stream that data to the master controller. The motor controllers 423A, 423B, and 423C are connected to the master controller via a high-speed communication network 425. Since joint A is an infinite rotation joint, the communication network 425 is routed through a slip ring 426. Additional electronic units 424A and 424B may be used to support the edge-contact grippers of the end-effectors 420A and 420B, respectively.
[0122] Referring now to FIGS. 4a-4e, the robot manipulator (400) of FIG. 3 may include suitable arm linkage mechanism(s). Suitable examples of arm linkage mechanisms include, for example, U.S. Patent No. 7,578,649 issued August 25, 2009; U.S. Patent 5,794,487 issued August 18, 1998; U.S. Patent 7,946,800 issued May 24, 2011; U.S. Patent 6,485,250 issued November 26, 2002; U.S. Patent 7,891,935 issued February 22, 2011; U.S. Patent 8,419,341 issued April 16, 2013; U.S. Patent Application No. 13 / 293,717 titled "Dual Arm Robot" filed November 10, 2011; and "Linear Vacuum Robot with Z Motion and" filed September 5, 2013. It can be found in U.S. Patent Application 13 / 861,693 titled “Articulated Arm”, the disclosure of which is incorporated herein by reference in its entirety. In the embodiments disclosed, at least one of the transfer arm, each transfer unit module 104, boom arm 143 and / or linear slide 144, may be derived from a traditional SCARA arm 315 (optional compatible segmental robotic arm) type design comprising a top arm 315U, a band-driven forearm 315F and a band-forced end-effector 315E (Fig. 4c), or may be derived from a telescoping arm or any other suitable arm design such as a Cartesian linear sliding arm 314 (Fig. 4b).Suitable examples of transport arms can be found, for example, in U.S. Patent Application No. 12 / 117,415, filed May 8, 2008, titled “Substrate Transport Apparatus with Multiple Movable Arms Utilizing a Mechanical Switch Mechanism,” and U.S. Patent No. 7,648,327, issued January 19, 2010, the disclosures of which are incorporated herein by reference in their entirety. The operation of said transport arms may be independent of one another (e.g., the extension / retraction of each arm is independent of the other arms), may be actuated via a loss motion switch, or may be operatively linked in a suitable manner so that said arms share at least one common drive axis. In other aspects, the above-mentioned delivery arms may have any other stored arrangements, such as a frog-leg arm 316 (Fig. 4a) configuration, a leap frog arm 317 (Fig. 4e) configuration, a di-symmetric arm 318 (Fig. 4d) configuration, etc.Suitable examples of delivery arms are U.S. Patent 6,231,297 issued May 15, 2001; 5,180,276 issued January 19, 1993; 6,464,448 issued October 15, 2002; 6,224,319 issued May 1, 2001; 5,447,409 issued September 5, 1995; 7,578,649 issued August 25, 2009; 5,794,487 issued August 18, 1998; 7,946,800 issued May 24, 2011; and 6,485,250 issued November 26, 2002. It can be found in 7,891,935, issued on February 22, 2011, and in U.S. patent application number 13 / 293,717, titled “Dual Arm Robot,” filed on November 10, 2011, and in 13 / 270,844, titled “Coaxial Drive Vacuum Robot,” filed on October 11, 2011, the disclosures of these applications are incorporated herein by reference in their entirety.
[0123] Referring further to FIGS. 2-4e, the robot manipulators 306, 311, and 400 described herein transport substrates S (see FIGS. 4a and 4b) between points in space, such as the substrate holding stations STN1-STN6 shown in FIG. 5a. To achieve the transport of substrates S, a motion control algorithm is executed in a suitable controller of an autonomous material handling platform 300, such as a robot controller (also referred to as a robot manipulator controller) 319, 323, 422, 423A-423C, and 810 (see FIGS. 2, 4 and 8a) connected to the robot manipulators 306, 311, and 400. The above motion control algorithm defines a desired substrate path in space, and the position control loop calculates desired control torques (or forces) to apply to each robot actuator responsible for moving their respective robot degrees of freedom in space.
[0124] The robot manipulators 306, 311, and 400 (which may be referred to as autonomous systems) are expected to perform a repetitive task of continuously transferring substrates S, and the robot manipulators require environmental conditions associated with processing such substrates. It is advantageous for the method and apparatus provided by the embodiments disclosed to monitor the robot manipulators (or other autonomous equipment of the autonomous material handling platform 300) over time and to determine (predictively diagnose) whether each robot manipulator 306, 311, and 400 can operate within expected parameters to handle its primary task, such as transporting and transferring substrates between substrate holding stations STN1-STN6.
[0125] According to the embodiments disclosed, for example, the health assessment of the robot manipulator 306, 311, 400 is performed by generating a base statistical signature (e.g., a statistical representation or baseline of the behavior of a given variable operating under typical environmental conditions) that defines each dynamic performance variable output by the robot manipulator 306, 311, 400 for a set of base moves / motions (the terms move and motion are used interchangeably herein) 820, 820A, 820B, 820C (see FIG. 8a) of the robot manipulator 306, 311, 400. The above base statistical signature is generated by registering the above-mentioned base statistical signature using a registration system 801R (which may be formed by or reside within any suitable storage unit such as storage unit 801) which is commutably coupled to a suitable controller of an autonomous material handling platform 300, such as controllers 319, 323, 422, 423A, 423B, 423C, 810, which specifies at least one dynamic performance variable output by the robot manipulator 306, 311, 400, executing a predetermined set of predetermined base motions.
[0126] Each of the above operational performance variables is specific to the autonomous system (such as the robot manipulators 306, 311, and 400), and this may be within a different group of autonomous systems (such as the group of autonomous systems forming the autonomous material handling platform 300) where the dynamic performance variables were acquired. As such, since each of the above dynamic performance variables is specific to one of the autonomous systems (of the group of autonomous systems), the base statistical signature of each autonomous system moves with each autonomous system. For example, the robot manipulator 306 placed in the standby section 301 of the autonomous material handling platform 300 has its own base statistical signature, and the robot manipulator 311 placed in the vacuum section 302 has its own base statistical signature.
[0127] If robot manipulator 311 was located within the aforementioned standby section 301, the base statistical signature of that robot manipulator 311 will still apply to that robot manipulator 311 when the robot manipulator 311 is placed within the standby section 301. In one aspect, the base statistical signature is associated with each autonomous system and / or the controller of that autonomous system in memory. Additionally, each robot manipulator may have operational characteristics that execute the base statistical signature of its respective robot manipulator. For example, robot manipulator 311 and other robot manipulators may be manufactured as robot manipulators of the same manufacturer and model. However, for example, due to manufacturing tolerances in robot drive systems and arm structures, the base statistical signature for robot manipulator 311 may not be applicable to other robot manipulators, and vice versa. Likewise, the base statistical signature for each robot moneylender travels with its respective robot moneylender (e.g., base statistical signature C for robot moneylender 311). pkbase ... moves with Robot Moneyplayer 311 and is unique to that Robot Moneyplayer, and base statistical signature C for Robot Moneyplayer 306 pkbase (it moves with robot manipulator 306 and is unique to that robot manipulator). Therefore, each device such as robot manipulator 311 is unique, and each normalized value or base statistical signature / value C for each predetermined base movement 501, 502, 503 of the predetermined motion base set 820, 820A-820C pkbaseAnd each mapped in situ process movement 501', 502', 503' of the other predetermined motion sets 830, 830A-830C is uniquely correlated only to the said unique device, and the said predetermined performance deterioration rate (such as the linear trend model LTM - see Fig. 11) is uniquely correlated only to the said unique device.
[0128] In one aspect, a system (such as the autonomous material handling platform 300 illustrated in FIG. 3) comprises or is provided with a plurality of different unique devices connected to and, for example, connected to the transfer device 311 (such as the aligner 307 listed in Table 1 and illustrated in FIG. 2, the robot manipulator 306, the fan filter unit 308, etc.), wherein each different unique device from the plurality of different unique devices App(i) (schematically indicated in FIG. 2a as each different unique device App1-Appn) corresponds to different corresponding normalized values C for each base movement 501, 502, 503 of the predetermined motion base set 820, 820A-820C. pkBasei (C pkBase(1-n) (including) and other normalized values C for each mapped in-situ process movement 501', 502', 503' of other predetermined motion sets 830, 830A-830C that correspond only to different corresponding unique devices App1-Appn from the plurality of different unique devices (i). pkOtheriIt is provided with. In one embodiment, each (or at least one) different unique device App1-Appn from a plurality of different devices App(i) has a configuration common to another of the different unique devices App1-Appn. For example, a robot manipulator 306 may have a configuration common to a robot manipulator 311. In other embodiments, each (or at least one) different unique device App1-Appn from a plurality of different unique devices App(i) has a configuration different from another of the different unique devices App(i). For example, the aligner 307 has a configuration different from the robot manipulator 306.
[0129] The dynamic performance variables of each autonomous device and / or system may be measured directly (i.e., continuous monitoring variables) or derived from available measurements (i.e., derived variables). Examples of such dynamic performance variables include the following:
[0130] Mechanical or electrical power;
[0131] Mechanical work;
[0132] Robot end-effector acceleration;
[0133] Motor PWM Duty: The PWM duty of the motor is a percentage of the input voltage supplied to each motor phase at a given time. The duty cycle in each of the motor phases is available to the health-monitoring and fault-diagnostic system;
[0134] Motor Current: Motor current represents the current flowing through each of the three phases of each of the motors. Motor current can be obtained as an absolute value or as a percentage of the maximum current. When obtained as an absolute value, the motor current has units of Amps. Motor current values can, in turn, be used to calculate motor torques using motor torque-current relationships;
[0135] Actual position, velocity, and acceleration: These are the position, velocity, and acceleration of each motor axis. For the rotation axes, the above position, velocity, and acceleration values are in degrees, degrees / sec, and degrees / sec, respectively. 2 (degrees / sec 2 ) units. For the translation axes, the above position, velocity, and acceleration values are mm, mm / sec, and mm / sec, respectively. 2 It is a unit;
[0136] Desired position, velocity, and acceleration: These are the position, velocity, and acceleration values commanded by the controller for the motor to have. These characteristics have units similar to the actual position, velocity, and acceleration mentioned above;
[0137] Position and velocity tracking errors: These are the differences between their respective desired and actual values. These characteristics have units similar to the actual position, velocity, and acceleration mentioned above;
[0138] Settling time: This is the time it takes for position and velocity tracking errors to stabilize within specified windows at the end of the motion;
[0139] Encoder Analog and Absolute Position Outputs: Motion position is determined by encoders, which output two types of signals—analog signals and absolute position signals. The analog signals are sine and cosine signals in mVolt units. The absolute position signals are non-variable integer values representing the number of passed analog sine cycles or integer multiples of analog sine cycles. Typically, digital outputs are read upon power-up, and thereafter, the axis position is determined solely from the analog signals;
[0140] Gripper Status: This is the status of the gripper—open or closed. In the case of a vacuum-operated edge-contact gripper, it is the blocked / unblocked state of one or more sensors;
[0141] Vacuum System Pressure: This is the vacuum level measured by the vacuum sensor. This is an analog sensor, and its output is binarized by an analog-to-digital converter. In the case of a suction gripper, the vacuum level indicates whether a wafer has been grasped;
[0142] Substrate-Presence Sensor Status: In a passive grip end effector, the wafer presence sensor output is a binary output. In a vacuum-operated edge-contact grip end effector, wafer presence is determined from the output statuses of two or more sensors, and the output statuses are binary;
[0143] Mapper Sensor Status: This is the status of the mapper sensor, indicating whether it is blocked or not at any given moment;
[0144] Substrate Mapper / Aligner Detector Light Intensity: This is a measure of the light intensity detected by the light detector of the substrate mapper or aligner. This signal is typically available as an integer value (e.g., which can range from 0 to 1024);
[0145] Substrate mapper sensor position capture data: This is an array of robot position valves whose state the mapper sensor changes;
[0146] Empty valve state: This is the commanded state of the vacuum valve. This specifies whether the solenoid operating the vacuum valve will be supplied with power;
[0147] Voltage at fuse output terminals: The voltage at the output terminals of each fuse in the motor control circuit is monitored. A blown fuse results in a low output terminal voltage;
[0148] Substrate alignment data: These are the angular orientation of the substrate alignment point and the substrate eccentricity vector reported by the aligner;
[0149] Position data at the transition of external substrate sensors: In some cases, the atmospheric and vacuum sections of the tool may be equipped with optical sensors that detect the leading edge and trailing edge of the substrate carried by the robot. Robot position data corresponding to these events is used for the one-time recognition of the eccentricity of the substrate on the robot end-effector;
[0150] Substrate cycle time: This is the time taken for a single substrate to be processed by the above tool, and is usually measured under constant flow conditions;
[0151] Minimum Ambient Pressure: This is the pressure measured by the pressure sensor within the atmospheric section of the above tool.
[0152] Special examples of continuous monitoring variables include the following:
[0153] name Shorten the unit Long unit T1 actual location deg angle T2 actual location deg angle Z actual position m meter T1 actual speed deg / sec Angle per second T2 actual speed deg / sec Angle per second Z actual speed m / sec meters per second T1 actual acceleration deg / sec2 Angle per square second T2 actual acceleration deg / sec2 Angle per square second Z actual acceleration m / sec2 meters per square second T1 Acceleration Command deg / sec2 Angle per square second T2 Acceleration Command deg / sec2 Angle per square second Z Acceleration Command m / sec2 millimeters per square second T1 position error deg angle T2 position error deg angle Z position error deg angle T1 actual torque N Newton-meter T2 Actual Torque N Newton-meter Z actual torque N Newton-meter T1 modeled torque N Newton-meter T2 modeled torque N Newton-meter Z-modeled torque N Newton-meter T1 actual current A ampere T2 actual current A ampere Z actual current A ampere Bus motor voltage V bolt Bus 24V rail V bolt Bus 12V rail V bolt Bus 5V rail V bolt Bus 3.3V rail V bolt Core temperature C Celsius FPGA core supply voltage V bolt FPGA I / O supply voltage V bolt Processor core supply voltage V bolt Processor I / O supply voltage V bolt DDR supply voltage V bolt T1 temperature C Celsius T2 temperature C Celsius Z temperature C Celsius T1 servo status bitwise Bitwise T2 servo status bitwise Bitwise Z servo status bitwise Bitwise T1 Encoder CRC Error Counter count bar T2 Encoder CRC Error Counter count bar Z Encoder CRC Error Counter count bar T1 Encoder Warning Flag Counter count bar T2 Encoder Warning Flag Counter count bar Z Encoder Warning Flag Counter count bar T1 Encoder Error Flag Counter count bar T2 Encoder Error Flag Counter count bar Z encoder error flag counter count bar T1 Command Location deg angle T2 Command Location deg angle Z Command Location m meter Radial Command Location m meter Radial Command Position Arm B m meter Theta Command Location deg angle Theta Command Position Arm B deg angle Radial position error m meter Radial position error arm B m meter Tangent position error m meter Tangent position error arm B m meter Command Composite Acceleration g g Command Composite Acceleration Arm B g g Actual compound acceleration g g Actual compound acceleration arm B g g CPU Core 0 Utilization % Percentage CPU Core 1 utilization % Percentage Total system RAM B byte Available system RAM B byte System RAM used B byte Percentage of system RAM used % Percentage Total disk space 0 B byte 0 available disk space B byte 0 disk space used B byte Used percentage of disk space: 0 % Percentage Processor PCB temperature C Celsius RTC Battery Low Flag Bool Boolean Number of controller boot cycles count count Controller hardware uptime days me Total cycle count count count Fan controller internal temperature C Celsius Fan 0 speed RPM RPM Fan 0 error flag Bool Boolean Fan 1 speed RPM RPM Fan 1 error flag Bool Boolean
[0154] Table 3: Continuous monitoring variables. Here, T1 and T2 are the robot manipulator drive rotation axes (there may be more or fewer rotation drive axes than two); Z is the robot drive Z axis; CPU is the robot controller (such as controllers 319, 323, 422, 423A-423C, 800); Fan 0, Fan 1 are the various fans of the robot manipulator; Theta is the rotation of the robot manipulator arm; and R is the extension of the robot manipulator arm.
[0155] Special examples of derived variables include the following:
[0156] name Shorten the unit Long unit T1 Servo state duration days me T2 servo state duration days me Z servo state duration days me T1 servo status bool Boolean T2 servo status bool Boolean Z servo status bool Boolean T1 mechanical power W watt T2 mechanical power W watt Z mechanical power W watt T1 electrical power W watt T2 electrical power W watt Z electrical power W watt T1 motor efficiency % Percentage T2 motor efficiency % Percentage Z motor efficiency % Percentage T1 Increased Energy J line T2 Increased Energy J line Z increase energy J line T1 increment position deg angle T2 increment position deg angle Z increasing position m meter Location loop average execution time us microseconds Location loop maximum execution time us microseconds Location loop average signal delay time us microseconds Location loop maximum signal delay time us microseconds Current loop average execution time us microseconds Current loop maximum execution time us microseconds IO loop average interval time ms milliseconds IO loop maximum interval time ms milliseconds Average execution time of IO loops ms milliseconds Maximum execution time of IO loop ms milliseconds T1 Encoder CRC Error Flag count bar T2 Encoder CRC Error Flag count bar Z encoder CRC error flag count bar T1 Encoder Warning Flag count bar T2 Encoder Warning Flag count bar Z encoder warning flag count bar T1 encoder error flag count bar T2 encoder error flag count bar Z encoder error flag count bar
[0157] Table 4: Derived Variables Such dynamic performance variables are calculated from raw or direct measurements such as motor position, speed, acceleration and control torque.
[0158] The predetermined base moves 501, 502, and 503 of the predetermined motion base sets 820, 820A-820C include at least one common base move of a statistically defined number that defines the base motion type (e.g., a move generated from sufficient sample moves collected to form a baseline and define a statistically meaningful batch). For example, the (each) motion base sets 820, 820A-820C (see FIG. 8a) for each base move 501, 502, 503 (e.g., base move 501 has motion base set 820A, base move 502 has motion base set 820B, and base move 503 has motion base set 820C) are the substantially minimum number N of moves sufficient to provide a statistically meaningful standard deviation based on a convergence criterion defined to define the motion base sets (or move sets) 820, 820A-820C for specific robot manipulators 306, 311, 400. min (See Fig. 6) (e.g., sample size). As such, each dynamic performance variable is specified for each robot manipulator 306, 311, 400 and is output by that robot manipulator.
[0159] The predetermined base movements 501, 502, and 503 of each predetermined motion base set 820, 820A-820C comprise a plurality of different base motion types, each of which is executed by the transfer device 306 in a number of common motions statistically defined for each base motion type. Each of the different base motion types has at least one different corresponding torque command feature and a position command feature defining a different common motion for each base motion type. In one embodiment, the predetermined base motion set 820, 820A-820C may be one or more movement / motion types. For example, each movement 501, 502, and 503 within the base motion set 820, 820A-820C may be simple movements or complex (e.g., mixed) movements defined by the torque command and position command defining each movement.
[0160] Simple movement is linear movement between two points (from point 0 to point 1 as illustrated in FIG. 5c) or movement within a circular arc between two points along one of the theta axis, extension axis, or Z axis of robot manipulators 306, 311, 400 (from point 1 to point 2 as illustrated in FIG. 5c) (e.g., 1 degree of freedom).
[0161] A complex or mixed movement is a movement in which two or more simple movements are mixed together as illustrated in FIG. 5b, wherein the movement extends from point 0 to point 2 by having a mixed path adjacent to point 1 that combines two straight lines moving from point 0 to point 1 and from point 1 to point 2 along at least two of the theta axis, extension axis, or Z axis of the robot manipulators 306, 311, 400 (e.g., two or more degrees of freedom).
[0162] Each of the above motion base sets 820, 820A-820C may also be defined by the positions of the movements within the set (e.g., the start and completion points of the movements), the load parameters of the movements within the set (e.g., whether the robot manipulators 306, 311, 400 are loaded (transporting substrates) or unloaded (not transporting substrates), and / or dynamic conditions at the initial and / or final positions of the movements (e.g., motion / stop, stop / stop, stop / motion, motion / motion, etc.). For example, referring to the complex movement in FIG. 5b, dynamic condition point 0 is stopped and the dynamic condition at point 2 is stopped. Referring to the two simple movements in FIG. 5c, the dynamic condition at point 0 is stopped and the dynamic condition at point 1 is moving; on the other hand, the dynamic condition at point 2 is stopped. As described above, the movement types Although robot manipulators with 1, 2, or 3 degrees of freedom have been described, it should be understood that the types of movement may include movements created using a suitable number of degrees of freedom or a single degree of freedom (as in vacuum pumps, board aligners, etc.).
[0163] Each move type is the minimum number of moves N that statistically define each move type. min Execute. For example, each dynamic performance variable or motion type can be represented in a histogram format as follows:
[0164]
[0165] Here, s is the base movement / motion signal provided in Tables 5 through 7. Signals s0 through s nare signals having scalar outputs and must be comparable across different template moves (which can also be referred to as base moves), that is, compare motor energy with respect to the baseline across different move types. n +1 to s n +1+mi The signals are vector output signals from Table 8 and cannot be compared globally across different template motion types marked with i.
[0166] metrics explanation unit peak Tracking error angle RMS Tracking error angle peak Torque error Newton-meter RMS Torque error Newton-meter peak electric current ampere RMS electric current ampere peak Duty cycle as a percentage of available torque Percentage peak mechanical power watt average Motor efficiency Percentage total mechanical energy line average Winding temperature Celsius total Vibration power over the configurable Hz range dB peak Vibration power dB / Hz peak Vibration Power Crest Factor dB / Hz
[0167] Table 5: Signals induced for each motor over the base movement, providing a scalar output.
[0168] metrics explanation unit condition peak Settling time candle Precision setting peak Set ring radial error meter Precision setting peak Settling tangent error meter Precision setting peak Handoff complex error meter At: R EX Change: Z State peak Tangent tracking error meter Change: R state RMS Tangent tracking error meter Change: R state peak Organized acceleration g
[0169] Table 6: Signals induced across the bass shift, per arm or end-effector, equipped with scalar output
[0170] metrics explanation unit peak Motor bus voltage drop bolt peak Accumulated motor mechanical power watt total Accumulated motor mechanical energy line
[0171] Table 7: System signals induced over base shift, equipped with scalar output
[0172] explanation unit Position error, per motor angle Actual torque, per motor Newton-meter Tangent error, per arm or end-effector meter Radial error, per arm or end-effector meter
[0173] Table 8: Signals induced across base shifts, equipped with vector outputs. These vector output signals have signals at each time sample along the trajectory, and therefore the number of these signals differs between different shifts, and there is no physical meaning in the evaluation at time samples from one shift to another. The base shift (type) index is indicated by i, and the history of a given index is indicated by j.
[0174] In this example, the last evaluated base move is as follows.
[0175]
[0176] And in this example, the third and final evaluated base shift is equal to the following equation.
[0177]
[0178] Referring to FIGS. 5a-5c, base moves such as base moves 501, 502, and 503 in each base move 820, 820A-820C may also be referred to as template moves. Base moves 501, 502, and 503 are repeated moves along a unique path. The base moves 501, 502, and 503 may consist of simple moves or complex moves as described above.
[0179] To evaluate system performance degradation and performance trends, feature data is analyzed along the unique path of the base shift with respect to the baseline. The base shifts 501, 502, and 503 may be defined theoretically and / or empirically. For example, the theoretical base shift is based on the process of the process tool and the expected design configuration to analyze the expected shifts in operation, and is then executed at any time before or after the installation of the in-situ process tool.
[0180] Empirical base shifts can be generated from in-situ process shift commands as shifts of desired occurrence commonality into sufficient statistical features generated to have meaningful statistical values that stabilize between predetermined rates of change convergence boundaries shown in FIG. 6 (wherein N in FIG. 6). minis the minimum number of moves sufficient to provide a statistically meaningful standard deviation based on a given convergence criterion (e.g., sample size). Generating empirical base moves may be a two-part process (applied similarly to the empirical generation of base statistical signatures). For example, generating empirical base moves may include the following: accessing the base move command histogram 700 (see FIG. 7) and identifying base moves using commands (e.g., torque, position, boundary parameters, command trajectory path (including velocity and move duration), load conditions, etc.) that map to base moves 501, 502, and 503 (e.g., said base moves correspond to base moves within a configurable tolerance); And a step of accessing each dynamic performance variable output by each robot manipulator 306, 311, 400 for the mapped motion from a storage unit 840 of a suitable registration system 801R that registers predetermined operational data specifying at least one dynamic performance variable output by a robot manipulator to perform another predetermined motion set 830 determination (described below).
[0181] The generation of the above empirical base movement is performed in near real-time, executed in the background, and accesses the register 840 without accessing the associated bidirectional communication / data channels of the controllers 319, 323, 422, 423A, 423B, 423C, 800 and the autonomous material handling platform 300. The above repositioning movement command histogram 700 includes motions commanded by the robot manipulator controllers (such as controllers 319, 323, 422, 423A, 423B, 423C, 800), including repositioning process motions executed by their respective robot manipulators 306, 311, 400. The above-mentioned original position command histogram 700 may be registered in a suitable storage unit 700R (see FIG. 8a) of the robot manipulator controller (such as controllers 319, 323, 422, 423A, 423B, 423C, 800) or another suitable controller of the autonomous material handling platform 300. As described herein, the robot manipulator controller analyzes the mapped motions from periodic access to the motion histogram 700 in the storage unit 700R.
[0182] For example, referring also to FIG. 8a, the motion resolver 800 analyzes the in-place processor motion commands of the robot controllers 319, 323, 422, 423A-423C, 810 from the robot manipulators 306, 311, 400 (see FIG. 2 and 3), wherein the in-place process motions 501', 502', 503' (see FIG. 5a) executed by the transfer device are mapped to the predetermined base motions 501, 502, 503 of a predetermined motion base set (described below) (each of which defines a corresponding template motion so that the in-place process motion is mapped onto its respective template motion), and another predetermined motion set of the robot controllers 319, 323, 422, 423A-423C, 810 (below (Described) is defined using the mapped in-place process motions 501', 502', and 503'. For example, in-place process motion 501' is mapped to base motion 501, in-place process motion 502' is mapped to base motion 502, and in-place process motion 503' is mapped to base motion 503. In a manner similar to that described above, each in-place process motion 501'-503' is defined by at least one of a torque command and a position command from the device controller, wherein at least one of the torque command and the position command defines the in-place process motion at least one degree of freedom of the robot manipulator 306, 311, 400.
[0183] The motion resolver 800 may be included as a module within robot controllers 319, 323, 422, 423A-423C, and 810, and the motion resolver 800 may be a remote processor communicably coupled to the robot controllers 319, 323, 422, 423A-423C, and 810, or the motion resolver 800 may be a separate processor communicably linked to the robot controllers 319, 323, 422, 423A-423C, and 810.
[0184] The above motion resolver 800 iterates through the original process moves 501', 502', and 503', for example, the minimum number of moves N required as determined by the standard deviation convergence illustrated in FIG. 6. minThe original process moves 501', 502', and 503' are identified using [this method]. For example, as mentioned above, to generate a baseline (e.g., to establish base moves 501, 502, and 503), there must be sufficient samples collected to define a statistically meaningful batch. The number of samples required to generate a baseline depends on the physical properties of the variables being analyzed. For example, defining typical statistics (mean and standard deviation) for the mechanical work of a given axis of motion of robot manipulators 306, 311, and 400 may take longer than the peak control torque of the same axis performing the same motion. To improve this situation, the size of the baseline is defined based on the statistical analysis of the collected data. For example, the standard deviation can be calculated up to the point where its value stabilizes within several boundaries during baseline data collection, as shown in Fig. 6. In Figure 6, the standard deviation of a given variable can be plotted for equal magnitude. As that equal magnitude increases, the standard deviation tends to converge within certain boundaries. These boundaries can be defined a priori or, for example, calculated in terms of the actual data set when the rate of plot variation is below approximately + / -10% variation; however, any suitable convergence methods and / or percentage of variation may be used.
[0185] Referring again to FIGS. 8a, FIGS. 5, and FIGS. 8b, at least the minimum number of movements N required min The movements constituting (e.g., movements used to define the baseline) may be referred to as a predetermined motion base set 820. Each of the base movements 501, 502, and 503 has its own predetermined motion base set 820A, 820B, and 820C, which is unique to the base movements 501, 502, and 503.
[0186] Referring again to FIG. 5, FIG. 8a and FIG. 8b, in one aspect, once the motion resolver 800 identifies and analyzes the predetermined motion base sets 820A, 820B, and 820C for each base move 501, 502, and 503, the in-place process moves 501', 502', and 503' that map to one of the base moves 501, 502, and 503 (as described above) are included within their respective predetermined motion base sets 820A, 820B, and 820C, thereby causing their respective predetermined motion base sets 820A, 820B, and 820C to be updated. In another aspect, the in-place process motions 501', 502', 503' mapping to one of the base movements 501, 502, 503 to a predetermined motion base set 820A, 820B, 820C may form a different set of motion type sets different from the predetermined motion base set 820A, 820B, 820C. The different set of the updated predetermined motion base set and / or motion type sets may be referred to as another predetermined motion set 830.
[0187] As described herein, the other predetermined motion base sets 830A, 830B, and 830C for their respective in-situ process movements 501', 502', and 503' are compared to the motion base sets 820A, 820B, and 820C for their respective base movements 501, 502, and 503 with respect to health assessments and predictive diagnoses of autonomous systems being monitored like the motion manipulator 300.
[0188] As described above, for example, a health assessment of the robot manipulators 306, 311, 400 (or other suitable autonomous equipment of the autonomous material handling platform 300) is performed by generating a base statistical signature (e.g., a baseline or statistical representation of a given variable behavior operating under typical ambient conditions) that defines each dynamic performance variable output by the robot manipulators 306, 311, 400 for a set of base moves 820, 820A, 820B, 820C (see FIG. 8a) of the robot manipulators 306, 311, 400.
[0189] In one embodiment, baseline metrics are captured / identified using a suitable processor 810P of the autonomous material handling platform 300 (which is substantially similar to processor 105 in one embodiment). The processor 810P may be included as a module within the robot controllers 319, 323, 422, 423A-423C, 810, and the processor 810P may be a remote processor communicably coupled to the robot controllers 319, 323, 422, 423A-423C, 810 (and motion resolver 800), or the processor 810P may be an individual processor communicably linked to the robot controllers 319, 323, 422, 423A-423C, 810 (and motion resolver 800). The processor 810P is coupled to the registration system 801R in a loading / unloading manner, and in other aspects, the processor 810P includes the registration system 801R.
[0190] The above baseline metrics are captured / identified, for example, by calculating the probability density function (PDF) of the above baseline statistical signature, wherein the probability density function can be expressed as follows:
[0191]
[0192] Here is the dataset mean, x is a dynamic performance variable, and ε is the standard deviation. Figure 9 shows a typical Gaussian distribution with mean and standard deviation. The upper specification limit and lower specification limit (USL and LSL, respectively) are also defined in Figure 9.
[0193] The base statistical signature of each dynamic performance variable of each robot manipulator 306, 311, 400 (see FIGS. 2 and 3) is normalized for each different base movement type (movement type set to base value) which defines the nominal / baseline of each dynamic performance variable specific to each robot manipulator 306, 311, 400 for each different base motion type(s) / movement type set(s). For example, for each motion of the aforementioned predetermined set of motion bases, the (process capacity index C) defined by the probability density function PDF of each dynamic performance variable output by their respective robot manipulators 306, 311, 400 pkBase The base value (such as) is determined.
[0194] Generally, process capacity index C pk is defined as follows:
[0195]
[0196] Here is the standard deviation. is the average value of samples collected for each variable. The above processor capacity index C pk is a process capacity index C that captures the mean and standard deviation of a population sample large enough to provide meaningful statistical data for the baseline of each dynamic performance variable. pkThey can be used as metrics to express as. The upper specification limit USL and lower specification limit LSL can be determined in any suitable manner, such as defining the upper and lower specification limits USL, LSL as a function of the measured standard deviations of the respective robot manipulators 306, 311, and 400 being measured. For example, as follows:
[0197]
[0198]
[0199] Here, N can be an integer greater than 3, and so the above C pk Ensure that can be a number greater than 1. For example, if N = 6, the above baseline process capacity index C pkBase can be defined as follows:
[0200]
[0201] In one aspect, C pkBase It can be set to 2.0, and the average of the baseline is based theoretically or empirically on the data set to identify the upper and lower specification limits USL, LSL. + / - 6 It can be set up so that 99.9% of the sampled movements are captured (as illustrated in FIGS. 9 and 10). In other aspects, when limits, for example, torque limits, maximum settling time, etc. are well established, the upper and lower specification limits USL, LSL can be set based on each individual signal.
[0202] In one aspect, also referring to FIG. 2a, the corresponding normalized values C for each different unique device App1-Appn pkBase (1-n) and other values C pkOther (1-n)This is registered in any suitable controller, such as the controller of one of the above devices App1-Appn. The normalized values C that are uniquely associated with each of the different unique devices App1-Appn. pkBase(1-n) and other values C pkOther (1-n) As described in great detail herein, a corresponding performance deterioration rate (indicated, for example, by their respective linear trend models LTM, see FIG. 11) is compared for each of the different unique devices App1-Appn to determine the device-based performance deterioration rate. For example, each device App1-Appn has its own linear trend models LTM1-LTMn as illustrated in FIG. 11.
[0203] Once baseline metrics are established for each measurement variable (raw and derived), batches of in-situ process moves 501'-503' are sampled during the operation of their respective robot manipulators 306, 311, and 400. For example, in-situ process moves 501', 502', and 503' are generated by controllers such as controllers 319, 323, 422, 423A, 423B, 423C, and 810 to identify different statistical signatures specific to the monitored robot manipulators 306, 311, and 400. As described above, each dynamic performance variable for a set of in-situ process moves is mapped to its respective base move (e.g., base move type / type set(s)—see Equations 1, 2, and 3). As described above, the mapped in-place process motions 501', 502', and 503' are used to define different predetermined motion sets 830, 830A-830C of their respective robot manipulators 306, 311, and 400.
[0204] Together with baseline moves 501-503, the above-mentioned in-situ process moves 501'-503' process the (different) statistical signatures of each dynamic performance variable of each robot manipulator 306, 311, 400 for each different in-situ (different) move type / type set(s).
[0205] (For example, the above other predetermined motion sets 830, 830A-830C are mapped to their respective predetermined motion base sets 820, 830A-830C, and for each of the different in-situ movement types (which may be simple or complex movements), an in-situ (different) value C that defines the in-situ performance of each dynamic performance variable of their respective robot manipulators 306, 311, 400) pkOther It is normalized to. The above original (other) value C pkOther is a process capacity index defined by the probability density function PDF of each of the dynamic performance variables output by robot manipulators 306, 311, and 400 executing the mapped in-situ process motions 501'-503' of the other predetermined motion sets 830, 830A-830C. The in-situ (other) value C pkOther Refers to the upper and lower limits USL, LSL of the baseline to determine the position of another predetermined motion set with respect to the predetermined motion base set shown in FIG. 10 (wherein the other predetermined motion set is identified as a "new batch" and the predetermined motion base set is identified as a "baseline"). C pkOther is a process capacity index that can be defined as follows:
[0206]
[0207] C being evaluated here pkOther It is the number of repetitions of. The above normalized in-situ (other) value C pkOtherThe normalized base value C for each of the monitored dynamic performance variables, as for each movement type and cross movement type. pkBase It is compared to.
[0208] The above original (other) value C pkOther and the above base value C pkBase Comparison between them may be performed by the processor 810P or any other suitable controller of the autonomous material handling platform 300, wherein each robot manipulator 306, 311, 400 is the predetermined motion base set 820, 820A-820C and the other predetermined motion set 830, 830A-830C (and the corresponding original (other) value C pkOther and the above base value C pkBase It is a common transfer device for both. The above original (other) value C pkOther and the above base value C pkBase The comparison between them performs a health assessment of each dynamic performance variable being monitored for specific devices, such as the respective robot manipulators 306, 311, and 400, by providing a tracking of how many performance variables deviate from or drift away from their baseline (see Fig. 10). The health assessment for each performance variable can be defined as the relative variation from its baseline as follows:
[0209]
[0210] This is a 100% evaluation of the above in-situ (other) value C pkOther and the above base value C pkBaseThis means that it indicates a complete statistical match between them. Equation 10 above represents one example of evaluating a given dynamic performance variable. In other forms, other methods of measuring the evaluation may be used, such as measuring the number of occurrences falling outside the upper and lower baseline limits, USL and LSL. Figure 10 illustrates an example of calculating a health evaluation in terms of its statistics for a given dynamic performance variable. In the example illustrated in Figure 10, 20% of the batch data samples lie outside the baseline range, and the in situ (other) value C pkOther Penalties apply to this.
[0211] Referring to FIGS. 11 and 12 as well as FIG. 10, the health evaluation of each dynamic performance variable of the respective robot manipulators 306, 311, and 400 is the above-mentioned original (different) value C pkOther This above base value C pkBase It can be defined in terms of the degree of deviation from. The said degree of deviation can be defined in terms of defined thresholds such as "path" and "error," and as described below, here "warning" may refer to "attention is requested" and "error" may refer to "immediate action is required." The said original (other) value C pkOther (and the above base value C pkBase Another aspect of tracking is that this tracking provides trend analysis, that is, it can estimate or estimate when the corresponding dynamic performance variable is predicted to reach a degree of variation at a different level.
[0212] Determining the amount by which each dynamic performance variable deviates from or is displaced from its baseline provides trend data (TD) for each dynamic performance variable, wherein said trend data (TD) defines the deterioration tendency of each dynamic performance variable. said trend data (TD) may be registered within any suitable register (TDR) of the autonomous material handling platform 300. FIG. 11 illustrates an example trend data chart of a dynamic performance variable; wherein the said in-situ (different) value C for predetermined points in time from different batch samples pkOther and the above base value C pkBase Each evaluation A1-An from the comparison is plotted on the chart above.
[0213] In FIG. 11, the slanted lines represent linear trend models LTM, LTM1-LTMn, which can be obtained in a suitable manner, such as using the least squares method; on the other hand, in other aspects, any suitable trend model may be used. For example, trend data defining the performance deterioration trend of each of the robot manipulator 306 (or any other suitable device of the autonomous material handling platform 300 (see FIG. 2)) and a number of different unique devices App1-Appn (see FIG. 2a) of the autonomous material handling platform 300 is registered in the register of any suitable controller / processor of the autonomous material handling platform 300 (e.g., the controller of each device or tool controller 314 or processor 810P). In one aspect, the processor 810P combines performance degradation trends corresponding to the transfer device, such as the transfer device 306, and each of a plurality of different unique devices App1-Appn of the autonomous material handling platform 300 to determine system performance degradation trends that define the performance degradation of the autonomous material handling platform 300.
[0214] Referring to the linear trend model LTM, this linear trend model LTM (which can represent unique devices such as the robot manipulator 306, robot manipulator 311, aligner 304, power supply PS of the autonomous material handling platform 300, etc.) is at time t warn It can be used to predict the estimated time (or cycle) for the evaluation metric to reach a defined warning threshold. Similarly, the time error can be estimated as the time (or cycle) for the robot manipulator 306, 311, 400 operation to reach a point where it is not recommended to continue. As can be seen in Fig. 11, the linear trend model LTM1-LTMn is determined for each different unique device App1-Appn. The linear trend model LTM1-LTMn can indicate the overall health of the system (such as the autonomous material handling platform 300) and the health of each of the different unique devices App1-Appn. Referring also to FIG. 2, for example, linear trend model LTM1 can correspond to power supply PS, linear trend model LTM2 can correspond to robot manipulator 306, linear trend model LTM3 can correspond to robot manipulator 311, and linear trend model LTMn can correspond to aligner 307.
[0215] As can be seen in FIGS. 11 and 12, the trend data TD may also provide health assessment warnings to be provided to the operator of the robot manipulator 306, 311, 400, for example, through any suitable display 140. For example, any suitable controller of the autonomous material handling platform 300, such as a processor 810P that may be separate from or included with the controllers 319, 323, 422, 423A, 423B, 423C, 810, may include a trend / assessment unit 870 (Fig. 8a) configured to transmit predetermined signals to display the health assessment of the robot manipulator 306, 311, 400 to the operator. In other embodiments, the trend / evaluation unit 870 may be part of the controllers 319, 323, 422, 423A, 423B, 423C, and 810. For example, the processor 810P may transmit a "warning" indication when the trend data TD reaches a first predetermined evaluation value WS, or may be visually visible, for example, in yellow; may provide an "error" indication in red when the trend data TD reaches a second predetermined evaluation value ES (for example, this is smaller than the first predetermined evaluation value WS); and may provide a "normal" indication (for example, all dynamic performance variables are within predetermined operating limits) in green when the trend data is above the first predetermined evaluation value WS. In other forms, the operating status of the autonomous system (e.g., normal, warning, and error) may be provided audibly, visually, or in any other suitable manner.
[0216] In one embodiment, the processor 810P aggregates the dynamic performance variables with the highest deterioration trends (e.g., lowest percentage evaluations) among at least one dynamic performance variable output by the transfer device and predicts the occurrence of the transfer device having performance below a predetermined performance state. For example, the overall health of robot manipulators 306, 311, and 400 can be measured as a worst-case evaluation across all dynamic performance variables monitored within a given batch of data samples. For example, five dynamic performance variables Var1–Var5 (e.g., T1 actual position, Z actual acceleration, bus motor voltage, T2 temperature to exemplify dissimilar variables being compared, and Theta command position) are measured and compared against their respective baselines, in which case the resulting evaluation values are as follows:
[0217] variable evaluation Var1 95% Var2 92% Var3 89% Var4 96% Var5 70%
[0218] Table 9: Evaluation Values In the example above, the evaluation of dynamic performance variable Var5 is the lowest among the five dynamic performance variables Var1-Var5, and the evaluation of the dynamic performance variable Var5 can be used to represent the overall current health evaluation of robot manipulators 306, 311, and 400 whose health is monitored by all five dynamic performance variables Var1-Var5. Since the evaluation can be compared globally for all of these entities based on the fact that the evaluations are measurements relative to their respective baselines, this can be done independently of the physical properties and meanings of each of these dynamic performance variables Var1-Var5.
[0219] As an example of the comparison of the performance variables described above, the processor 810P compares the performance degradation trend of the transfer device 306 with the performance degradation trend of each of a plurality of different unique devices App1-Appn, and determines whether the performance degradation trend of the transfer device 306 or the performance degradation trend of the other of the plurality of different unique devices App1-Appn is the dominant performance degradation trend, and whether the dominant performance degradation trend is the cause of the determination of the performance degradation trend of the system. For example, at time t s In this case, the linear trend model LTM2 for the robot manipulator 306 has the lowest evaluation, and in this case, this lowest evaluation is considered the overall health of the autonomous material handling platform 300 as described above with respect to Table 9. Over time, other linear trend models, such as the linear trend model LTM1, may show a faster rate of performance deterioration. In this example, for instance, the overall health of the autonomous material handling platform can be determined based on the linear trend model LTM1 at time t0, for instance, in this case, at time t warnLTM1 An alert is generated based on the linear trend model LTM1 at time t. errorLTM1 Errors are generated based on the linear trend model LTM1.
[0220] While the overall health of the above-described autonomous material handling system can be determined by a linear trend model having the lowest evaluation value at any given point in time, said linear trend models also provide indications or fingerprints regarding which device App1-Appn is the cause or primary contributor of system errors or warnings. For example, the power supply PS can affect other devices App1-Appn, such as by failing to apply sufficient voltage to the robot manipulator 306 (corresponding to the linear trend model LTM2). As can be seen in FIG. 11, at time t as a result of the deterioration of power supply PS performance warnLTM1 A warning may occur at time t regarding the deterioration in the performance of Robot Manipulator 306. warnLTM2 A warning may occur; however, the robot manipulator 306 is functioning properly, but the power supply PS is supplying insufficient voltage to the robot manipulator 306. These two warnings are indicators that the power supply PS and the robot manipulator 306 must be checked for repair, and suggest that there may be some correlation between the deterioration in the performance of the power supply PS and the deterioration in the performance of the robot manipulator 306.
[0221] In another aspect, referring to FIG. 5a and FIG. 8a, aspects of the disclosed embodiments may provide the health of the system as a combined aggregate feature and health prediction. It should be noted that the combined aggregate feature and health prediction of the system is different from combining / aggregating different deterioration trends of the components of the system to determine an overall system deterioration trend. For example, the combined aggregate feature and health prediction To determine the deterioration trend of a system equipped with multiple devices, they may be considered similar, in which case the system and its multiple devices are treated as a single unique device, whereas the deterioration trends for each unique device of the system are also determined as described above. In this embodiment, the base moves 501-503 and the in-place process motions 501'-503' correspond to their respective unique devices as described above. The base moves 501-503 and the in-place process motions 501'-503' may differ for each different device of a common type (e.g., the base moves 501-503 and in-place process motions 501'-503' of robot moneyplate 306 may differ from the base moves 501-503 and in-place process motions 501'-503' of robot moneyplate 311). Base motion sets 820, 820A-820C and other predetermined base motion sets 830, 830A-830C for a unique system (such as the autonomous material handling platform 300) can be determined by base motion set 890 (see FIG. 8a), wherein the base motions in the base motion set 890 are determined by combining a plurality of one or more base motions 501-503, wherein each of the plurality of one or more base motions is uniquely correlated to a unique device (e.g., power supply, robot manipulator, wafer sensors, etc.) (such as those described in Tables 1 and 2 above) communicably connected to form a single aggregate motion 890AG. The single aggregate motion is a unique system (such as the autonomous material handling platform 300) and (of each device operating in the single aggregate motion) Candidates related combined association dynamic performance variables (e.g., It is uniquely correlated with ), and in this case is a scalar value and and is a vector value, which is the system performance normalized value C pkbase (system μ devices) It is intended to generate and regarding the mapped motions Another value C uniquely related to the system of the device(s) pkOther (system μ devices) It is intended to generate.
[0222] Referring to FIG. 14, in which a device (such as those listed in Tables 1 and 2) is placed within a system (such as the autonomous material handling platform 300), the health determination of the system can be generated by repeating the system health determination (Fig. 14, block 1400), wherein the repetition of the system health determination comprises (1) repeating the determination of deterioration trends (such as those indicated by the linear trend models LTM, LTM1-LTMn) for each device of the system (or at least for a replaced device) and determining the overall system health by combining the device deterioration trends to control one of the deterioration trends, for example, as described with respect to Table 9 (Fig. 14, block 1401); and (2) determining a new system aggregate deterioration trend of the combined aggregate features as described above (Fig. 14, block 1402); (3) If the replacement device has improved or reduced the overall system deterioration trend and the new device has reduced the deterioration trend, the device is replaced again, and / or the devices are mixed and matched to improve the overall system deterioration trend (Fig. 14, Block 1403).
[0223] In one aspect, weighting processing of the above deterioration trends (Fig. 15, Block 1500) may be applied to the linear trend models LTM, LTM1-LTMn of each device of the system by a suitable processor of the system (such as the tool controller 314). When applying weighting processing, the tool controller 314 determines whether the deterioration trend of any one or more devices is dominant (e.g., the greatest deterioration) or otherwise indicates failure at a predicted time of failure outside a predetermined time range of desired time; or any one of the one or more devices may be identified as the first device predicted to fail, and the range (time range) between the first device predicted to fail and the last device predicted to fail may be determined (Fig. 15, Block 1502). A history of past failures (if any) may also be determined and stored in the memory of the system and (if any) examined by the tool controller 314 to determine which devices are most likely to fail first (Fig. 15, Block 1503). From the determination made above, the tool controller 314 may determine whether the failure frequency of a device does not match that of the system (e.g., the failure frequencies of other devices) (Fig. 15, Block 1504). The tool controller 314 may also identify device characteristics related to system performance (e.g., whether the system can operate with or cannot operate with a failed device) (Fig. 15, Block 1505). In one aspect, device characteristics related to system performance may be categorized as critical (such as when the system cannot function without the device) or routine (when the system can function without the device).The above device characteristics may include, but are not limited to, the primacy of the device, the difficulty in finding a replacement for the device, the accessibility of the device within the system (whether the device is easily accessible for replacement / whether it is difficult to access and difficult to replace), the packaging of the device (e.g., whether a failed motor in a robot manipulator will require the replacement of the robot manipulator, whereas a failed power supply will require the replacement of only the power supply), or other factors that may affect system downtime and / or the availability of replacement devices.
[0224] The weights given to the deterioration trends for each device may be determined, for example, by the tool controller 314, based on the device characteristics related to the failure frequency of the device and system performance. Weighting of device deterioration trends improves or reduces the impact of said device deterioration trends on the overall deterioration trend of the system, in which case the overall system health assessment is based on the weighted deterioration trends of each of the devices of said system.
[0225] As a non-restrictive example, linear trend models corresponding to recently replaced / repaired devices may be weighted less than devices that have been in service for some time, so that the recently replaced / repaired devices have less influence on the overall system health assessment than devices that have been in service for a longer period. In another aspect, the linear trend models LTM, LTM1-LTMn may be weighted so that linear trend models for devices known to fail frequently do not contribute to the overall system health assessment, or contribute only to a limited extent. In another aspect, the system health assessment may not include any weighting factors applied to the linear trend models LTM, LTM1-LTMn.
[0226] Now, referring to FIGS. 2, 3, 5A, 8A, 8B, and 13, the exemplary health assessment operation will be described according to the appearance of the disclosed embodiment. Predetermined operation data is registered using a registration system 801R, which is communicably coupled to device controllers 319, 323, 422, 423A-423C, and 810 (Fig. 13, Block 1300). The predetermined operation data embodies at least one dynamic performance variable output by the transfer device executing a predetermined motion base set 820, 820A, 820B, and 820C of predetermined base motions. Base value C using, for example, the processor 810P, which is communicably coupled to the registration system 810R. pkBase This is determined (Fig. 13, Block 1310). The base value C above. pkBase It is executed by the probability density function PDF of each dynamic performance variable output by the transfer device 306, 311, 400 for each motion among the above-determined motion base sets 820, 820A, 820B, and 820C.
[0227] The above-mentioned in-place process motions 501'-503' are analyzed by, for example, a motion resolver 800, which is communicably coupled to the device controllers 319, 323, 422, 423A-423C, and 810 (Fig. 13, Block 1320). The in-place process motions 501'-503', executed by the transfer device 306 in correspondence with the analyzed in-place process motion commands, are mapped to the predetermined base motions 501-503 among the predetermined motion base sets 820, 820A, 820B, and 820C. Other predetermined motion sets 830, 830A, 830B, and 830C of the transfer device are defined using the mapped in-place motions 501'-503' (Fig. 13, Block 1330).
[0228] Predetermined operation data specifying at least one dynamic performance variable output by the transfer device executing the other predetermined motion set is, for example, registered by the registration system 801R (Fig. 13, Block 1340). The processor 810P specifies another value C defined by the probability density function PDF of each of the dynamic performance variables output by the transfer device executing the mapped in-place process motions 501'-503' of the other predetermined motion set 830, 830A-830C. pkOther Determines (Fig. 13, Block 1350).
[0229] The above other value C pkOther and the above base value C pkBaseFor each dynamic performance variable output by the transfer device corresponding to the predetermined motion base set and the other predetermined motion set, respectively, it is compared, for example, by the processor 810P (Fig. 13, block 1360), wherein the transfer device is a transfer device common to both the predetermined motion base set and the other predetermined motion set. The health of the transfer device is evaluated based on the comparison as described above, and any suitable health evaluation notice may be transmitted to the operator of the autonomous material handling platform 300 as described above.
[0230] According to one or more aspects of the disclosed embodiments, a method for health assessment of a system including a transfer device is:
[0231] A step of registering predetermined operation data that implements at least one dynamic performance variable output by a transfer device executing a predetermined set of predetermined motion bases of predetermined base motions, using a registration system communicably coupled to a device controller;
[0232] A base value (C) defined by the probability density function of each of the dynamic performance variables output by the transfer device for each motion of the above-determined set of motion bases. pkBase A step of determining ) using a processor communically coupled to the above registration system;
[0233] A step of analyzing in situ process motion commands of the device controller from the transfer device using a motion resolver communicably coupled to the device controller, wherein in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and another predetermined motion set of the transfer device is limited using the mapped in situ process motions;
[0234] Predetermined operation data specifying the at least one dynamic performance variable output by the transfer device executing the other predetermined motion set is registered using the registration system, and other values (C) defined by the probability density function of each dynamic performance variable output by the transfer device executing the mapped original position process motions of the other predetermined motion set. pkOther A step of determining ) using the above processor; and
[0235] For each of the dynamic performance variables output by the transfer device, corresponding to the aforementioned predetermined motion base set and the aforementioned other predetermined motion set, the other value and the base value (C pkBase The method includes comparing the transfer device using the processor, wherein the transfer device is a unique transfer device common to both the predetermined motion base set and the other predetermined motion set, and evaluating the health of the transfer device based on the comparison.
[0236] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-place process motion substantially maps onto a corresponding template motion among the template motions.
[0237] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0238] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command defines a template motion in at least one degree of freedom of the transfer device motion.
[0239] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, in a register of the device controller, and the processor analyzes the mapped motions from periodic access to the motion histogram in the register.
[0240] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0241] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0242] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0243] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering trend data for each dynamic performance variable using the registration system, wherein the trend data defines the deterioration trend of each dynamic performance variable.
[0244] According to one or more aspects of the disclosed embodiments, the method further comprises the step of aggregating, using the processor, the dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device, and predicting the occurrence of the transfer device having performance below a predetermined performance state.
[0245] According to one or more aspects of the disclosed embodiments, the method further comprises the step of providing to an operator of the transfer device, using the processor, an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0246] According to one or more aspects of the disclosed embodiments, a method for health assessment of a system including a transfer device is:
[0247] A step of registering predetermined operation data that implements at least one dynamic performance variable output by a transfer device executing a predetermined set of motion bases arranged to limit the statistical definition of predetermined base motions, using a registration system communicably coupled to a device controller;
[0248] A step of determining, using a processor communicably coupled to the registration system, a normalized value that statistically defines the nominal performance of each of the dynamic performance variables output by the transfer device for each motion of the aforementioned predetermined set of motion bases;
[0249] A step of analyzing in situ process motion commands of the device controller from the transfer device using a motion resolver communicably coupled to the device controller, wherein in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and another predetermined motion set of the transfer device is limited using the mapped in situ process motions;
[0250] A step of registering, using the registration system, predetermined operation data that specifies the at least one dynamic performance variable output by the transfer device executing the other predetermined motion set, and determining, using the processor, another normalized value that statistically defines the in-place process performance of each dynamic performance variable output by the transfer device executing the mapped in-place process motion of the other predetermined motion set; and
[0251] A step of comparing the other normalized value and the normalized value for each of the dynamic performance variables of the transfer device, corresponding to the aforementioned predetermined base motion set and the aforementioned other predetermined motion set, using the processor, and determining the rate of performance deterioration from the nominal performance of the transfer device based on the comparison, wherein the device is unique, and each normalized value (C) for each predetermined base motion of the aforementioned predetermined motion base set pkbase ) and each different value (C) for each mapped in-place process motion of the other predetermined motion sets mentioned above. pkOther ) is uniquely correlated only with the above-mentioned unique device, and the above-mentioned predetermined performance deterioration rate is uniquely correlated only with the above-mentioned unique device, including the step.
[0252] According to one or more aspects of the disclosed embodiments, the method further comprises the step of providing the system with a plurality of different unique devices and the transfer device connected to each other, wherein each different unique device from the plurality of different unique devices (i) has different corresponding normalized values (C) for each base motion of the predetermined base motion set. pkBasei ) and other normalized values (C) for each mapped in-situ process motion of the other predetermined motion set that corresponds uniquely only to the different corresponding unique device (i) from a plurality of different unique devices. pkOtheri has ).
[0253] According to one or more aspects of the disclosed embodiments, the method comprises, for each different unique device (i), the corresponding normalized values (C pkBasei ) and other normalized values (C) uniquely correlated to the different corresponding unique device (i). pkOtheri A step of registering ) using a controller respectively coupled to the different corresponding unique devices, and for each different unique device (i), the uniquely correlated normalized values (C pkBasei ) and other normalized values of that different unique device (i) (C pkOtheri The method further includes the step of determining the corresponding performance deterioration rate for the different unique device (i) from the comparison of ) on the device based on the device (i=1...n).
[0254] According to one or more aspects of the disclosed embodiments, each different unique device from the plurality of different unique devices is in common with the transfer device.
[0255] According to one or more aspects of the disclosed embodiments, each different unique device from the plurality of different unique devices is configured differently from the transfer device.
[0256] According to one or more aspects of the disclosed embodiments, the method further includes the step of registering trend data defining the performance deterioration trend of each of the plurality of different unique devices of the system and the transfer device in the register of the controller.
[0257] According to one or more aspects of the disclosed embodiments, the method further comprises the step of determining system performance deterioration trends that define performance deterioration of the system by combining performance deterioration trends corresponding to the transfer device and each of the plurality of different unique devices of the system using the processor.
[0258] According to one or more aspects of the disclosed embodiments, the method further comprises the steps of: using the processor to compare the performance deterioration trend of the transfer device with the performance deterioration trend of each of the plurality of different unique devices; and using the processor to determine whether the performance deterioration trend of the transfer device or the performance deterioration trend of each of the plurality of different unique devices is a control performance deterioration trend and whether the control performance deterioration trend is a determining cause of the performance deterioration trend of the system.
[0259] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-situ process motion substantially maps onto a corresponding template motion among the template motions.
[0260] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0261] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command is defined by a template motion within at least one degree of freedom of motion of the transfer device.
[0262] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, in a register of the device controller, and the processor analyzes the mapped motions from periodic access of the motion histogram in the register.
[0263] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0264] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0265] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0266] According to one or more aspects of the disclosed embodiments, the method further comprises the step of registering trend data for each dynamic performance variable using the registration system, wherein the trend data defines the deterioration trend of each dynamic performance variable.
[0267] According to one or more aspects of the disclosed embodiments, the method further comprises the step of aggregating, using the processor, the dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device, and predicting the occurrence of the transfer device having performance below a predetermined performance state.
[0268] According to one or more aspects of the disclosed embodiments, the method further comprises the step of providing to an operator of the transfer device, using the processor, an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0269] According to one or more aspects of the disclosed embodiments, a health assessment device for evaluating the health of a system including a transfer device comprises:
[0270] A registration system communically coupled to the transfer device controller of the above-mentioned transfer device, wherein the registration system is,
[0271] It is configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing a predetermined motion base set of predetermined base motions, and
[0272] A registration system configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing another predetermined set of motions;
[0273] As a motion resolver communicably coupled to the above-mentioned device controller, the motion resolver is,
[0274] The above device controller is configured to analyze in situ process motion commands from the transfer device, and the in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and
[0275] A motion resolver configured to limit another predetermined set of motions of the above transfer device using the mapped original position process motions; and
[0276] As a processor communicably coupled to the above-mentioned registration system, the processor,
[0277] A base value (C) defined by the probability density function of each of the dynamic performance variables output by the transfer device for each motion of the above-determined set of motion bases. pkBase It is configured to determine ),
[0278] Another value (C) defined by the probability density function of each dynamic performance variable output by the transfer device executing the mapped in-place process motions of the other predetermined motion set. pkOther It is configured to determine ),
[0279] For each of the dynamic performance variables output by the transfer device, corresponding respectively to the aforementioned predetermined motion base set and the aforementioned other predetermined motion set, the other value and the base value (C pkBase It is configured to compare ), and
[0280] It includes a processor configured to evaluate the health of the transfer device based on the above comparison,
[0281] The above transfer device is a common transfer device for both the above-determined motion base set and the other above-determined motion set.
[0282] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-place process motion substantially maps onto a corresponding template motion among the template motions.
[0283] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0284] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command defines a template motion in at least one degree of freedom of the transfer device motion.
[0285] According to one or more aspects of the disclosed embodiments, the transfer device controller comprises a register configured to register a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, and the processor is further configured to analyze the mapped motions from periodic access to the motion histogram in the register.
[0286] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0287] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0288] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0289] According to one or more aspects of the disclosed embodiments, the registration system is further configured to register trending data for each dynamic performance variable, said trending data defines the deterioration trend of each dynamic performance variable.
[0290] According to one or more aspects of the disclosed embodiments, the processor is further configured to aggregate dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device using the processor, and to predict the occurrence of the transfer device having performance below a predetermined performance state.
[0291] According to one or more aspects of the disclosed embodiments, the transfer device is further configured to provide the operator of the transfer device with an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0292] According to one or more aspects of the disclosed embodiments, a health assessment device for evaluating the health of a system including a transfer device comprises:
[0293] A registration system communically coupled to the transfer device controller of the above-mentioned transfer device, wherein the registration system is,
[0294] It is configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing a predetermined motion base set arranged to limit the statistical definition of predetermined base motions, and
[0295] A registration system configured to register predetermined operation data that implements at least one dynamic performance variable output by the transfer device executing another predetermined set of motions;
[0296] As a motion resolver communicably coupled to the above-mentioned device controller, the motion resolver is,
[0297] The above device controller is configured to analyze in situ process motion commands from the transfer device, and the in situ process motions executed by the transfer device are mapped to the predetermined base motions of the predetermined motion base set, and
[0298] A motion resolver configured to limit another predetermined set of motions of the above transfer device using the mapped original position process motions; and
[0299] As a processor communicably coupled to the above-mentioned registration system, the processor,
[0300] It is configured to determine a normalized value that statistically defines the nominal performance of each of the dynamic performance variables output by the transfer device for each motion of the above-determined set of motion bases, and
[0301] It is configured to determine other normalized values that statistically define the in-situ process performance of each dynamic performance variable output by the transfer device executing the mapped in-situ process motion of the other predetermined motion set, and
[0302] It is configured to compare the other normalized value and the normalized value for each of the dynamic performance variables of the transfer device, corresponding to the above-determined base motion set and the other-determined motion set, respectively, and
[0303] It includes a processor configured to determine the performance deterioration rate of the transfer device from the nominal performance based on the above comparison,
[0304] The above transfer device is a common transfer device for both the above-determined base motion set and the other-determined motion set.
[0305] According to one or more aspects of the disclosed embodiments, each of the predetermined base motions defines a template motion, and each in-place process motion substantially maps onto a corresponding template motion among the template motions.
[0306] According to one or more aspects of the disclosed embodiments, each template motion is defined by at least one of a torque command and a position command from the device controller.
[0307] According to one or more aspects of the disclosed embodiments, at least one of the torque command and the position command defines a template motion in at least one degree of freedom of the transfer device motion.
[0308] According to one or more aspects of the disclosed embodiments, the transfer device controller comprises a register configured to register a histogram of motions commanded by the device controller, including in-place process motions executed by the transfer device, and the processor is further configured to analyze the mapped motions from periodic access to the motion histogram in the register.
[0309] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include at least one common base motion that statistically defines a number of base motion types.
[0310] According to one or more aspects of the disclosed embodiments, the predetermined base motions of the predetermined set of motion bases include a plurality of different base motion types, each of which is executed by the transfer device in a number of common motions statistically defined for each base motion type.
[0311] According to one or more aspects of the disclosed embodiments, each of the different base motion types has at least one different corresponding torque command feature and position command feature that defines each of the different common motions as each base motion type.
[0312] According to one or more aspects of the disclosed embodiments, the registration system is further configured to register trending data for each dynamic performance variable, said trending data defines the deterioration trend of each dynamic performance variable.
[0313] According to one or more aspects of the disclosed embodiments, the processor is further configured to aggregate dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the transfer device using the processor, and to predict the occurrence of the transfer device having performance below a predetermined performance state.
[0314] According to one or more aspects of the disclosed embodiments, the transfer device is further configured to provide the operator of the transfer device with an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
[0315] It should be understood that the foregoing description is merely illustrative of the aspects of the disclosed embodiments. Various alternatives and modifications may be devised by those skilled in the art without departing from the aspects of the disclosed embodiments. Accordingly, the aspects of the disclosed embodiments are intended to include all such alternatives, modifications, and variations that fall within the scope of the appended claims. Furthermore, the mere fact that different features are described in mutually different dependent or independent claims does not indicate that a combination of these features cannot be advantageously used, and such combinations remain within the scope of the aspects of the invention.
Claims
Claim 1 A method for evaluating the health of a system including a transfer device, the method comprises: registering, using a registration system communically coupled to a device controller, predetermined operational data that implements at least one dynamic performance variable output by at least one device component of the transfer device executing a predetermined base speed or pose command base set of predetermined base speed or pose commands of the transfer device; determining, using a processor communically coupled to the registration system, a base value defined by a probability density function of each of the dynamic performance variables output by the at least one device component for each speed or pose command of the predetermined speed or pose command base set; analyzing in situ process speeds or pose commands of the device controller from the at least one device component using a resolver communically coupled to the device controller, wherein in situ process speeds or poses executed by the at least one device component are mapped to the predetermined base speeds or pose commands of the predetermined speed or pose command base set, and another predetermined speed or pose of the transfer device corresponding to the transfer device of the predetermined speed or pose command base set. A step of limiting the command set using the mapped in-situ process speeds or poses;A method comprising the step of registering, using the registration system, predetermined operating data implementing the at least one dynamic performance variable output by the at least one device component executing the other predetermined speed or pose command set, and determining, using the processor, another value defined by the probability density function of each of the dynamic performance variables output by the at least one device component executing the mapped in-situ process speeds or poses of the other predetermined speed or pose command set—this is for comparing the other value with the base value—and evaluating the health of the transfer device based on the comparison. Claim 2 A method according to claim 1, further comprising the step of comparing the other value and the base value using the processor for each of the dynamic performance variables output by the at least one device component, each corresponding to the predetermined speed or pose command base set and the other predetermined speed or pose command set. Claim 3 A method according to claim 1, wherein the transfer device is a unique transfer device common to both the predetermined speed or pose command base set and the other predetermined speed or pose command set. Claim 4 A method according to claim 1, wherein each of the predetermined base commands limits the template command speed or pose, and each original position process maps to one of the corresponding template commands. Claim 5 The method of claim 1 further comprises the step of registering a histogram commanded by the device controller, which includes an in-situ process executed by at least one device component, in a register of the device controller, wherein the processor analyzes the mapped in-situ process from periodic access to a velocity or pose histogram in the register. Claim 6 A method according to claim 1, wherein the predetermined base command of the predetermined speed or pose command base set comprises at least one common base command of a statistically defined number that limits the base command type. Claim 7 A method according to claim 1, wherein the predetermined base command of the predetermined speed or pose command base set comprises a plurality of different base command types, each of the command types being executed by the at least one device component in a number of common commands statistically determined for each base command type. Claim 8 A method according to claim 7, wherein each of the different base component types comprises at least one different corresponding command feature and position command feature that limits each of the different common commands to each base command type. Claim 9 A method according to claim 1, further comprising the step of registering trend data for each dynamic performance variable using the registration system, wherein the trend data defines the deterioration trend of each dynamic performance variable. Claim 10 A method according to claim 9, further comprising the step of aggregating dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the at least one device component using the processor, and predicting the occurrence of the transfer device having performance under a predetermined performance state. Claim 11 A method according to claim 10, further comprising the step of providing to the operator of the transfer device, using the processor, an indication of the predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables. Claim 12 A health evaluation device for evaluating the health of a system including a transfer device, wherein the health evaluation device comprises: a registration system communicably coupled to a transfer device controller of the transfer device, wherein the registration system is configured to register predetermined operational data that implements at least one dynamic performance variable output by at least one device component of the transfer device executing a predetermined speed or pose command base set of predetermined base commands of the transfer device, and is configured to register predetermined operational data that implements at least one dynamic performance variable output by the at least one device component executing another predetermined speed or pose command set; and a resolver communicably coupled to the transfer device controller, wherein the resolver is configured to analyze in situ process commands of the transfer device controller from the at least one device component, and the in situ process executed by the at least one device component is mapped to the predetermined base commands of the predetermined speed or pose command base set, and another predetermined speed or pose command set of the transfer device corresponding to the transfer device of the predetermined speed or pose command base set is mapped using the mapped in situ process. Motion resolver configured to limit;A health evaluation device comprising a processor coupled to the above-mentioned registration system, wherein the processor is configured to determine a base value defined by a probability density function of each of the dynamic performance variables output by the at least one device component for each command of the above-mentioned predetermined speed or pose command base set, and determine another value defined by a probability density function of each of the dynamic performance variables output by the at least one device component executing the above-mentioned mapped in-situ process of the other predetermined speed or pose command set—this is for comparing the other value with the base value—and is configured to evaluate the health of the transfer device based on the comparison. Claim 13 A health assessment device according to claim 12, wherein the processor is further configured to compare the other value and the base value for each of the dynamic performance variables output by the at least one device component, each corresponding to the predetermined speed or pose command base set and the other predetermined speed or pose command set. Claim 14 In paragraph 12, the transfer device is a health assessment device that is a common transfer device for both the predetermined speed or pose command base set and the other predetermined speed or pose command set. Claim 15 A health assessment device according to claim 12, wherein each of the above-mentioned predetermined base commands limits a template command, and each in-situ process maps to one of the corresponding template commands. Claim 16 In claim 12, the transfer device controller comprises a register configured to register a histogram commanded by the transfer device controller, which includes an in-situ process executed by the at least one device component, and the processor further configured to analyze the mapped in-situ process from periodic access of the histogram within the register, a health assessment device. Claim 17 A health assessment device according to claim 12, wherein the predetermined base command of the predetermined speed or pose command base set comprises at least one common base command of a statistically defined number that limits the base command type. Claim 18 In paragraph 12, the predetermined base commands of the predetermined speed or pose command base set comprise a plurality of different base command types, each of which is executed by the at least one device component in a number of common commands statistically determined for each base command type, a health assessment device. Claim 19 A health assessment device according to claim 18, wherein each of the above-mentioned different base command types has at least one different corresponding command feature and position command feature that limits each of the different common commands to each base command type. Claim 20 In paragraph 12, the above-mentioned registration system is further configured to register trending data for each dynamic performance variable, said trending data defining the deterioration trend of each dynamic performance variable, a health assessment device. Claim 21 A health assessment device according to claim 20, wherein the processor aggregates dynamic performance variables having the highest deterioration trend among the at least one dynamic performance variable output by the at least one device component using the processor, and is further configured to predict the occurrence of the transfer device having performance under a predetermined performance state. Claim 22 A health assessment device according to claim 21, further configured to provide to the operator of the transfer device an indication of a predicted issuance of the transfer device having performance under a predetermined performance state based on the aggregation of the dynamic performance variables.
Citation Information
Patent Citations
Composite longitudinal vibration mechanical filter and its manufacture
JP1992240912A
Intelligent condition-monitoring and fault diagnostic system
KR101322434B1
Predictive maintenance for third party support equipment
KR1020130095817A
Condition Monitoring Of An Industrial Robot
US20140222352A1
Predictive maintenance for industrial products using big data
US20140336791A1
