Determining substrate position threshold based on optical characteristics

Abstract

The system includes an optical sensor including a light-emitting element and a light-receiving element. The optical sensor is configured to detect an optical transmission ratio of the substrate in response to a substrate support supporting the substrate within an optical path of a first optical sensor. The system further includes a processing device communicatively coupled to the optical sensor. The processing device is configured to determine an optical threshold based on the optical transmission ratio of the substrate. The processing device is further configured to determine a position of the substrate relative to a substrate handling robot end effector based on the sensor data output by the position sensor and the optical threshold.

Description

[Technical Field]

[0001] FIELD OF THE DISCLOSURE Embodiments of the present disclosure relate generally to determining the optical transmission ratio of a substrate, and more particularly to systems, methods, and devices for determining a threshold for substrate position probing based on optical characteristics of the substrate. [Background technology]

[0002] In substrate processing, accurately determining the position of a substrate as it is being handled is of utmost importance. In particular, accurately determining the position can directly affect the quality of the processed substrate. The position of a substrate can be affected by a variety of factors, including acceleration during handling. Summary of the Invention

[0003] The following is a simplified summary of the present disclosure to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is not intended to identify key or critical elements of the disclosure, nor to delineate the scope or claims of particular embodiments of the disclosure. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.

[0004] Some embodiments described herein involve a system including a first optical sensor including a light-emitting element and a light-receiving element. The first optical sensor is configured to detect an optical transmittance ratio of the substrate in response to a substrate support supporting the substrate within an optical path of the first optical sensor. The system further includes a processing device communicatively coupled to the first optical sensor. The processing device is configured to determine an optical threshold based on the optical transmittance ratio of the substrate. The processing device is further configured to determine a position of the substrate relative to a substrate handling robot end effector based on the sensor data output by the position sensor and the optical threshold.

[0005] Additional or related embodiments described herein include a method including receiving optical transmission data from a first optical sensor. The optical transmission data indicates an optical transmission ratio of a substrate positioned within an optical path of the first optical sensor. The method further includes determining an optical threshold based on the optical transmission data. The method further includes receiving position sensor data indicating a position of the substrate along a substrate transport path. The method further includes determining a position of the substrate relative to a substrate handling robot end effector based on the position sensor data and the optical threshold.

[0006] In a further embodiment, a non-transitory machine-readable storage medium includes instructions that, when executed by a processing device, cause the processing device to perform operations, including receiving data including one or more of optical transmission data associated with a substrate, first reflectance data associated with a first surface of the substrate, or second reflectance data associated with a second surface of the substrate. The operations further include inputting the one or more of the optical transmission data, the first reflectance data, or the second reflectance data into a trained machine learning model. The operations further include receiving output from the trained machine learning model indicating predicted substrate identification data. The predicted substrate identification data corresponds to one or more of a predicted corresponding substrate type, a predicted first characteristic of the substrate, or a predicted second characteristic of the substrate.

[0007] In accordance with these and other aspects of the present disclosure, numerous other features are also provided. Other features and aspects of the present disclosure will become more fully apparent from the following detailed description, claims, and accompanying drawings.

[0008] The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings, in which like references indicate similar elements. It should be noted that various references in this disclosure to "an embodiment" or "one embodiment" do not necessarily refer to the same embodiment, but rather that such references mean at least one. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a schematic top view of an exemplary manufacturing system according to aspects of the present disclosure. [Figures 2A-2C] FIG. 1 is a simplified side view of a system for determining optical properties of a substrate, according to aspects of the present disclosure. [Figure 3] FIG. 1 illustrates an exemplary computer system architecture according to aspects of the present disclosure. [Figure 4] 1 is a flowchart of a method for updating an optical threshold corresponding to a substrate ID, according to an aspect of the present disclosure. [Figure 5] FIG. 1 illustrates a model training and application workflow for determining predicted substrate identification, according to aspects of the present disclosure. [Figure 6A] 1 is a flowchart of a method for generating a training dataset for training a machine learning model, according to an aspect of the present disclosure. [Figure 6B] 1 is a flowchart of a method for generating predicted substrate identification data using a trained machine learning model, according to an aspect of the present disclosure. [Figure 7] 1 is a flowchart of a method for determining a substrate position based on a determined optical threshold of the substrate, according to an aspect of the present disclosure. [Figure 8] 1 is a diagrammatic representation of a machine, an exemplary form of computing device, capable of executing a set of instructions to cause the machine to perform any one or more of the techniques discussed herein. DETAILED DESCRIPTION OF THE INVENTION

[0010] Embodiments of the present disclosure are directed to systems and methods for determining a substrate position threshold based on optical characteristics. Substrates are often handled by substrate handling robots within a manufacturing system. The substrate handling robot can lift a substrate (e.g., via the robot's end effector) and transport the substrate from one station to another within the manufacturing system. The robot then places the transported substrate within a new station (e.g., a process chamber, etc.). During transport, the substrate experiences accelerations that can cause the substrate to slide on the robot end effector. Changes in position on the end effector can cause the substrate to be mispositioned. Misplaced substrates can lead to incorrect processing of the substrate, which can result in damage to the substrate. The damaged substrate may be discarded after processing.

[0011] Conventional systems utilize position sensors to determine the position of a substrate relative to a robot end effector. In these conventional systems, a transported substrate interrupts a light beam emitted by an emitter of the position sensor. The time the beam is interrupted is captured. Using the robot's position (e.g., tracked by software) in combination with the time the light beam is interrupted, the software can determine the position of the substrate relative to the robot end effector. However, some substrates transmit a certain amount of light and do not completely interrupt the light beam. In some cases, some substrates (e.g., glass substrates) transmit a large portion of the light beam between the transmitter and the receiver. The intensity of the light beam received by the receiver of the position sensor can be reduced, but the beam cannot be completely interrupted. Therefore, the position sensor of the conventional system can accurately detect the position of the substrate. Because the light beam of the position sensor can penetrate at least some substrates, the conventional system cannot accurately determine the position of the substrate relative to the robot end effector. The position of the substrate can be more accurately determined by setting a light intensity threshold (e.g., an optical threshold) that indicates the presence of a substrate relative to the substrate position sensor to trigger a state change in the sensor.

[0012] Furthermore, a substrate manufacturing system can process multiple types of substrates. Some types of substrates can include glass substrates, silicon substrates, and bonded substrates (e.g., silicon bonded to glass). Some types of substrates can have different edge shapes and / or varying degrees of polishing. Furthermore, some substrates can have different coatings compared to other substrates and / or different coatings on different surfaces of the substrate (e.g., a first coating on the top surface of the substrate and a different second coating on the bottom surface of the substrate). Each substrate type can utilize different settings for determining substrate position. For example, a glass substrate can transmit more light than a silicon substrate, requiring adjustment of the light intensity threshold (e.g., optical threshold) corresponding to the position sensor. Without threshold adjustment, the position sensor cannot detect a change in light intensity sufficient to trigger a state change in the sensor (e.g., a state change indicating the presence of a substrate in the position sensor). The adjustment can be performed manually by adjusting the threshold corresponding to the position sensor. This adjustment can be slow or even impractical, especially when multiple types of substrates are routinely processed by the manufacturing system. Detection of substrates of this type currently being handled and / or processed can enable automatic updating of the optical threshold as discussed herein above.

[0013] Aspects and embodiments of the present disclosure address the above and other shortcomings of conventional systems by providing a system (e.g., an optical property measurement system) for measuring optical properties of a substrate. In some embodiments, the system includes a first optical sensor having a light-emitting element and a light-receiving element. The light-emitting element can emit a light beam (e.g., a radiation beam, a light ray, etc.) along an optical path of the first optical sensor toward the light-receiving element. In some embodiments, a substrate support holds the substrate within the optical path. For example, a substrate support (e.g., a robot end effector, a substrate aligner, etc.) can hold the substrate such that the optical path intersects the substrate. Thus, the light beam emitted by the light-emitting element can travel through the substrate to the light-receiving element (e.g., the substrate transmits the light beam). Due to the optical properties of the substrate (e.g., transmission ratio, reflectivity, etc.), the light-emitting element can receive the light beam with a reduced intensity.

[0014] In some embodiments, the system includes a processing device communicatively coupled to a first optical sensor. The processing device can receive optical transmission data from the first optical sensor. The processing device can determine an optical transmission ratio (e.g., a ratio of transmitted light to received light) corresponding to a substrate on the substrate support. In some embodiments, the processing device determines an optical threshold based on the optical transmission ratio of the substrate. The optical threshold can correspond to a threshold condition for determining the position of the substrate (e.g., by a position sensor). For example, a threshold amount of light (e.g., corresponding to the optical threshold) detected by a light-receiving element of the position sensor can indicate the presence of a corresponding substrate. In another example, the timing at which the light-receiving element of the position sensor detects the threshold amount of light (e.g., corresponding to the optical threshold) can correspond to the timing at which the substrate "breaks the beam" of the position sensor as the substrate is moved past the position sensor by the substrate handling robot. In some embodiments, the processing device can track the movement of the substrate handling robot as the robot moves the substrate based on one or more robot position sensors (e.g., on the robot). Based on the optical threshold and sensor data output by the position sensor, the processing device can determine the position of the substrate relative to a substrate handling robot end effector.

[0015] In some embodiments, the system may include a second optical sensor and / or a third optical sensor. Each of the second optical sensor and / or the third optical sensor may be configured to detect the reflectivity of a surface of the substrate (e.g., the top and / or bottom surfaces of the substrate). Data from the second and / or third optical sensors may be used by a processing device to determine one or more characteristics of the substrate. In some examples, data collected by the second optical sensor may indicate the reflectivity of the top surface of the substrate. The reflectivity of the top surface may further indicate a coating (e.g., a coating or film) on the top surface of the substrate. In some examples, data collected by the third optical sensor may indicate the reflectivity of the bottom surface of the substrate. The reflectivity of the bottom surface may further indicate a coating (e.g., a coating or film) on the bottom surface of the substrate. By identifying one or more coatings (e.g., or films) on the substrate, the substrate may be successfully identified, and settings for determining substrate position (e.g., light intensity thresholds or optical thresholds described herein, etc.) may be automatically updated.

[0016] Embodiments of the present disclosure provide advantages over the conventional systems described above. In particular, some embodiments described herein can detect the optical transmission ratio of a substrate being handled within a substrate manufacturing system. The optical transmission ratio can be used to accurately determine the position of a substrate being handled by a substrate handling robot relative to a robot end effector. Accurately determining the position of a substrate can result in more accurate substrate handling and more precise substrate processing compared to conventional systems. Additionally, the systems and methods described herein can predict substrate characteristics (e.g., via machine learning algorithms), thereby enabling automatic updating of parameters related to determining the position of a substrate. By automatically identifying the type of substrate being handled, the systems and methods of the present disclosure can provide faster substrate handling and increased manufacturing system throughput.

[0017] 1 is a schematic top view of an exemplary manufacturing system 100 according to an embodiment of the present disclosure. The manufacturing system 100 is capable of performing one or more processes on a substrate 102. The substrate 102 may be any suitably rigid, flat body of fixed dimensions, such as, for example, a silicon-containing disk or wafer, a patterned wafer, a glass plate, or the like, suitable for fabricating electronic devices or circuit components thereon.

[0018] The manufacturing system 100 may include a process tool 104 and a factory interface 106 coupled to the process tool 104. The process tool 104 may include a housing 108 having a transfer chamber 110. The transfer chamber 110 may include one or more process chambers (also referred to as processing chambers) 114, 116, 118 arranged around and coupled to the transfer chamber 110. The process chambers 114, 116, 118 may be coupled to the transfer chamber 110 via respective ports, such as slit valves.

[0019] The process chambers 114, 116, and 118 can be adapted to perform any number of processes on the substrate 102. The same or different substrate processes can be performed in each process chamber 114, 116, and 118. The substrate processes can include atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, hardening, pre-cleaning, metal or metal oxide removal, and the like. In one example, a PVD process can be performed in one or both process chambers 114, an etching process can be performed in one or both process chambers 116, and an annealing process can be performed in one or both process chambers 118. Other processes can also be performed on the substrate 102 in the process chambers. The process chambers 114, 116, and 118 can each include a substrate support assembly. The substrate support assembly can be configured to hold the substrate 102 in a fixed position while a substrate process is being performed.

[0020] In some embodiments, the process chambers 114, 116, 118 may include a carousel (also referred to as a susceptor). The carousel may be disposed within the interior volume of the process chambers 114, 116, 118 and may be configured to rotate about an axial center of the process chambers 114, 116, 118 during a process (e.g., a deposition process) to ensure uniform distribution of process gases. In some embodiments, the carousel may include one or more end effectors configured to handle one or more objects. For example, the end effectors may be configured to hold substrates, process kits, and / or process kit carriers. According to embodiments described herein, one or more sensors may be disposed in the process chambers 114, 116, 118, and the one or more sensors may be configured to detect the placement of an object on the carousel end effector.

[0021] The transfer chamber 110 can also include a transfer chamber robot 112. The transfer chamber robot 112 can include one or more arms, each arm including one or more end effectors at the end of each arm. The end effectors can be configured to handle a particular object, such as a substrate. Alternatively or additionally, the end effectors can be configured to handle a process kit (i.e., use a process kit carrier). In some embodiments, the transfer chamber robot 112 can be a Selective Compliance Assembly Robot Arm (SCARA) robot, such as a two-link SCARA robot, a three-link SCARA robot, a four-link SCARA robot, etc.

[0022] A load lock 120 may also be coupled to the housing 108 and the transfer chamber 110. The load lock 120 may be configured to be coupled to the transfer chamber 110 on one side and to the factory interface 106. In some embodiments, the load lock 120 may have an environmentally controlled atmosphere that allows the load lock 120 to vary from a vacuum environment (where substrates may be transferred to and from the transfer chamber 110) to an atmospheric pressure inert gas environment (where substrates may be transferred to and from the factory interface 106) or a near-atmospheric pressure inert gas environment. In some embodiments, the load lock 120 may be a stacked load lock and may have a pair of upper and a pair of lower internal chambers located at different vertical levels (e.g., one above the other). In some embodiments, the pair of upper internal chambers may be configured to receive processed substrates from the transfer chamber 110 for removal from the process tool 104, and the pair of lower internal chambers may be configured to receive substrates from the factory interface 106 for processing within the process tool 104. In some embodiments, the load lock 120 may be configured to perform a substrate process (eg, an etch or a pre-clean) on one or more substrates 102 received within the load lock 120 .

[0023] The factory interface 106 may be any suitable enclosure, such as, for example, a front-end equipment module (EFEM). The factory interface 106 may be configured to receive substrates 102 from substrate carriers 122 (e.g., front-opening integrated pods (FOUPs)) docked to various load ports 124 of the factory interface 106. A factory interface robot 126 (shown in dashed lines) may be configured to transfer substrates 102 between the substrate carriers 122 (also referred to as containers) and the load locks 120. In other and / or similar embodiments, the factory interface 106 may be configured to receive replacement parts (e.g., process kits) from a replacement parts storage container 123. The factory interface robot 126 may include one or more robot arms and may be or include a SCARA robot. In some embodiments, the factory interface robot 126 may have more links and / or more degrees of freedom than the transfer chamber robot 112. The factory interface robot 126 may include an end effector at the end of each robot arm. The end effector can be configured to lift and handle a specific object, such as a substrate or a process kit. Alternatively or additionally, the end effector can be configured to handle an object, such as a process kit (e.g., using a process kit carrier). In some embodiments, the factory interface robot 126 can place the substrate on the aligner 140 before and / or after transporting the substrate to the load lock 120. The aligner 140 can be configured to align the substrate to a target alignment (e.g., a target orientation) for transport and / or processing.

[0024] Any conventional robot type can be used for the factory interface robot 126. Transfers can be performed in any order or direction. In some embodiments, the factory interface 106 can be maintained, for example, in a non-reactive gas environment at a slight positive pressure (e.g., using nitrogen as the non-reactive gas).

[0025] In some embodiments, the transfer chamber 110, the process chambers 114, 116, and 118, and the load lock 120 can be maintained at a vacuum level. The manufacturing system 100 can include one or more vacuum ports coupled to one or more stations of the manufacturing system 100. For example, a first vacuum port 130a can couple the factory interface 106 to the load lock 120. A second vacuum port 130b can be coupled to the load lock 120 and can be disposed between the load lock 120 and the transfer chamber 110.

[0026] In some embodiments, one or more sensors may be included at one or more stations of the manufacturing system 100. For example, one or more sensors may be included in the transfer chamber 110 at or near the ports (i.e., entrances) of the process chambers 114, 116, 118. An end effector of a robot arm (e.g., of the transfer chamber robot 112) may move the substrate 102 or process kit (i.e., use a process kit carrier) past one or more sensors as it moves the substrate 102 and / or process kit into or out of the process chambers 114, 116, 118. Each sensor may be configured to detect the substrate 102 or process kit and / or carrier as the end effector moves the substrate 102 or process kit and / or carrier into or out of the process chambers 114, 116, 118. In some embodiments, one or more sensors included in the manufacturing system 100 may be configured to sense one or more optical properties of the substrate. For example, one or more optical sensors may be included in the factory interface 106 to determine optical transmittance data and / or reflectance data for the substrate.

[0027] The manufacturing system 100 can also include a system controller 128. The system controller 128 can be and / or include a computing device such as a personal computer, a server computer, a programmable logic controller (PLC), a microcontroller, etc. The system controller 128 can include one or more processing devices, which can be general-purpose processing devices such as a microprocessor, a central processing unit, etc. More specifically, the processing device can be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or a processor implementing other instruction sets or combinations of instruction sets. The processing device can also be one or more special-purpose processing devices such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, etc. The system controller 128 can include a data storage device (e.g., one or more disk drives and / or solid-state drives), a main memory, a static memory, a network interface, and / or other components. The system controller 128 may execute instructions to implement any one or more of the techniques and / or embodiments described herein. These instructions may be stored on a computer-readable storage medium, which may include a main memory, a static memory, a secondary storage, and / or a processing device during execution of the instructions.

[0028] In some embodiments, the system controller 128 can receive optical transmission data corresponding to the substrate 102 from the optical sensor. The optical transmission data can indicate the optical transmission ratio of the substrate. In some embodiments, the optical transmission data is collected as a substrate support (e.g., an end effector of the factory interface robot 126, an aligner 140, etc.) moves the substrate 102 through the optical path of the optical sensor. The system controller 128 can determine a threshold for transmitted light (e.g., transmitted through the substrate 102) based on the optical transmission data. In some embodiments, the system controller 128 can use this threshold (e.g., an optical threshold) to determine the presence of a substrate at a substrate position sensor station, as described herein.

[0029] In some embodiments, the system controller 128 can execute instructions to cause an end effector of a robot arm (e.g., of the transfer chamber robot 112) to move an object (i.e., a substrate 102 and / or a process kit) from a first station (e.g., the load lock 120) of the manufacturing system 100 to a second station (e.g., the process chambers 114, 116, 118) of the manufacturing system 100. In response to executing these instructions to move the object from the first station to the second station, the system controller 128 can receive a set of signals from one or more sensors (e.g., signals from one or more substrate position sensors) included at an entrance to the first station or the second station. Each signal can indicate a position of the object placed at the end effector of the robot arm. In some embodiments, the system controller 128 can determine the position of the object placed at the end effector of the robot arm based on each signal received from the one or more sensors. For example, the system controller 128 can identify a first set of coordinates corresponding to the center of the object based on the received signals. The system controller 128 can compare a first set of coordinates corresponding to the center of the object with a second set of coordinates corresponding to the center of the end effector to determine a correspondence between the first set of coordinates and the second set of coordinates. According to embodiments described herein, the system controller 128 can determine whether the center of the object is located at a target position relative to the center of the end effector based on the determined coordinate correspondence. In some embodiments, the system controller 128 adjusts the threshold for indicating the presence of an object in the position sensors based on the optical threshold determined above. For example, the system controller 128 can detect a change in the light intensity of one or more position sensors corresponding to the optical threshold to determine that an object (e.g., substrate 102) being transported (e.g., on the end effector) has passed the position sensor.

[0030] While embodiments of the present disclosure are directed to identifying the center of an object placed on an end effector of a robot arm (e.g., of a transfer chamber robot or a factory interface robot), it should be noted that embodiments of the present disclosure may also be applied to identifying the center of an object placed on other components of a manufacturing system. For example, according to embodiments described herein, an object may be placed on an end effector of a rotating carousel or rotating susceptor in process chambers 114, 116, and 118. According to the aforementioned embodiments, one or more sensors may be positioned within the process chamber to detect the position of the object on the rotating carousel. According to embodiments described herein, the one or more sensors may transmit signals to system controller 128, as described above, and system controller 128 may detect the center of the object on the rotating susceptor based on the received signals.

[0031] 2A-2C show simplified side views of systems for determining optical properties of a substrate according to aspects of the present disclosure. FIG. 2A shows a simplified side view of system 200A according to aspects of the present disclosure. FIG. 2B shows a simplified side view of system 200B according to aspects of the present disclosure. FIG. 2C shows a simplified side view of system 200C according to aspects of the present disclosure. In some embodiments, each of systems 200A, 200B, and 200C is an optical property measurement tool.

[0032] 2A , a substrate 202 is positioned on a substrate support, such as a robot end effector 226. The robot end effector can be a factory interface robot or a transfer chamber robot (e.g., factory interface robot 126 or transfer chamber robot 112 of FIG. 1 ). In some embodiments, the end effector 226 can transport the substrate 202 from a first station (e.g., substrate carrier 122 of FIG. 1 ) to a second station (e.g., load lock 120 of FIG. 1 ). In some embodiments, the end effector 226 can transport the substrate 202 from the factory interface chamber 206 through a vacuum port 230 (e.g., vacuum port 130 a or 130 b of FIG. 1 , a slit valve, etc.) to the load lock 220. One or more sensors positioned proximate the vacuum port 230 can detect optical properties of the substrate 202 as it is transported to the vacuum port 230. In some embodiments, as described herein, one or more sensors may detect the optical transmission ratio of substrate 202 and / or the reflectivity of one or more surfaces of substrate 202.

[0033] In some embodiments, system 200A includes a first optical sensor composed of a light emitting element and a light receiving element. In some embodiments, the light emitting element is light transmitter 230A, and the light receiving element is light receiver 230B. Light transmitter 230A can be configured to emit a light beam (e.g., a light ray, a laser beam, a beam of optical radiation, etc.) substantially perpendicular to the top surface of substrate 202. In some examples, light transmitter 230A is a light emitting diode (LED) or a laser diode. In some embodiments, an optical fiber (e.g., fiber optic) delivers the light beam from a remote source to light transmitter 230A. Light receiver 230B can be configured to receive the light beam (e.g., from light transmitter 230A). The light beam received by light receiver 230B can travel along an optical path from light transmitter 230A to light receiver 230B. In some examples, light receiver 230B is a photodiode, a line sensor, or a position sensing device (PSD). In some embodiments, the optical fiber receives the light beam from the optical receiver 230B and delivers the light beam to a remote sensor.

[0034] In some embodiments, the light beam emitted by the light transmitter 230A is transmitted through the substrate 202 (e.g., as the substrate 202 is transported by the end effector 226 through the optical path between the light transmitter 230A and the light receiver 230B). The intensity of the light beam received by the light receiver 230B can indicate the optical transmittance of the substrate 202. The optical transmittance ratio of the substrate 202 can be calculated by the controller 228 (e.g., the system controller 128 of FIG. 1 ) based on the ratio of the intensity of the light beam received (e.g., by the light receiver 230B) to the intensity of the light beam transmitted (e.g., by the light transmitter 230A). In some embodiments, either one of the light transmitter 230A or the light receiver 230B can include a lens for focusing the light beam. In some embodiments, the first optical sensor (e.g., the light transmitter 230A and the light receiver 230B) is communicatively coupled to the controller 228. The sensor data from the first light sensor may be received by the controller 228.

[0035] To accurately determine the position of the substrate 202 by one or more position sensors of the substrate processing system, the optical transmission ratio of the substrate 202 can be used to determine an optical threshold corresponding to a threshold amount of light transmitted. For example, a substrate can have an optical transmission ratio of 0.5, which corresponds to 50% (e.g., one-half) of the emitted light intensity being transmitted through the substrate. Thus, as the substrate is transported through the beam of the substrate position sensor, the position sensor detects a 50% reduction in light at the position sensor's receiver. The corresponding optical threshold for detecting the substrate may be slightly above 50% (e.g., about 55%, 60%, etc.) so that the position sensor can successfully and accurately detect the presence of the substrate and therefore accurately determine the position of the substrate. In this example, a reduction in received light below about 55% (or about 60%) indicates that the edge of the substrate has passed the position sensor. In another example, a substrate can have an optical transmission ratio of 0.95, which corresponds to 95% of the emitted light intensity being transmitted through the substrate. Thus, as the substrate is transported through the beam of the substrate position sensor, the position sensor detects a 95% reduction in light at the position sensor's receiver. Using the optical threshold from the previous example (e.g., about 55% or about 60%), the sensor would not be able to detect the presence of the substrate. Instead, using an optical threshold of about 97% (e.g., corresponding to a reduction in light below 97%), the position sensor can detect the presence of the substrate. In this example, a reduction in received light below about 97% indicates that the edge of the substrate has passed the position sensor.

[0036] In some embodiments, as described herein, the optical transmission ratio of the substrate 202 may be used in conjunction with other optical properties (e.g., reflectance, etc.) to determine the identity of the substrate 202 (e.g., the identity of the substrate type, the identity of one or more coatings on the substrate, etc.) In some embodiments, the controller may determine whether the substrate 202 is a glass substrate, a silicon substrate, a bonded substrate, etc. based on the optical properties of the substrate 202.

[0037] In some embodiments, system 200A includes a second optical sensor comprised of optical transmitter 232A and optical receiver 232B. In some embodiments, the second optical sensor may include lens 233. Optical transmitter 232A and optical receiver 232B may be substantially similar to optical transmitter 230A and optical receiver 230B (e.g., have similar structure, components, characteristics, etc.). In some embodiments, optical transmitter 232A is configured to emit a light beam at an oblique angle relative to the top surface of substrate 202. The light beam may be focused onto the top surface of substrate 202 by lens 233. A portion of the light beam may reflect from the top surface of substrate 202 and be received by optical receiver 232B. In some embodiments, the intensity of the portion of the light beam received by optical receiver 232B may be used to determine the reflectivity of the top surface of substrate 202. In some examples, the ratio of the intensity of the light beam received by the light receiver 232B to the intensity of the light beam emitted by the light transmitter 232A indicates the reflectivity of the top surface of the substrate 202. The reflectivity of the top surface of the substrate 202 can be used to identify the substrate 202 (e.g., the type of substrate, the coating on the top surface of the substrate, etc.). In some embodiments, the identification of the substrate 202 is performed using a machine learning algorithm described later herein. In some embodiments, the second light sensor (e.g., the light transmitter 232A and the light receiver 232B) is communicatively coupled to the controller 228. Sensor data from the second light sensor can be received by the controller 228.

[0038] In some embodiments, system 200A includes a third optical sensor comprised of optical transmitter 234A and optical receiver 234B. In some embodiments, the third optical sensor may include lens 233. Optical transmitter 234A and optical receiver 234B may be substantially similar to optical transmitter 232A and optical receiver 232B. In some embodiments, optical transmitter 234A is configured to emit a light beam at an oblique angle relative to the bottom surface of substrate 202. The light beam may be focused onto the bottom surface of substrate 202 by lens 235. A portion of the light beam may reflect from the bottom surface of substrate 202 and be received by optical receiver 234B. In some embodiments, the intensity of the portion of the light beam received by optical receiver 234B may be used to determine the reflectivity of the bottom surface of substrate 202. In some examples, the ratio of the intensity of the light beam received by the light receiver 234B to the intensity of the light beam emitted by the light transmitter 234A indicates the reflectivity of the bottom surface of the substrate 202. The reflectivity of the bottom surface of the substrate 202 can be used to identify the substrate 202 (e.g., the type of substrate, the coating on the bottom surface of the substrate, etc.). In some embodiments, the identification of the substrate 202 is performed using a machine learning algorithm described later herein. In some embodiments, the optical threshold can be automatically updated (e.g., by the controller 228) based on the identification of the substrate 202. In some embodiments, a third light sensor (e.g., the light transmitter 234A and the light receiver 234B) is communicatively coupled to the controller 228. Sensor data from the third light sensor can be received by the controller 228.

[0039] In some embodiments, controller 228 can determine the coating on the top surface of substrate 202 and / or the coating on the bottom surface of substrate 202 based on the reflectance data collected via optical receiver 232B and / or optical receiver 234B. In some embodiments, controller 228 makes this determination via a machine learning algorithm, as described later herein. In some embodiments, controller 228 can determine a corrective action based on the optical transmittance sensor data and / or the reflectance sensor data. For example, controller 228 can determine that the coating on the top surface of substrate 202 does not match the expected top surface coating. Additionally, controller 228 can determine that the coating on the bottom surface of substrate 202 matches the expected top surface coating. In response to determining the top and / or bottom substrate coatings, the controller can determine that substrate 202 should be flipped.

[0040] In some embodiments, the first optical sensor (optical transmitter 230A and optical receiver 230B), the second optical sensor (optical transmitter 232A and optical receiver 232B), and / or the third optical sensor (optical transmitter 234A and optical receiver 234B) can be supported within factory interface chamber 206 by brackets (not shown). In some embodiments, the brackets can be coupled to an interior wall of factory interface chamber 206.

[0041] 2B , system 200B includes the optical sensor of system 200A. In some embodiments, substrate 202 is supported by a pedestal of aligner 240 in factory interface chamber 206. The optical sensor can be supported (e.g., by a bracket) proximate to aligner 240. In some examples, a factory interface robot retrieves substrate 202 (e.g., from a load lock or substrate carrier) and places substrate 202 on aligner 240. While substrate 202 is being aligned to a target orientation by aligner 240, the optical sensor can determine one or more optical characteristics of substrate 202 (e.g., optical transmission ratio, reflectance, etc.). After substrate 202 is aligned and the optical characteristics are determined, the factory interface robot can remove substrate 202 from the aligner.

[0042] 2C , system 200C includes an optical sensor of system 200A. In some embodiments, the optical sensor is supported by a sensor support structure 250 coupled to the robot end effector 226. The optical sensor can measure optical properties of the substrate 202 while the substrate 202 is supported by the end effector 226. In some embodiments, the sensor support structure 250 is a generally Y-shaped bracket coupled at a distal end to the end effector 226. The sensor support structure 250 can be configured to move between an extended position and a retracted position in response to handling of the substrate 202 by the end effector 226. For example, when the end effector 226 lifts the substrate 202 from the first station, the sensor support structure 250 can be in a retracted position (not shown) to provide clearance for the substrate 202. After lifting the substrate, the sensor support structure 250 can move to an extended position (shown in FIG. 2C ) so that the substrate 202 intersects the optical path of a first optical sensor (e.g., comprised of optical transmitter 230A and optical receiver 230B). The optical sensor can then measure the optical properties of the substrate 202. In preparation for placing the substrate at a second (e.g., new) station, the sensor support structure 250 can move to a retracted position to provide clearance for the substrate 202 to be lowered from the end effector 226. In some embodiments, the sensor support structure 250 extends and retracts linearly (e.g., along a substantially linear path). In some embodiments, the sensor support structure 250 extends and retracts by pivoting about an axis (e.g., a vertical axis).

[0043] 3 illustrates an exemplary computer system architecture 300 according to aspects of the present disclosure. The computer system architecture 300 includes a client device 320, manufacturing equipment 322, an optical property measurement tool 326, a prediction server 312 (e.g., for generating prediction data, providing model adaptation, using a knowledge base, etc.), and a data store 350. The prediction server 312 can be part of a prediction system 310. The prediction system 310 can further include server machines 370 and 380. In some embodiments, the computer system architecture 300 can include or be part of a manufacturing system for processing substrates or the optical property measurement tool 326. Further details regarding the optical property measurement tool 326 are provided with respect to FIGS. 2A-2C .

[0044] Components of client device 320, manufacturing equipment 322, optical characteristic measurement tool 326, prediction system 310, and / or data store 350 can be coupled to each other via network 340. In some embodiments, network 340 is a public network that provides client device 320 access to prediction server 312, data store 350, and other publicly available computing devices. In some embodiments, network 340 is a private network that provides client device 320 access to manufacturing equipment 322, optical characteristic measurement tool 326, data store 350, and / or other privately available computing devices. Network 340 may include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.

[0045] The client devices 320 may include computing devices such as personal computers (PCs), laptops, mobile phones, smartphones, tablet computers, netbook computers, network-connected televisions ("smart TVs"), network-connected media players (e.g., Blu-ray players), set-top boxes, over-the-top (OTT) streaming devices, operator boxes, etc.

[0046] The manufacturing equipment 322 can fabricate products according to a recipe. In some embodiments, the manufacturing equipment 322 can include or be part of a manufacturing system that includes one or more stations (e.g., process chambers, transfer chambers, load locks, factory interfaces, etc.) configured to perform various operations on substrates.

[0047] The optical property measurement tool 326 may be a tool (e.g., a system) for determining one or more optical properties of the object being measured. The optical property measurement tool 326 may be configured to generate data associated with the optical transmittance and / or reflectance of the object being measured by the optical measurement tool 326. In some embodiments, the optical property measurement tool 326 corresponds to system 200A, system 200B, or system 200C. In some embodiments, such data (e.g., optical transmission data, first reflection data, second reflection data, etc.) may be stored in a data store 350, where the data may be accessed (e.g., via network 340). The optical property measurement tool 326 may include one or more sensors (e.g., multiple optical sensors) configured to detect optical properties and generate data associated with the object being measured. In some embodiments, the optical property measurement tool 326 includes a first optical sensor configured to detect the optical transmittance of light through the substrate. In some embodiments, the optical property measurement tool 326 includes a second optical sensor and / or a third optical sensor configured to detect a first reflectance of the first surface of the substrate and / or a second reflectance of the second surface of the substrate. In some embodiments, the optical property measurement tool 326 can generate optical transmission data and / or reflectance data based on the intensity of light received (e.g., by each light-receiving element of the corresponding optical sensor). In some embodiments, the optical measurement tool 326 can be included in a system used to manufacture components (e.g., processing chamber parts) of the manufacturing equipment 322. In some embodiments, the optical properties (e.g., optical transmittance, reflectance, etc.) measured by the optical property measurement tool 326 can be used to identify substrates being handled and / or processed by the manufacturing equipment 322. In some embodiments, the optical properties can be used to accurately determine the location of a substrate undergoing transport (e.g., within the manufacturing equipment 322).

[0048] The data store 350 can be a memory (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or another type of component or device capable of storing data. The data store 350 can include multiple storage components (e.g., multiple drives or multiple databases) that can span multiple computing devices (e.g., multiple server computers). The data store 350 can store optical transmission data and reflectance data (e.g., generated by the optical property measurement tool 326).

[0049] One or more portions of data store 350 can be configured to store data that is not accessible to users of the manufacturing system. In some embodiments, all data stored in data store 350 can be inaccessible to manufacturing system users. In other or similar embodiments, a portion of the data stored in data store 350 is inaccessible to users, while another portion of the data stored in data store 350 is accessible to users. In some embodiments, the inaccessible data stored in data store 350 is encrypted using an encryption mechanism unknown to the users (e.g., the data is encrypted using a private encryption key). In other or similar embodiments, data store 350 can include multiple data stores, with data that is inaccessible to users stored in a first data store and data that is accessible to users stored in a second data store.

[0050] In some embodiments, prediction system 310 includes server machine 370 and server machine 380. Server machine 370 includes training set generator 372 capable of generating a training dataset (e.g., a set of data inputs and a set of target outputs) for training, validating, and / or testing a machine learning model 390 or a set of machine learning models 390. Some operations of training set generator 372 are described in more detail below with respect to Figures 5 and 6A. In some embodiments, training set generator 372 can partition the training data into a training set, a validation set, and a test set.

[0051] The server machine 380 may include a training engine 382. An engine may refer to hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, a processing device, etc.), software (e.g., instructions executed on a processing device, a general-purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. The training engine 382 may be capable of training one machine learning model 390 or a set of machine learning models 390. The machine learning model 390 may refer to a model artifact resulting from the training engine 382's use of training data. The training data may include training inputs and corresponding target outputs (correct responses for each training input). The training engine 382 may discover patterns in the training data that map the training inputs to target outputs (predicted responses). The training engine 382 may then ultimately provide a machine learning model 390 that captures these patterns. The machine learning model 390 may include a linear regression model, a partial least squares regression model, a Gaussian regression model, a random forest model, a support vector machine model, a neural network, a ridge regression model, etc. In some embodiments, machine learning model 390 is a physics-based model instead of or in addition to being a machine learning model.

[0052] The training engine 382 may also be capable of validating the trained machine learning models 390 using a corresponding set of features of the validation set from the training set generator 372. In some embodiments, the training engine 382 may assign a performance grade to each of the set of trained machine learning models 390. The performance grade may correspond to the accuracy of each trained model, the speed of each model, and / or the efficiency of each model. The training engine 382 may select trained machine learning models 390 having a performance grade that satisfies performance criteria to be used by the prediction engine 314, according to some embodiments described herein. Further details regarding the training engine 382 are provided with respect to FIG. 6A .

[0053] The prediction server 312 includes a prediction engine 314 that can provide data from the optical property measurement tool 326 (e.g., optical transmission data and / or reflectance data) as input to a trained machine learning model 390. The prediction engine 314 can run the trained model 390 on the input to obtain one or more outputs. In embodiments, the trained model 390 is trained on training data that includes historical substrate optical transmission data, historical substrate reflectance data, and corresponding substrate identification data, such as corresponding substrate type data and historical substrate characteristic data (e.g., historical substrate coating data). As described further with respect to FIG. 6B , in some embodiments, the prediction engine 314 processes the input data (e.g., optical transmittance and / or optical reflectance) using the model 390 to predict a substrate identification.

[0054] It should be noted that in some other embodiments, the functionality of server machines 370 and 380 and prediction server 312 may be provided by more or fewer machines. For example, in some embodiments, server machines 370 and 380 may be combined into a single machine. In other embodiments, server machines 370 and 380 and / or prediction server 312 may be combined into a single machine. In general, functionality described in one embodiment as being performed by server machine 370, server machine 380, and / or prediction server 312 may also be performed on client device 320. Additionally, functionality attributed to particular components may also be performed by different or multiple components operating together.

[0055] 4 is a flowchart of a method 400 for updating an optical threshold corresponding to a substrate ID, according to an aspect of the present disclosure. Method 400 is performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one embodiment, method 400 may be performed by a computer system, such as computer system architecture 300 of FIG. 3. In other or similar embodiments, one or more operations of method 400 may be performed by one or more other machines not shown.

[0056] At block 402, processing logic measures the light intensity of an optical sensor in the absence of a substrate. The optical sensor may be a first optical sensor of an optical characteristic measurement tool (e.g., comprised of optical transmitter 230A and optical receiver 230B of FIGS. 2A-2C). Measuring the light intensity in the absence of a substrate may provide a benchmark light intensity value and / or an initial normalized calibration value.

[0057] In block 404, the light intensity of the light sensor is measured as the substrate is moved through the light path of the light sensor (e.g., by a substrate handling robot end effector). A portion of the light emitted by the light emitting element of the light sensor can be transmitted through the substrate to the light receiving element of the light sensor. The light intensity measured in block 404 may be substantially less than the intensity measured in block 402. In some embodiments, the tendency of the substrate to transmit light may affect the light intensity measured in block 404. For example, the less light the substrate transmits, the weaker the light intensity measured in block 404 will be; and the more light the substrate transmits, the stronger the light intensity measured in block 404 will be.

[0058] At block 406, processing logic determines the optical transmission ratio of the substrate. The optical transmission ratio may be the ratio of the light intensity measured at block 404 to the light intensity measured at block 402. In some embodiments, the optical transmission ratio of the substrate is stored as a characteristic for a substrate ID. The substrate ID may be a set of data corresponding to the substrate (e.g., substrate type, substrate coating, substrate optical transmission ratio, etc.). The substrate ID may be used to automatically update manufacturing system parameters (e.g., optical thresholds described herein, etc.) for handling and / or processing the substrate.

[0059] At block 408, processing logic calculates an optical threshold (e.g., a detected light intensity threshold of a substrate position sensor described herein) corresponding to the substrate ID. In some embodiments, the optical threshold is calculated (e.g., determined) based on an optical transmission ratio.

[0060] At block 410, processing logic updates the optical threshold for the corresponding substrate ID being processed. One or more position sensors can use the updated optical threshold to determine the position of the substrate during handling. In some embodiments, the new (e.g., updated) optical threshold is saved as a characteristic for the substrate ID.

[0061] 5 illustrates a model training workflow 505 and a model application workflow 517 for determining a predicted substrate identification for substrate optical property data, according to one embodiment. The model training workflow 505 and the model application workflow 517 may be performed by processing logic executed by a processor of a computing device. One or more of these workflows 505, 517 may be performed, for example, by one or more machine learning models implemented on the processing device and / or other software and / or firmware executing on the processing device.

[0062] The model training workflow 505 is for training one or more machine learning models (e.g., deep learning models) to determine predicted substrate identification. The model application workflow 517 is for applying the one or more trained machine learning models to perform substrate identification. Each of the substrate optical property data 512 can include optical transmission data and / or reflectance data (e.g., first reflectance data of a first surface and / or second reflectance data of a second surface) corresponding to one or more measured substrates. In some embodiments, the substrate optical property data can be generated via system 200A, 200B, or 200C described herein above.

[0063] Various machine learning outputs are described herein. Specific numbers and arrangements of machine learning models are described and illustrated. However, it should be understood that the number and types of machine learning models used, as well as the arrangements of such machine learning models, can be modified to achieve the same or similar end results. Therefore, the arrangements of the machine learning models described and illustrated are merely examples and should not be construed as limiting.

[0064] In some embodiments, one or more machine learning models are trained to perform one or more substrate identification tasks. Each task may be performed by a separate machine learning model. Alternatively, a single machine learning model may perform each of the tasks or portions of the tasks. For example, a first machine learning model may be trained to determine substrate identification, and a second machine learning model may be trained to determine corresponding corrective actions. Additionally or alternatively, different machine learning models may be trained to perform different combinations of these tasks. In one example, one or several machine learning models may be trained. The trained machine learning (ML) model may be a single shared neural network with multiple shared layers and multiple separate higher-level output layers, each of which outputs a different prediction, classification, identification, etc. For example, a first higher-level output layer may determine substrate identification based on input data corresponding to a first substrate, and a second higher-level output layer may determine substrate identification based on input data corresponding to a second substrate.

[0065] One type of machine learning model that can be used to perform some or all of the above tasks is an artificial neural network, such as a deep neural network. An artificial neural network generally includes a feature representation component with a classifier or recurrent layer that maps features to a target output space. A convolutional neural network (CNN), for example, hosts multiple layers of convolutional filters. Deep learning is a type of machine learning algorithm that uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. Deep neural networks can learn in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. Deep neural networks include a hierarchy of multiple layers, with different layers learning different levels of representation corresponding to different levels of abstraction. In deep learning, each level learns to transform its input data into slightly more abstract and complex representations. In particular, the deep learning process can learn which features naturally best fit into which levels. The "deep" in "deep learning" refers to the number of layers through which data is transformed. More precisely, deep learning systems have substantial Contribution Allocation Path (CAP) depth. A CAP is a chain of transformations from input to output. A CAP describes the potentially causal connections between input and output. For forward neural networks, the CAP depth can be the depth of the network, which can be the number of hidden layers plus one. For recurrent neural networks, where a signal can propagate through a layer more than once, the CAP depth is potentially unlimited.

[0066] Training a neural network can be accomplished in a supervised learning manner, which involves feeding a training data set of labeled inputs through the network, observing its outputs, defining an error (by measuring the difference between the output and the label value), and using techniques such as deep gradient descent and backpropagation to adjust the mass of the network across all of its layers and nodes to minimize the error. In many applications, repeating this process across many labeled inputs in the training data set results in a network that can provide the correct output when presented with inputs different from those present in the training data set.

[0067] For the model training workflow 505, a training dataset including hundreds, thousands, tens of thousands, hundreds of thousands, or more instances of substrate optical property data 512 (e.g., optical transmission data, reflectance data, etc.) should be used to form the training dataset. The data may include, for example, a substrate optical transmission ratio determined using a given number of measurements. The data may further include, for example, a first reflectance value of a first surface of the substrate (e.g., a top surface of the substrate) and / or a second reflectance value of a second surface of the substrate (e.g., a bottom surface of the substrate). This data may be processed to generate one or more training datasets 536 for training one or more machine learning models. The training data items in the training dataset 536 may include the substrate optical property data 512, corresponding substrate identification data (e.g., identification of the type of the corresponding substrate, identification of a coating on the corresponding substrate, etc.), and / or one or more images of the substrate.

[0068] To accomplish training, processing logic inputs a training data set 536 into one or more untrained machine learning models. Before inputting a first input to the machine learning models, the machine learning models may be initialized. Processing logic trains the untrained machine learning models based on the training data set to generate one or more trained machine learning models that perform the various operations described above. Training may be performed by inputting input data, such as substrate optical property data 512, images, and / or substrate identification data, into the machine learning models one at a time.

[0069] A machine learning model processes this input to generate an output. An artificial neural network includes an input layer consisting of values ​​in the data points. The next layer is called the hidden layer, and each node in the hidden layer receives one or more of the input values. Each node includes parameters (e.g., masses) to apply to the input values. Thus, each node essentially inputs the input values ​​into a multivariate function (e.g., a nonlinear mathematical transformation) to produce an output value. The next layer can be another hidden layer or an output layer. In either case, the nodes in the next layer receive output values ​​from the nodes in the previous layer, and each node applies masses to those values ​​and then generates its own output value. This can be done at each layer. The final layer is the output layer, where there is one node for each class, prediction, and / or output that the machine learning model can produce.

[0070] Thus, the output may include one or more predictions or inferences (e.g., predicted substrate type identification, predicted substrate coating, etc.). Processing logic may compare the output predicted substrate identification with past substrate identifications. Processing logic may determine an error (i.e., classification error) based on the difference between the predicted substrate identification and the target substrate identification. Processing logic may adjust the mass of one or more nodes in the machine learning model based on the error. An error term or delta may be determined for each node in the artificial neural network. Based on this error, the artificial neural network adjusts one or more of its parameters (masses for one or more inputs of the node) for one or more of its nodes. Parameters may be updated using a backpropagation method, such that nodes in the highest layer are updated first, followed by nodes in the next layer, and so on. An artificial neural network includes multiple layers of "neurons," each layer receiving as input values ​​from neurons in the previous layer. The parameters for each neuron include a mass associated with values ​​received from each of the neurons in the previous layer. Accordingly, adjusting the parameters may include adjusting masses assigned to each of the inputs to one or more neurons in one or more layers within the artificial neural network.

[0071] After the model parameters are optimized, model validation can be performed to determine whether the model has improved and to determine the current accuracy of the deep learning model. After one or more training rounds, the processing logic can determine whether a stopping criterion has been met. The stopping criterion can be a target accuracy level, a target number of processed data points from the training dataset, a target amount of change to the parameters relative to one or more previous data points, combinations thereof, and / or other criteria. In one embodiment, the stopping criterion is met when at least a minimum number of data points have been processed and at least a threshold accuracy has been achieved. The threshold accuracy can be, for example, 70%, 80%, or 90% accuracy. In one embodiment, the stopping criterion is met when the accuracy of the machine learning model stops improving. If the stopping criterion is not met, further training is performed. If the stopping criterion is met, training can be completed. After the machine learning model is trained, the model can be tested using a reserved portion of the training dataset. After one or more trained machine learning models 538 are generated, these machine learning models 538 can be stored in model storage 545 and added to the substrate identification engine 530.

[0072] According to one embodiment, for the model application workflow 517, the input data 562 can be input into one or more substrate identification determiners 567, each of which can include a trained neural network or other model. Additionally or alternatively, the one or more substrate identification determiners 567 can apply a data processing algorithm to determine the substrate identification. The input data can include substrate optical transmission data and / or substrate reflectance data (e.g., measured / generated using an optical property measurement tool described herein). The input data can additionally, optionally, include one or more images of the substrate being measured. Based on the input data 562, the substrate identification determiner 567 can output one or more substrate identification predictions 569. The substrate identification predictions 569 can include a predicted identification (e.g., type, coating, etc.) of the substrate being analyzed.

[0073] The action determiner 572 can determine one or more actions 570 to take based on the substrate identification prediction 569. In one embodiment, the action determiner 572 compares the substrate identification prediction 569 to one or more expected substrate identifications. If one or more of the substrate identification predictions 569 do not match the expected substrate identifications, the action determiner 572 can determine that a corrective action is recommended. In such cases, the action determiner 572 can output a recommendation or notification to perform the corrective action (e.g., flipping the substrate, updating the process recipe, etc.). In some embodiments, the action determiner 572 automatically updates process parameters based on the substrate identification prediction 569 satisfying one or more criteria. In some examples, the substrate identification prediction 569 can include an estimated optical threshold (e.g., a light intensity threshold) for locating the substrate by a position sensor. In some embodiments, the estimated optical threshold can be used to determine the position of the substrate relative to a robot end effector while the substrate is being transported.

[0074] 6A is a flowchart of a method 600A for generating a training dataset for training a machine learning model to perform substrate identification, according to an aspect of the present disclosure. Method 600A is performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one embodiment, method 600A may be performed by a computer system, such as computer system architecture 300 of FIG. 3. In other or similar embodiments, one or more operations of method 600A may be performed by one or more other machines not shown.

[0075] At block 610, processing logic initializes the training set T to an empty set (e.g., {}).

[0076] At block 612, processing logic acquires substrate optical property information (e.g., data associated with one or more optical properties of the substrate, such as optical transmittance, reflectance, etc.) associated with a substrate handled in the substrate processing system. As described above, the substrate optical property data may be acquired by an optical property measurement tool (e.g., optical property measurement tool 326 of FIG. 3 ) or a system for determining substrate optical transmittance and / or substrate reflectance (e.g., system 200A, system 200B, or system 200C), as described herein. In some embodiments, processing logic acquires historical substrate optical property data corresponding to substrates previously handled in one or more substrate processing systems. In some embodiments, the historical substrate optical property data is uploaded and / or input into the processing logic (e.g., by a user, such as an engineer or technician) from one or more data sheets corresponding to past substrates.

[0077] At block 614, processing logic retrieves substrate identification information associated with the substrate. In some embodiments, the substrate identification information may include information such as substrate type and / or information associated with one or more substrate coatings (e.g., on the top or bottom surface of the substrate). In some embodiments, processing logic retrieves past substrate identification information corresponding to past identifications of the substrate (e.g., past substrate types, past substrate coatings, etc.). In some embodiments, the past substrate identification information is uploaded and / or input into the processing logic from one or more data sheets corresponding to past substrates.

[0078] At block 616, processing logic generates training inputs based on the information obtained for the substrate optical properties at block 612. In some embodiments, the training inputs may include a normalized set of sensor data (e.g., normalized optical transmission data, normalized reflectance data, etc.).

[0079] At block 618, processing logic may generate target outputs based on the substrate identification information obtained at block 614. The target outputs may correspond to substrate identification metrics (data indicative of an identified substrate type and / or an identified substrate coating) for substrates handled within the substrate processing system.

[0080] At block 620, processing logic generates an input / output mapping. The input / output mapping refers to training inputs that include or are based on information about substrate optical properties and target outputs for the training inputs, where the target outputs identify the substrate identity, and the training inputs are associated with (or mapped to) the target outputs. At block 622, processing logic adds the input / output mapping to a training set T.

[0081] At block 624, processing logic determines whether training set T includes a sufficient amount of training data for training the machine learning model. Note that in some embodiments, training set T may be determined to be sufficient based solely on the number of input / output mappings in the training set, while in some other embodiments, training set T may be determined to be sufficient based on one or more other criteria (e.g., the degree of diversity of the training examples) in addition to or instead of the number of input / output mappings. In response to determining that training set T includes a sufficient amount of training data for training the machine learning model, processing logic provides training set T for training the machine learning model. In response to determining that the training set does not include a sufficient amount of training data for training the machine learning model, method 600 returns to block 612.

[0082] At block 626, processing logic provides a training set T for training the machine learning model. In some embodiments, the training set T is provided to a training engine 382 of a server machine 380 (e.g., of FIG. 3 ) to perform the training. In the case of a neural network, for example, input values ​​of a given input / output mapping (e.g., spectral data and / or chamber data for a previous substrate) are input to the neural network, and output values ​​of the input / output mapping are stored in output nodes of the neural network. The connection mass in the neural network is then adjusted according to a learning algorithm (e.g., backpropagation, etc.), and the procedure is repeated for other input / output mappings in the training set T. After block 626, the machine learning model (e.g., machine learning model 390 of FIG. 3 ) can be used to provide predicted substrate identifications for substrates handled in the substrate processing system.

[0083] 6B is a flowchart of a method 600B for generating a predicted substrate identification using a trained machine learning model, according to an aspect of the present disclosure. Method 600B is performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one embodiment, method 600B may be performed by a computer system, such as computer system architecture 300 of FIG. 3. In other or similar embodiments, one or more operations of method 600B may be performed by one or more other machines not shown.

[0084] At block 652, processing logic receives data associated with one or more optical properties (e.g., optical transmission ratio, reflectance, etc.) of a substrate handled by the substrate processing system. In some embodiments, the data is received from an optical property measurement tool (e.g., optical property measurement tool 326 of FIG. 3 ) or a system for determining substrate optical transmittance (e.g., ratio) and / or substrate reflectance (e.g., system 200A, system 200B, or system 200C), as described herein. The data may be raw sensor data or may be data that has been processed (e.g., by a processing device, computing device, etc.) to determine the substrate optical transmission ratio and / or substrate reflectance.

[0085] At block 654, processing logic inputs the data received at block 652 into a trained machine learning model. In some embodiments, the trained machine learning model is trained using the techniques described herein with reference to FIG. 3, FIG. 5, and / or FIG. 6A. The trained machine learning model can be trained with data inputs including historical substrate optical property data (e.g., historical optical transmission data, historical reflectance data, etc.). The trained machine learning model can be trained with target output data including historical substrate identification data (e.g., historical corresponding substrate types, historical substrate coatings, etc.). The trained machine learning model can be trained to output one or more predicted substrate identifications based on the data inputs associated with the substrate optical properties. In some embodiments, the trained machine learning model is trained to output data indicative of one or more predicted substrate identifications.

[0086] At block 656, processing logic receives output from the trained machine learning model including a predicted substrate identification corresponding to the measured substrate. In some embodiments, the predicted substrate identification (e.g., predicted substrate type, predicted substrate coating, etc.) may influence the determination of the substrate position (e.g., relative to the position of a substrate handling robot end effector handling the substrate). The predicted substrate identification may be used to accurately determine optical thresholds for precisely locating the substrate and / or for performing corrective action, as described herein.

[0087] 7 is a flowchart of a method 700 for determining a substrate position based on a determined optical threshold of the substrate, according to an aspect of the present disclosure. Method 700 is performed by a system that may include hardware (such as circuitry described herein, dedicated logic, optical measurement tools, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one embodiment, method 700 may be performed by a computer system such as computer system architecture 300 of FIG. 3. In other or similar embodiments, one or more operations of method 700 may be performed by one or more other machines not shown.

[0088] At block 702, processing logic receives optical transmission data from an optical sensor (e.g., a first optical sensor comprised of optical transmitter 230A and optical receiver 230B of FIGS. 2A-2C). In some embodiments, the optical transmission data indicates an optical transmission ratio of a substrate positioned in the optical path of the optical sensor. The substrate may be supported in the optical path by a substrate support, such as a substrate handling robot end effector or an aligner stage. In some embodiments, the substrate handling robot end effector transports the substrate from a first station (e.g., a substrate carrier) to a second station (e.g., a load lock) through the optical path of the optical sensor.

[0089] At block 704, processing logic determines an optical threshold based on the optical transmission data. In some embodiments, the optical threshold corresponds to an optical transmission ratio of the substrate. For example, the optical transmission ratio may indicate that a certain percentage (e.g., 50%) of light is transmitted through the substrate. In some embodiments, the optical threshold may be slightly higher than the optical transmission ratio. In some examples, the optical threshold is such that the substrate position sensor can detect the presence of the substrate when the substrate passes through the beam of the position sensor. For example, for a substrate having an optical transmission ratio of 0.2 (e.g., corresponding to 20% of light being transmitted through the substrate), the optical threshold may be approximately 25%. Thus, when the position sensor detects a transmitted light intensity of 25% or less, the sensor data may indicate the presence of a substrate.

[0090] At block 706, processing logic receives position sensor data indicating the position of the substrate along the substrate transport path. The processing logic can correlate the time at which the position sensor indicates the presence of the substrate (e.g., position sensor trigger) with the position (e.g., related to the time domain) at which the substrate handling robot transports the substrate along the substrate transport path.

[0091] At block 708, processing logic determines the position of the substrate relative to a substrate handling robot end effector transporting the substrate. The position may be based on position sensor data and optical thresholds. In some examples, processing logic may determine a displacement of the substrate from the end effector. In some examples, processing logic determines a distance by which the center of the substrate is displaced relative to the center of the robot end effector. The substrate may be displaced due to motion of the substrate from acceleration of the end effector.

[0092] FIG. 8 shows a diagrammatic representation of a machine in the exemplary form of a computing device 800 capable of executing a set of instructions to cause the machine to perform any one or more of the techniques discussed herein. In alternative embodiments, the machine may be connected (e.g., networked) to other machines in a local area network (LAN), an intranet, an extranet, or the Internet. The machine may operate in the capacity of a server or client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), a tablet computer, a set-top box (STB), a personal digital assistant (PDA), a cellular telephone, a web appliance, a server, a network router, switch, or bridge, or any machine capable of executing (sequentially or otherwise) a set of instructions that specify actions to be taken by the machine. Moreover, while only a single machine is shown, the term “machine” should also be taken to include any group of machines (e.g., computers) that individually or jointly execute a set (or sets) of instructions to perform any one or more of the techniques discussed herein. In an embodiment, computing device 800 may correspond to one or more of server machine 370, server machine 380, or prediction server 312, as described herein.

[0093] The exemplary computing device 800 includes a processing device 802, a main memory 804 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM), e.g., synchronous DRAM (SDRAM), etc.), a static memory 806 (e.g., flash memory, static random access memory (SRAM), etc.), and a secondary memory (e.g., a data storage device 828), which communicate with each other via a bus 808.

[0094] The processing device 802 may represent one or more general-purpose processors, such as a microprocessor, a central processing unit, or the like. More specifically, the processing device 802 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing a combination of instruction sets. The processing device 802 may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, or the like. The processing device 802 may also be or include a system-on-chip (SoC), a programmable logic controller (PLC), or other type of processing device. The processing device 802 is configured to execute processing logic for performing the operations discussed herein.

[0095] Computing device 800 may further include a network interface device 822 for communicating with a network 864. Computing device 800 may also include a video display unit 810 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 812 (e.g., a keyboard), a cursor control device 814 (e.g., a mouse), and a signal generation device 820 (e.g., a speaker).

[0096] The data storage device 828 may include a machine-readable storage medium (or more specifically, a non-transitory machine-readable storage medium) 824 having stored thereon one or more sets of instructions 826 that implement any one or more of the techniques or functions described herein. A non-transitory storage medium refers to a storage medium other than a carrier wave. The instructions 826 may also reside, completely or at least partially, within the main memory 804 and / or within the processing device 802 during execution thereof by the computing device 800, the main memory 804, and the processing device 802, which also constitute computer-readable storage media.

[0097] While the exemplary embodiment shows computer-readable storage medium 824 as a single medium, the term "computer-readable storage medium" should be interpreted to include a single medium or multiple media (e.g., centralized or distributed databases, and / or associated caches and servers) that store one or more sets of instructions. The term "computer-readable storage medium" should also be interpreted to include any medium capable of storing or encoding a set of instructions for execution by a machine that cause the machine to perform any one or more of the techniques of this disclosure. Thus, the term "computer-readable storage medium" should be interpreted to include, but is not limited to, solid-state memory and optical and magnetic media.

[0098] The above description sets forth numerous specific details, such as examples of specific systems, components, methods, etc., to provide a thorough understanding of some embodiments of the present disclosure. However, it will be apparent to those skilled in the art that at least some embodiments of the present disclosure can be practiced without these specific details. In other instances, well-known components or methods have not been described in detail or have been presented in simple block diagram form to avoid unnecessarily obscuring the present disclosure. Thus, the specific details described are merely exemplary. It is contemplated that particular embodiments may vary from these exemplary details and still be within the scope of the present disclosure.

[0099] Throughout this specification, a reference to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment. Thus, the appearances of the phrase "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. In addition, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or." When the term "about" or "approximately" is used herein, it is intended to mean that the stated nominal value is accurate to within ±10%.

[0100] Although the method operations herein are illustrated and described in a particular order, the order of the method operations may be changed, such that certain operations may be performed in reverse order, and certain operations may be performed at least partially concurrently with other operations. In alternative embodiments, instructions or sub-operations of separate operations may be performed intermittently and / or alternately.

[0101] It is understood that the above description is intended to be illustrative, and not limiting. Many other embodiments will be apparent to those skilled in the art upon reading and understanding the above description. Accordingly, the scope of the present disclosure should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. a first photosensor including a light emitting element and a light receiving element, the first photosensor configured to detect a light transmission ratio of the substrate in response to a substrate support supporting the substrate within an optical path of the first photosensor; a processing device communicatively coupled to the first optical sensor, the processing device comprising: determining an optical threshold based on the optical transmission ratio of the substrate; and determining a position of the substrate relative to a substrate handling robot end effector based on sensor data output by a position sensor and the optical threshold. system.

2. a second light sensor including a second light emitting element and a second light receiving element, the second light sensor configured to detect a first reflectance of the first surface of the substrate; the processing device is further configured to determine a first property of the first surface of the substrate based on the first reflectance. The system of claim 1 .

3. a third optical sensor including a third light emitting element and a third light receiving element, the third optical sensor configured to detect a second reflectance of the second surface of the substrate; the processing device is further configured to determine a second property of the second surface of the substrate based on the second reflectance. The system of claim 2 .

4. The system of claim 3 , wherein the first property is associated with a first coating on the first surface of the substrate and the second property is associated with a second coating on the second surface of the substrate.

5. The system of claim 3 , wherein the processing device is further configured to determine a corresponding substrate type based on one or more of the optical transmission ratio, the first reflectance, or the second reflectance.

6. 10. The system of claim 1, wherein the first optical sensor is positioned in a factory interface chamber of a manufacturing system proximate to a vacuum port, and the optical path of the first optical sensor at least substantially intersects a transport path of the substrate through the vacuum port.

7. 10. The system of claim 1, wherein the first optical sensor is positioned in a factory interface chamber of a manufacturing system proximate to an aligner configured to align the substrate to a target orientation, and the optical transmission ratio is detected in response to the substrate being placed on the aligner.

8. The system of claim 1 , wherein the first optical sensor is coupled to the substrate handling robot end effector.

9. 9. The system of claim 8, wherein the light-emitting element is coupled to a first side of a generally Y-shaped bracket and the light-receiving element is coupled to a second side of the generally Y-shaped bracket, the generally Y-shaped bracket configured to move between an extended position and a retracted position in response to handling of the substrate by the end effector.

10. receiving optical transmission data from a first optical sensor, the optical transmission data indicative of an optical transmission ratio of a substrate positioned within an optical path of the first optical sensor; determining an optical threshold based on the optical transmission data; receiving position sensor data indicative of a position of the substrate along a substrate transport path; determining a position of the substrate relative to a substrate handling robot end effector based on the position sensor data and the optical threshold; and A method comprising:

11. determining the optical transmission ratio of the substrate; positioning the substrate in the optical path of the first optical sensor on a substrate support; emitting light radiation through a light emitting element of the first light sensor; and detecting, via a light receiving element of the first optical sensor, an intensity of radiation transmitted through the substrate along the optical path.

12. the light emitting element is coupled to a first side of a generally Y-shaped bracket of the substrate handling robot end effector, and the light receiving element is coupled to a second side of the generally Y-shaped bracket, and the method comprises: The method of claim 11 , further comprising moving the generally Y-shaped bracket between an extended position and a retracted position in response to handling of the substrate by the end effector.

13. determining a first characteristic of the first surface of the substrate based on first reflectance data output by a second optical sensor; determining a second characteristic of the second surface of the substrate based on second reflectance data output by a third optical sensor; and The method of claim 10 further comprising:

14. determining a corresponding substrate type based on one or more of the optical transmission ratio, the first reflectance data, or the second reflectance data; 14. The method of claim 13, further comprising:

15. determining the corresponding substrate type; inputting data including one or more of the light transmission data, the first reflectance data, or the second reflectance data into a trained machine learning model; and receiving output from the trained machine learning model indicative of predicted substrate identification data, wherein the predicted substrate identification data corresponds to one or more of a predicted corresponding substrate type, a predicted first characteristic of the substrate, or a predicted second characteristic of the substrate.

16. training a machine learning model to create the trained machine learning model, wherein the machine learning model is trained using (i) training input data including one or more of past optical transmission data, past first reflectance data, or past second reflectance data, and (ii) target output data including past substrate identification data, wherein the past substrate identification data corresponds to one or more of a past corresponding substrate type, a past first characteristic, or a past second characteristic; 16. The method of claim 15.

17. A non-transitory machine-readable storage medium containing instructions that, when executed by a processing device, cause the processing device to perform operations, the operations including: receiving data including one or more of optical transmission data associated with a substrate, first reflectance data associated with a first surface of the substrate, or second reflectance data associated with a second surface of the substrate; inputting one or more of the light transmission data, the first reflectance data, or the second reflectance data into a trained machine learning model; and receiving output from the trained machine learning model indicative of predicted substrate identification data, wherein the predicted substrate identification data corresponds to one or more of a predicted corresponding substrate type, a predicted first characteristic of the substrate, or a predicted second characteristic of the substrate.

18. 20. The non-transitory machine-readable storage medium of claim 17, wherein the trained machine learning model is trained using (i) training input data including one or more of past optical transmission data, past first reflectance data, or past second reflectance data, and (ii) target output data including past substrate identification data, wherein the past substrate identification data corresponds to one or more of a past corresponding substrate type, a past first characteristic, or a past second characteristic.

19. receiving the data, receiving the light transmission data from a first light sensor, the light transmission data indicating a light transmission ratio of the substrate; receiving the first reflectance data from a second optical sensor configured to detect a first reflectance of the first surface of the substrate; and receiving the second reflectance data from a third optical sensor configured to detect a second reflectance of the second surface of the substrate.

20. the processing device: positioning the substrate on a substrate support within an optical path of the first optical sensor; emitting light radiation through a light emitting element of the first light sensor; 20. The non-transitory machine-readable storage medium of claim 19, for performing operations further comprising detecting, via a light-receiving element of the first photosensor, an intensity of radiation transmitted through the substrate along the optical path.

Images