A method, system, device, and medium for visual monitoring of semiconductor devices.

Abstract

This application provides a method, system, device, and medium for visual monitoring of semiconductor equipment. The method includes collecting multi-source data during the operation of the semiconductor equipment; dividing the task into multiple sub-tasks based on action primitive types; calculating the effective execution window of each action primitive; constructing task fragments accordingly; forming a task fragment set; initially selecting currently executing or not yet executed task fragments from the task fragment set to form a task fragment candidate set; comparing any two task fragments in the task fragment candidate set to filter out task fragment pairs with overlapping effective execution windows, generating a task fragment pair set; and performing time conflict and spatial conflict determination on each task fragment pair in the task fragment pair set. This application uses two-dimensional lightweight simulation to achieve visual monitoring of the semiconductor equipment's operating arm operation process, improving real-time performance while enabling rapid monitoring and intuitive display of potential interference risks from multiple operating arms.

Description

Technical Field

[0001] This application relates to semiconductor technology, and more particularly to a method, system, device, and medium for visual monitoring of semiconductor devices. Background Technology

[0002] In semiconductor wafer manufacturing equipment, common robotic arms for wafer transfer include fixed 180° opposing dual-arm robotic arms, non-fixed 180° dual-arm robotic arms, and extended robotic arms containing more than two arms. When these robotic arms perform wafer handling, loading, and handover between multiple process cavities, the spatial trajectories and task timing of the arms are prone to overlap or contention.

[0003] Existing technologies typically avoid conflicts through low-level scheduling logic, but operators find it difficult to intuitively understand the operational status through numerical values ​​or light status. Once the scheduling logic fails, conflicts are often only discovered at the physical level, leading to wafer damage or process interruption. While existing 3D animation monitoring is intuitive, it consumes too many resources and is unsuitable for web-based and low-performance terminals.

[0004] Therefore, there is a need for a visual monitoring method, system, device, and medium for semiconductor devices that can achieve real-time monitoring and conflict warning in a lightweight manner. Summary of the Invention

[0005] This application provides a method, system, device, and medium for visual monitoring of semiconductor devices to solve the problems of the prior art.

[0006] In a first aspect, this application provides a method for visually monitoring semiconductor devices, including:

[0007] Data modeling involves collecting multi-source data during the operation of semiconductor equipment, breaking down tasks into multiple sub-tasks based on action primitive types, calculating the effective execution window for each action primitive, and constructing task fragments accordingly to form a set of task fragments.

[0008] A task fragment pair set is generated. In the task fragment set, task fragments that are being executed or have not yet been executed are initially selected to form a task fragment candidate set. Any two task fragments in the task fragment candidate set are compared, and task fragment pairs with overlapping effective execution windows are selected to generate a task fragment pair set.

[0009] Two-dimensional conflict determination involves performing temporal and spatial conflict determinations for each task segment pair in the task segment pair set, wherein:

[0010] The time conflict is determined when the target workstations of the two task segments in the task segment pair are the same, and the duration of the time overlap of their effective execution windows is greater than the global time safety interval.

[0011] The spatial conflict determination is performed in a two-dimensional simulation parameter space. Based on the global angle safety interval, the global radius safety interval, and the operating arm buffer width, the angle interval and the radius interval are safely expanded to obtain a first normalized angle interval, a second normalized angle interval, a first radius interval, and a second radius interval. When the angle interval satisfies the two-dimensional angle intersection determination rule and the radius interval satisfies the two-dimensional radius intersection determination rule, the task segment is determined to have a spatial conflict.

[0012] In one possible design, the task segment that is being executed or has not yet been executed is determined by the forward monitoring time window. When the effective execution window of the task segment overlaps with the forward monitoring time window, the task segment is included in the task segment candidate set.

[0013] The forward-looking monitoring time window is based on the current system time and consists of the de-jitter buffer time and the forward-looking duration. The lower bound of the forward-looking monitoring time window is the current system time minus the de-jitter buffer time, and the upper bound of the forward-looking monitoring time window is the current system time plus the forward-looking duration.

[0014] In one possible design, the spatial conflict determination includes:

[0015] Angle interval calculation and safety expansion: For each pair of task segments, calculate the lower bound and upper bound of the angle interval of the two task segments respectively to obtain the first angle interval and the second angle interval, and perform safety expansion according to the global angle safety interval.

[0016] Radius interval calculation and safety expansion: For each pair of task segments, the lower bound and upper bound of the radius interval of the two task segments are calculated respectively, and the expansion is performed according to the global radius safety interval and the operating arm buffer width to obtain the first radius interval and the second radius interval.

[0017] Angle normalization: Normalize the first angle interval and the second angle interval to... This forms the first normalized angle interval and the second normalized angle interval;

[0018] Two-dimensional intersection determination is performed based on the two-dimensional angle intersection determination rules and the two-dimensional radius intersection determination rules, respectively, for the first normalized angle interval, the second normalized angle interval, the first radius interval, and the second radius interval.

[0019] Spatial conflict determination is based on the results of two-dimensional intersection determination.

[0020] In one possible design, the two-dimensional angle intervals are closed and intersecting under the condition that the lower bound of the first normalized angle interval is not greater than the upper bound of the second normalized angle interval, and the lower bound of the second normalized angle interval is not greater than the upper bound of the first normalized angle interval.

[0021] In one possible design, the two-dimensional angle intersection determination rule is:

[0022] If either the first normalized angle interval or the second normalized angle interval covers a complete circle, then the task segment is determined to have a two-dimensional angle intersection.

[0023] When both the first normalized angle interval and the second normalized angle interval do not cross zero, if the angle interval closure intersection condition is met, it is determined that the task segment pair has a two-dimensional angle intersection.

[0024] When the first normalized angle interval crosses zero and the second normalized angle interval does not cross zero, the first sub-interval and the second sub-interval of the first normalized angle interval are compared with the second normalized angle interval twice. If any comparison satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

[0025] When the first normalized angle interval does not cross zero and the second normalized angle interval crosses zero, the first sub-interval and the second sub-interval of the second normalized angle interval are compared with the first normalized angle interval twice. If either comparison satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

[0026] When both the first normalized angle interval and the second normalized angle interval cross zero, the first and second sub-intervals of the first normalized angle interval are compared one by one with the first and second sub-intervals of the second normalized angle interval. When any pair of sub-intervals satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

[0027] In one possible design, the two-dimensional radius intersection determination rule is that when the lower bound of the first radius interval is not greater than the upper bound of the second radius interval, and the lower bound of the second radius interval is not greater than the upper bound of the first radius interval, the task segment is determined to have two-dimensional radius intersection.

[0028] In one possible design, the visualization monitoring method further includes:

[0029] Conflict classification and early warning: Based on the judgment results of two-dimensional conflict determination, the conflict level of task segments with two-dimensional conflict is determined, and the corresponding operating arm is highlighted in the two-dimensional simulation interface, the target workstation involved is marked and flashed, and the conflict level prompt and arrival time or in progress prompt are displayed.

[0030] Secondly, this application provides a visual monitoring system for semiconductor devices, comprising:

[0031] The data modeling module collects multi-source data during the operation of semiconductor equipment, breaks down tasks into multiple sub-tasks based on action primitive types, calculates the effective execution window for each action primitive, and constructs task fragments accordingly to form a set of task fragments.

[0032] The task fragment pair generation module initially selects task fragments that are being executed or have not yet been executed from the task fragment set to form a task fragment candidate set. It compares any two task fragments in the task fragment candidate set, filters out task fragment pairs whose effective execution windows have time overlap, and generates a task fragment pair set.

[0033] The two-dimensional conflict determination module performs temporal conflict determination and spatial conflict determination for each task segment pair in the task segment pair set, wherein:

[0034] The time conflict is determined when the target workstations of the two task segments in the task segment pair are the same, and the duration of the time overlap of their effective execution windows is greater than the global time safety interval.

[0035] The spatial conflict determination is based on the global angle safety interval, the global radius safety interval, and the operating arm buffer width in the two-dimensional simulation parameter space. The angle interval and the radius interval are safely expanded to obtain the first normalized angle interval, the second normalized angle interval, the first radius interval, and the second radius interval. When the angle interval satisfies the two-dimensional angle intersection determination rule and the radius interval satisfies the two-dimensional radius intersection determination rule, the task segment is determined to have a spatial conflict.

[0036] The conflict classification and early warning module, based on the judgment results of two-dimensional conflict determination, determines the conflict level of task segments with two-dimensional conflicts, highlights the corresponding operating arm in the two-dimensional simulation interface, marks and flashes the target workstation involved, and displays the conflict level prompt and arrival time or in progress prompt.

[0037] Thirdly, this application provides an electronic device, comprising:

[0038] Processor; and,

[0039] Memory for storing the executable instructions of the processor;

[0040] The processor is configured to perform any of the possible methods described in the first aspect by executing the executable instructions.

[0041] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement any of the possible methods described in the first aspect.

[0042] This application provides a method, system, device, and medium for visual monitoring of semiconductor equipment. The method uses a unified data modeling and filtering process based on a chain of "action primitives, effective execution windows, task segments, and task segment pair sets." First, the effective execution window is calculated based on the action primitives, and a set of task segments is constructed. Then, a set of task segment pairs is generated only for segments with overlapping effective execution windows. In the two-dimensional simulation parameter space, safety expansion is performed on the angle and radius intervals based on the global angle safety interval, the global radius safety interval, and the operator arm buffer width. Finally, a two-dimensional conflict determination result is given based on the two-dimensional angle intersection determination rule and the two-dimensional radius intersection determination rule. This process balances real-time performance and computational efficiency, and reduces false alarms and missed alarms through safety expansion, facilitating online early warning and maintenance decisions, and improving equipment safety and maintainability.

[0043] Furthermore, by introducing a forward-looking monitoring time window, upcoming and recently occurring segments can be included in the monitoring, enabling predictive early warning and improving the convergence stability of time-based judgments. Combined with angle normalization and safety expansion, encoder errors, manufacturing tolerances, and trajectory fitting deviations can be effectively covered, further reducing false alarms and missed alarms. In conjunction with a conflict classification and early warning mechanism, target workstations can be highlighted, marked, and conflict levels displayed on the 2D simulation interface. The solution is applicable to fixed 180° opposed dual manipulators, non-fixed 180° manipulators, and multiple manipulators, offering low engineering integration and deployment costs and strong adaptability. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0045] Figure 1 This is a flowchart illustrating a method for visually monitoring a semiconductor device according to an example embodiment of this application;

[0046] Figure 2 This is a schematic diagram of the data modeling process according to an example embodiment of this application;

[0047] Figure 3This is a schematic diagram of the spatial conflict determination process according to an example embodiment of this application;

[0048] Figure 4 This is a schematic diagram illustrating the conflict classification and early warning process according to an example embodiment of this application;

[0049] Figure 5 This is a schematic diagram of the structure of a visual monitoring system for semiconductor devices according to an example embodiment of this application;

[0050] Figure 6 This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application.

[0051] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0052] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0053] This embodiment proposes a visual monitoring method, system, device, and medium for semiconductor devices, applicable to scenarios where a single robotic arm is equipped with multiple manipulators. These multiple manipulators include fixed 180° opposing dual manipulators, non-fixed 180° dual manipulators, and structures with two or more manipulators. This method provides lightweight monitoring of the manipulator operation process through two-dimensional simulation, avoiding the high computational overhead of three-dimensional rendering, and can intuitively and in real-time present potential interference risks.

[0054] Specifically, firstly, based on information such as task data (TaskHeader), trajectory data (KinematicEnvelopeData), action primitive timeline (PrimitiveTimeline), arm buffer width (ArmBuffer), and global safety threshold parameter (SafetyThresholdPara), the effective execution window of each action primitive is calculated, and standardized task segments are constructed, thus forming a fragmented data model in a two-dimensional simulation parameter space (angle-radius representation). Subsequently, judgments are made in both the time and spatial dimensions: in the time dimension, time (resource) contention among multiple manipulators at the same target workstation is identified; in the spatial dimension, potential spatial intersections at workstation entrances or within trajectory intervals are identified based on the safety expansion of angle and radius intervals. Furthermore, the judgment results are classified into resource conflicts, spatial conflicts, and compound conflicts, and warnings are provided in the visualization interface through manipulator highlighting, target workstation marker flashing, and conflict level labels.

[0055] It should be noted that the robotic arm comprises two or more manipulators. In this case, the robotic arm refers to the overall actuator, while the manipulator is a sub-unit on the robotic arm that specifically performs actions such as wafer gripping, handling, loading, and resetting.

[0056] This application employs lightweight two-dimensional simulation to achieve visualized monitoring of the operation process of semiconductor equipment manipulators. Compared with traditional three-dimensional rendering, this significantly reduces the computational and rendering burden and improves the real-time performance of the system. Simultaneously, two-dimensional visualization is used for conflict monitoring, identifying and intuitively presenting spatial interference and time (resource) contention based on angle-radius representation. This facilitates quick understanding and timely handling by operators. Without altering the original task scheduling logic, it enables rapid monitoring and intuitive display of potential interference risks from multiple manipulators, thereby improving the safety and maintainability of equipment operation.

[0057] Figure 1 This is a flowchart illustrating a method for visually monitoring a semiconductor device according to an example embodiment of this application. Figure 1 As shown, the semiconductor device visualization monitoring method provided in this embodiment includes:

[0058] Step S101: Data modeling. Collect multi-source data during the operation of semiconductor equipment. Based on the action primitive type, the task is divided into multiple sub-tasks, and the effective execution window of each action primitive is calculated. Based on this, task fragments are constructed to form a set of task fragments.

[0059] In this embodiment, a unified data model is first performed on the task execution process of the manipulator to serve as the basis for subsequent temporal and spatial conflict determination. By defining and associating task data (TaskHeader), trajectory data (KinematicEnvelopeData), action primitive timeline (PrimitiveTimeline), manipulator buffer width (ArmBuffer), and global safety threshold parameter (SafetyThresholdPara), the originally unstructured motion process is transformed into fragmented data units with clear temporal and spatial boundaries. This modeling method ensures that the motion states of different manipulators in the two-dimensional simulation parameter space (angle-radius representation) can be described with a unified structure, facilitating two-dimensional conflict detection and visualization.

[0060] It should be noted that this embodiment uses a motion primitive (SegmentType) modeling approach, decomposing each task into several minimum motion units in a two-dimensional simulation parameter space. After data acquisition, the effective execution window is calculated for each primitive record according to the rule of "actual takes precedence over plan." Subsequently, index and field mapping is completed when constructing the task segment, and all task segments are aggregated into a task segment set. The arm buffer width (ArmBuffer) and the global safety threshold parameter (SafetyThresholdPara) are device-level and system-level parameters, respectively, and are loaded during task segment generation for subsequent judgment. To facilitate tracking, the recording order of the task segment set (TaskSegmentSet) is consistent with the recording order of the action primitive timeline (PrimitiveTimeline).

[0061] The task fragment set is as follows:

[0062] .

[0063] The structure of each task fragment record in the task fragment set is as follows:

[0064]

[0065] in,

[0066] This is the i-th task segment;

[0067] TaskID is the task fragment. Task identifier;

[0068] ArmID is a mission fragment. Operating arm markings;

[0069] SequenceNo is a task fragment. In the action primitive type (SegmentType), the execution order number corresponding to an action primitive is SequenceNo. Under the same (TaskID, ArmID), SequenceNo is incremented from 1 and is unique.

[0070] SegmentType is a task segment The types of action primitives include Rotate, Extend, Dock, Retract, and Handover. ;

[0071] and , for task fragments rotation start and end angles and the start and end radius of the extension / retraction ;

[0072] StationRef is the target workstation corresponding to this action primitive.

[0073] EffectiveStartTime and EffectiveEndTime are task segments. The valid start time and valid end time;

[0074] , and , for task fragments The global security threshold parameter;

[0075] ArmBuffer is a task fragment. The operating arm buffer width.

[0076] The task segment serves as a unified input for subsequent time conflict and spatial conflict determination, enabling the operation of different manipulators to be analyzed and visualized using a standardized data model in a two-dimensional simulation parameter space.

[0077] Step S102: Task fragment pair set generation. In the task fragment set, task fragments that are being executed or have not yet been executed are initially selected to form a task fragment candidate set. Any two task fragments in the task fragment candidate set are compared, and task fragment pairs with overlapping effective execution windows are selected to generate a task fragment pair set.

[0078] In this embodiment, to achieve proactive monitoring of the operational risks of multiple manipulators, a forward-looking monitoring time window mechanism is introduced. This mechanism uses the current system time as a baseline, consisting of a debouncing buffer time and a forward-looking duration, to uniformly include upcoming or recently occurring action segments within the monitoring scope. This mechanism, on the one hand, provides predictive warnings of potential conflicts in the short future timeframe through "forward looking," and on the other hand, shields instantaneous jitter and measurement noise through "debouncing buffering," ensuring stable convergence of judgments in the time dimension. Simultaneously, in the spatial dimension, safety expansion of the angle and radius intervals (based on global angle safety intervals, global radius safety intervals, and manipulator buffer width) is used to cover manufacturing tolerances and measurement errors, thereby significantly reducing the probability of false alarms and missed alarms. The combination of forward-looking convergence in the time dimension and safety expansion in the spatial dimension achieves low-computing-power, low-latency, and intuitively visual online risk monitoring without altering the original task scheduling logic. Specifically:

[0079] First, based on the task fragment set, the task fragments are filtered according to the task fragment filtering criteria to form a candidate task fragment set.

[0080] This embodiment performs preliminary screening, selecting task segments that are currently being executed or have not yet been executed from the task segment set, and then determining spatial and temporal conflicts for them. Executed task segments do not require spatial and temporal conflict determination and are therefore removed to save computing resources for the semiconductor device and improve its operating efficiency.

[0081] The task segments that are being executed or have not yet been executed are determined by the forward monitoring time window. When the effective execution window of the task segment overlaps with the forward monitoring time window, the task segment is included in the task segment candidate set.

[0082] The forward-looking monitoring time window is based on the current system time and consists of the de-jitter buffer time and the forward-looking duration. The lower bound of the forward-looking monitoring time window is the current system time minus the de-jitter buffer time, and the upper bound is the current system time plus the forward-looking duration. Specifically:

[0083] Get the current system time Now, the preset lookahead duration T_lookahead (e.g., preset lookahead duration is 200 seconds), and the preset de-jitter buffer time. (For example, the preset de-shake buffer time is 15 seconds);

[0084] Iterate through each task segment record in the TaskSegmentSet;

[0085] If a task fragment record meets the task fragment filtering criteria, it is added to the task fragment candidate set.

[0086] After traversing the set of task fragments, the final set of candidate task fragments is generated.

[0087] The criteria for selecting task segments are as follows:

[0088] when and If there is an intersection, it is determined that the task fragment selection criteria are met, and the task fragment is recorded. Add it to the task fragment candidate set for the next step of spatial conflict determination and temporal conflict determination.

[0089] It should be noted that when and If there is an intersection, then the valid execution window falls into the intersection. The task segment records within a given time period, including tasks currently being executed and those about to be executed (falling within the forward-looking monitoring time window). Task segments that do not participate in the two-dimensional conflict determination include two categories: first, historical task segments that have ended before "current system time - debouncing buffer time", and second, long-term task segments whose start time is later than "current system time + forward-looking duration".

[0090] Then, based on the effective execution window overlap determination of the task fragment candidate set, a set of task fragment pairs requiring two-dimensional conflict determination is generated. Specifically:

[0091] Iterate through the candidate set of task segments and compare any two different task segment records in the candidate set;

[0092] If two task fragment records meet the task fragment pair filtering criteria, they are added to the task fragment pair set.

[0093] After traversing the candidate set of task fragments, the final set of task fragment pairs is generated.

[0094] Among them, the task fragments are the effective execution windows for filtering two task fragments. and There is a time overlap, that is, if If the task fragment pair meets the filtering criteria, the task fragment pair will be added to the task fragment pair set for subsequent spatial and temporal conflict determination.

[0095] To improve the real-time performance and reliability of online monitoring, this embodiment adopts a two-level mechanism: first, filtering based on the forward-looking monitoring time window, and then filtering based on the effective execution window. Firstly, the forward-looking monitoring time window filtering (with a time range from "current system time − debouncing buffer time" to "current system time + forward-looking duration") can exclude completed historical segments and long-term segments in advance, retaining only currently executing and soon-to-be-executed segments as candidates. This enables predictive early warning for the short-term future and suppresses instantaneous jitter and measurement noise through debouncing buffering, significantly reducing computational burden and interface flicker, and improving system responsiveness and stability. Secondly, filtering based on overlapping effective execution window times only pairs segments that may simultaneously occupy resources or space in time for subsequent judgment. This reduces invalid comparisons and false alarms, and, combined with the "actual takes precedence over plan" effective window calculation, improves timing accuracy and judgment convergence. Thus, without changing the original task scheduling logic, it achieves low-latency, visualized, and early warning-enabled two-dimensional conflict monitoring with lower computing power.

[0096] Step S103: Two-dimensional conflict determination, performing time conflict determination and spatial conflict determination for each task segment pair in the task segment pair set.

[0097] In this embodiment, to balance real-time performance and accuracy, both time-based and spatial conflict assessments are performed on each task segment. Time-based conflict assessment addresses resource-exclusive scenarios at the same target workstation, quickly identifying contention risks within the same workstation and time window without involving complex spatial calculations. Spatial conflict assessment, within the two-dimensional simulation parameter space, utilizes safety expansion of angle and radius intervals, along with established two-dimensional angle and radius intersection rules, to identify trajectory overlap risks from different workstations or nearby entrances. These two assessments complement each other. The time dimension reduces invalid alarms and locates resource bottlenecks, while the spatial dimension covers trajectory interference and suppresses measurement errors through safety expansion, thereby achieving low-latency, stable convergence online monitoring under low-computing-power conditions.

[0098] The time conflict determination is based on the following condition: if the target workstations of the two task segments in the task segment pair are the same and the duration of the overlap of their effective execution windows is longer than the global time safety interval, then the task segment pair is determined to have a time conflict; otherwise, it is not determined to be a time conflict.

[0099] Time conflict determination includes:

[0100] Determine the task fragment pairs respectively Target workstation;

[0101] When task fragments If the target workstations are the same, then determine whether the duration of time overlap of the effective execution windows of the task fragment pair is greater than the global time safety interval. ;

[0102] When the duration of the time overlap of the effective execution windows of the task fragment pair is greater than the global time safety interval. If so, it is determined that there is a time conflict between the task segments.

[0103] If the task segment is determined to be time-conflicted, a time conflict record is generated. The structure of the time conflict record is as follows:

[0104] ,

[0105] in, , , for task fragments Index information;

[0106] , For task fragments Action primitive types;

[0107] For task fragments The time overlap of the effective execution window;

[0108] This is the global time safety interval.

[0109] The spatial conflict determination is performed in a two-dimensional simulation parameter space. Based on the global angle safety interval, the global radius safety interval, and the operating arm buffer width, the angle interval and the radius interval are safely expanded to obtain a first normalized angle interval, a second normalized angle interval, a first radius interval, and a second radius interval. When the angle interval satisfies the two-dimensional angle intersection determination rule and the radius interval satisfies the two-dimensional radius intersection determination rule, the task segment is determined to have a spatial conflict.

[0110] It should be noted that for spatial conflict determination of fixed 180° opposing dual manipulators, the nominal orientations of the two manipulators differ by approximately 180° at any given time. This embodiment still employs a two-dimensional conflict determination mechanism based on two-dimensional angle intersection and two-dimensional radius intersection, requiring no additional branch processing. When there is time overlap within the effective time window of a task segment pair, a spatial conflict is determined only when both angle and radius intersections of the task segments exist simultaneously. In practice, if the angle intervals of each task segment (already calculated according to global angle safety intervals) are... If the angle of safety expansion is significantly less than 180° and the movements are confined to their respective radius ranges, angular intersections generally do not occur. Even if the radius ranges intersect at close working radii, spatial conflicts will not be triggered. Spatial conflicts may only be identified in the following situations: for example, the angle range span is large (including covering more than a semicircle after safety expansion), the time window for task segment division is too wide, or... The setting is too large, causing the angle intervals of the fixed 180° opposed dual manipulators to intersect within the effective execution window, and the radius intervals also intersect. Therefore, it can be seen that this method does not have systematic false alarms under the fixed 180° opposed dual manipulator, and is also applicable to non-fixed 180° dual manipulators and multi-arm equipment, without the need to adjust the judgment process.

[0111] Step S104, Conflict Classification and Early Warning: Based on the judgment result of the two-dimensional conflict determination, the conflict level of the task segments with two-dimensional conflicts is determined, and the corresponding operating arm is highlighted in the two-dimensional simulation interface, the target workstation involved is marked and flashed, and the conflict level prompt and arrival time or in progress prompt are displayed.

[0112] In this embodiment, when a two-dimensional conflict is determined to exist between a certain task segment, the conflict is classified and visualized according to the following rules: Conflict Level 1 (only temporal conflict exists, no spatial conflict), Conflict Level 2 (only spatial conflict exists, no temporal conflict), and Conflict Level 3 (both temporal and spatial conflicts exist). The interface layer highlights the corresponding operating arm, marks and flashes the relevant target workstation, and displays the conflict level and arrival / in-progress prompts (arrival time can be calculated from the starting point of the overlapping interval and the current system time).

[0113] Figure 2 This is a schematic diagram illustrating the data modeling process according to an example embodiment of this application. For example... Figure 2 As shown, the data modeling method provided in this embodiment includes:

[0114] Step S1011: Collect multi-source data during the operation of semiconductor equipment.

[0115] In this embodiment, the collected multi-source data includes static configuration data and dynamic operational data, wherein:

[0116] Static configuration data is pre-configured by the device configuration database and loaded at runtime based on index lookup tables, including:

[0117] Arm Buffer Width: This describes the physical width and safety margin that the manipulator needs to reserve during its spatial movement to avoid interference risks caused by errors in the actual size and trajectory of the manipulator.

[0118] Global safety threshold parameter (SafetyThresholdPara):

[0119] Global angle safety interval The minimum angle difference that needs to be reserved when determining whether the two arms have conflicting angles.

[0120] Global radius safety interval The minimum radius difference that needs to be reserved when determining whether the two arms have conflicting radii.

[0121] Global time safety interval The minimum non-overlapping duration that must be satisfied when determining whether two time windows conflict.

[0122] Dynamic operating data is provided by the control system and sensors of the semiconductor equipment during operation, including:

[0123] Task data (TaskHeader):

[0124] Task ID is a unique identifier for a task, used to globally track the execution status of the task.

[0125] Arm ID (ArmID): The identifier of the arm to which the task belongs.

[0126] Default target workstation (DefaultStationID): The default target workstation for the task. This can be filled in for single-site tasks; otherwise, it can be left blank and specified by the primitive record.

[0127] Kinematic Envelope Data:

[0128] Rotation start and end angles This indicates the range of rotational movement of the manipulator within this segment, and is bound to (TaskID, ArmID, SequenceNo). The initial value is the default value, and the rotation start and end angles are updated in real time when the task runs. .

[0129] Expansion / retraction start and end radius This indicates the range of extension and retraction of the manipulator within this segment. It is bound to (TaskID, ArmID, SequenceNo) and initialized with a default value. The extension and retraction start and end radii are updated in real time as the task runs. .

[0130] PrimitiveTimeline: This records the timing information of each primitive in execution order, including the planned start time, planned end time, actual start time, and actual end time for each primitive (SegmentType) in each task. The PrimitiveTimeline is a collection of implementations of the primitives.

[0131] ,

[0132] in, Record the task fragment for action primitive i.

[0133] The structure of action primitive records is as follows:

[0134] ;

[0135] in,

[0136] TaskID is a task identifier used to associate with task data (TaskHeader);

[0137] ArmID is the operator arm identifier, used to associate with task data (TaskHeader);

[0138] SequenceNo is the execution order number, which is the execution order number corresponding to an action primitive in the action primitive type (SegmentType). Under the same (TaskID, ArmID), SequenceNo increments from 1 and is unique.

[0139] SegmentType is the action primitive type. Based on this type, each task is broken down into multiple sub-tasks, including Rotate, Extend, Dock, Retract, and Handover. ;

[0140] PlannedStartTime is the planned start time, which is the planned start time corresponding to an action primitive in the SegmentType.

[0141] PlannedEndTime is the planned end time, which is the planned end time corresponding to an action primitive in the SegmentType.

[0142] ActualStartTime is the actual start time, which is the actual start execution time of an action primitive in the SegmentType.

[0143] ActualEndTime is the actual end time, which is the actual execution end time of an action primitive in the SegmentType.

[0144] StationRef is the target workstation corresponding to this action primitive.

[0145] Step S1012: Based on the action primitive sequence, generate a unified valid execution window for each action primitive record.

[0146] In this embodiment, after obtaining the action primitive timeline data, an effective execution window is generated for each action primitive record in the action primitive timeline. The effective execution window includes an effective start time (EffectiveStartTime) and an effective end time (EffectiveEndTime). The effective start time (EffectiveStartTime) and effective end time (EffectiveEndTime) are determined according to the rule of "actual takes precedence over planned." That is, when the actual start time (ActualStartTime) and the actual end time (ActualEndTime) are available, they are taken as the effective start time (EffectiveStartTime) and effective end time (EffectiveEndTime), respectively; otherwise, the effective start time (EffectiveStartTime) and effective end time (EffectiveEndTime) are taken from the corresponding planned start time (PlannedStartTime) and planned end time (PlannedEndTime). Specifically:

[0147] If the actual start time in the action primitive timeline is not empty, then the effective start time is the actual start time; otherwise, the effective start time is the planned start time.

[0148] If the actual end time in the action primitive timeline is not empty, then the effective end time is the actual end time; otherwise, the effective end time is the planned end time.

[0149] The effective execution window is the difference between the effective end time (EffectiveEndTime) and the effective start time (EffectiveStartTime), which can be represented as:

[0150] ,

[0151] in, .

[0152] Step S1013: Based on the effective execution window, construct a task fragment for each action primitive record in the action primitive sequence to form a task fragment set.

[0153] In this embodiment, it specifically includes:

[0154] For each action primitive record in PrimitiveTimeline Construct corresponding task fragments respectively. This forms a set of task fragments. The union key (TaskID, ArmID, SequenceNo) and Maintain consistency.

[0155] And maintain the same order within the same (TaskID, ArmID) field, that is, increment with SequenceNo.

[0156] Will The effective start time and effective end time are used as task segments. Binding to time boundaries. Represented as:

[0157] ;

[0158] .

[0159] Will Write the action primitive type into the task fragment:

[0160] ;

[0161] .

[0162] Within the active execution window, read or calculate the motion trajectory data (KinematicEnvelopeData) parameters for this segment from the motion trajectory data (KinematicEnvelopeData):

[0163] First, obtain the valid start time and valid end time of the i-th task segment, that is:

[0164] , .

[0165] Then, obtain the rotation start and end angles and the scaling start and end radii within the valid execution window, i.e.:

[0166] ;

[0167] .

[0168] Add the workstation configuration data to the corresponding task segment, specifically:

[0169]

[0170] The global security threshold parameters and the unique indexes (TaskID, ArmID, SequenceNo) of the task fragments are added to the corresponding task segments to complete the construction of the task fragments. The set of task fragments is as follows:

[0171] .

[0172] The structure of each task fragment record in the task fragment set is as follows:

[0173]

[0174] in,

[0175] This is the i-th task segment;

[0176] TaskID is the task fragment TS i Task identifier;

[0177] ArmID is a mission fragment TS i Operating arm markings;

[0178] SequenceNo is a task fragment. In the action primitive type (SegmentType), the execution order number corresponding to an action primitive is SequenceNo. Under the same (TaskID, ArmID), SequenceNo is incremented from 1 and is unique.

[0179] SegmentType is a task segment (TS). i The types of action primitives include Rotate, Extend, Dock, Retract, and Handover. ;

[0180] and , for task fragments rotation start and end angles and the start and end radius of the extension / retraction ;

[0181] StationRef is the target workstation corresponding to this action primitive.

[0182] EffectiveStartTime and EffectiveEndTime are task segments. The valid start time and valid end time;

[0183] , and , for task fragments The global security threshold parameter;

[0184] ArmBuffer is a task fragment. The operating arm buffer width.

[0185] Figure 3 This is a schematic diagram illustrating the spatial conflict determination process according to an example embodiment of this application. Figure 3 As shown, the spatial conflict determination method provided in this embodiment includes:

[0186] Step S1031, Angle Interval Calculation and Safety Expansion: For each pair of task segments, calculate the lower bound and upper bound of the angle interval for the two task segments respectively to obtain the first angle interval and the second angle interval, and perform safety expansion according to the global angle safety interval.

[0187] In this embodiment, firstly, the task fragment pairs are calculated. The lower and upper bounds of the angle intervals are determined, and the first and second angle intervals are defined.

[0188] ,

[0189] ,

[0190] but ,

[0191] in, For task fragments Lower bound of the angle interval; For task fragments Upper bound of the angle interval; AngleRange i This is the first angle interval.

[0192] ,

[0193] ,

[0194] but ,

[0195] in, For task fragments Lower bound of the angle interval; For task fragments Upper bound of the angle interval; This is the second angle interval.

[0196] It should be noted that the above calculation is performed by subtracting the global angle safety interval from the lower bound of the angle interval. Add a global angle safety interval to the upper bound of the angle range. This achieves safety expansion to cover manufacturing tolerances, sensor and encoder quantization errors, and trajectory fitting deviations, reducing the risk of missed detections due to boundary fitting.

[0197] Step S1032, Radius Interval Calculation and Safety Expansion: For each task segment pair, calculate the lower and upper bounds of the radius intervals of the two task segments respectively, and expand them according to the global radius safety interval and the operating arm buffer width to obtain the first radius interval and the second radius interval.

[0198] Compute task fragment pairs The lower and upper bounds of the radius intervals are determined, and the first and second radius intervals are identified.

[0199] The first radius interval is:

[0200] ,

[0201] ,

[0202] but ,

[0203] in, This is the lower bound of the radius interval of the first radius interval; This is the upper bound of the first radius interval; This is the first radius interval.

[0204] The second radius interval is:

[0205] ,

[0206] ,

[0207] but ,

[0208] in, This is the lower bound of the second radius interval; This is the upper bound of the second radius interval; This is the second radius interval.

[0209] It should be noted that the above calculation is performed by subtracting the lower bound of the radius interval. Add to the upper bound of the radius interval This achieves safety expansion, simultaneously covering the envelope thickening caused by measurement, encoder errors, trajectory fitting deviations, and the thickness and attitude swing of the manipulator, thereby reducing the risk of missed detections and false alarms caused by critical edge contact.

[0210] Step S1033: Angle normalization, normalize the first angle interval and the second angle interval to... This forms the first normalized angle interval and the second normalized angle interval.

[0211] In this embodiment, the task fragments are paired Rotation start and end angle range and Normalization to interval, when or When the boundary spans 0° or 360°, or It needs to be split into two sub-intervals to determine whether it intersects with the other. That is:

[0212] The first angle interval Normalization processing, and Normalization to Interval:

[0213] if Then the first normalized angle interval covers the entire circle, and the first normalized angle interval is... ;

[0214] if and Then the first normalized angle interval is one that does not cross zero. ;

[0215] if and Then the first normalized angle interval is a zero-crossing interval, and the first normalized angle interval is divided into the first sub-interval. and the second subinterval ,Right now .

[0216] The second angle interval Normalization processing, and Normalization to Interval:

[0217] if Then the second normalized angle interval covers the entire circle, and the second normalized angle interval is... ;

[0218] if and Then the second normalized angle interval is one that does not cross zero. ;

[0219] if and Then the second normalized angle interval crosses zero, and the second normalized angle interval is divided into the first sub-interval. and the second subinterval ,Right now, .

[0220] Step S1034: Two-dimensional intersection determination. Based on the two-dimensional angle intersection determination rules and the two-dimensional radius intersection determination rules, two-dimensional intersection determination is performed on the first normalized angle interval, the second normalized angle interval, the first radius interval, and the second radius interval, respectively.

[0221] In this embodiment, the two-dimensional angle intersection determination rule is as follows:

[0222] If either the first normalized angle interval or the second normalized angle interval covers a complete circle, then the task segment is determined to have a two-dimensional angle intersection.

[0223] When both the first and second normalized angle intervals do not cross zero, if the angle interval closure intersection condition is met, it is determined that the task segment pair has a two-dimensional angle intersection.

[0224] When the first normalized angle interval crosses zero and the second normalized angle interval does not cross zero, the first sub-interval and the second sub-interval of the first normalized angle interval are compared with the second normalized angle interval twice. If any comparison satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

[0225] When the first normalized angle interval does not cross zero and the second normalized angle interval crosses zero, the first sub-interval and the second sub-interval of the second normalized angle interval are compared with the first normalized angle interval twice. If either comparison satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

[0226] When both the first normalized angle interval and the second normalized angle interval cross zero, the first and second sub-intervals of the first normalized angle interval are compared one by one with the first and second sub-intervals of the second normalized angle interval. When any pair of sub-intervals satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

[0227] The two-dimensional angle intervals are closed and intersecting under the following conditions: the lower bound of the first normalized angle interval is not greater than the upper bound of the second normalized angle interval, and the lower bound of the second normalized angle interval is not greater than the upper bound of the first normalized angle interval.

[0228] The rule for determining the intersection of two-dimensional radii is as follows:

[0229] When the lower bound of the first radius interval is not greater than the upper bound of the second radius interval, and the lower bound of the second radius interval is not greater than the upper bound of the first radius interval, it is determined that the task segment pair has a two-dimensional radius intersection.

[0230] Step S1035: Spatial conflict determination. Spatial conflict determination is performed based on the results of the two-dimensional intersection determination.

[0231] In this embodiment, when a pair of task segments intersects in both two-dimensional angles and two-dimensional radii, then the pair of task segments has a spatial conflict.

[0232] When a spatial conflict exists between task segments, a spatial conflict record is generated. The structure of this record is as follows:

[0233]

[0234] in, , , for task fragments Index information;

[0235] , For task fragments Action primitive types;

[0236] For task fragments The intersection of the effective execution windows;

[0237] This is a snapshot of the interval values ​​used in the determination.

[0238] , This is a snapshot of the global security threshold parameters;

[0239] and , for task fragments The operating arm buffer width.

[0240] Figure 4 This is a schematic diagram illustrating the conflict classification and early warning process according to an example embodiment of this application. Figure 4 As shown, the conflict classification and early warning method provided in this embodiment includes:

[0241] Step S1041: Based on the results of the time conflict and space conflict determination, determine the conflict level of the task segment pairs.

[0242] In this embodiment, the conflict level determination includes:

[0243] When there is only a time conflict between task fragments, the conflict level is level one and the conflict type is resource conflict.

[0244] When a task fragment pair has only spatial conflicts, the conflict level is level two and the conflict type is spatial conflict.

[0245] When there are temporal and spatial conflicts between task segments, the conflict level is level three and the conflict type is a compound conflict.

[0246] Step S1042: When it is determined that there is a conflict between task segments, a warning time window is generated.

[0247] In this embodiment, when a two-dimensional conflict exists between task segments, the overlap time of the effective execution windows of the task segments is obtained as the occurrence time window of the two-dimensional conflict. The difference between its starting point and the current system time Now can be used as the arrival time (for visualization). If its starting point is earlier than the current system time Now, it indicates that the two-dimensional conflict is "in progress".

[0248] Step S1043: Provide a visual warning output in the two-dimensional simulation screen.

[0249] In this embodiment, when a conflict is determined to exist between task segments, a visual warning is output in the two-dimensional simulation screen, including:

[0250] Highlight the corresponding operating arm;

[0251] The target workstations involved (target workstations of task fragment pairs) are marked and flashed;

[0252] Display the conflict level (Level 1, 2, or 3) and arrival time or "In Progress" message near the conflict object or in the sidebar;

[0253] Overlaying snapshots of key intervals used for determination and the time window in which the conflict occurred assists in localization.

[0254] The two-dimensional conflict detection in this embodiment is only used for monitoring and early warning. Its output will not change the existing task queue or path, and it does not participate in task insertion, reordering, or path optimization, thus strictly distinguishing it from the "dynamic task scheduling" control layer.

[0255] Figure 5 This is a schematic diagram illustrating the structure of a visual monitoring system for semiconductor devices according to an example embodiment of this application. Figure 5 As shown, the semiconductor device visualization monitoring system 500 provided in this embodiment includes: a data modeling module 510, a task fragment pair generation module 520, a two-dimensional conflict determination module 530, and a conflict classification and early warning module 540.

[0256] The data modeling module 510 collects multi-source data during the operation of semiconductor equipment, breaks down tasks into multiple sub-tasks based on action primitive types, calculates the effective execution window of each action primitive, and constructs task fragments accordingly to form a set of task fragments.

[0257] The task fragment pair generation module 520 initially selects task fragments that are being executed or have not yet been executed from the task fragment set to form a task fragment candidate set. It compares any two task fragments in the task fragment candidate set, filters out task fragment pairs whose effective execution windows have time overlap, and generates a task fragment pair set.

[0258] The two-dimensional conflict determination module 530 performs temporal conflict determination and spatial conflict determination for each task segment pair in the task segment pair set, wherein:

[0259] The time conflict is determined when the target workstations of the two task segments in the task segment pair are the same, and the duration of the time overlap of their effective execution windows is greater than the global time safety interval.

[0260] The spatial conflict determination is performed in a two-dimensional simulation parameter space. Based on the global angle safety interval, the global radius safety interval, and the operating arm buffer width, the angle interval and the radius interval are safely expanded to obtain a first normalized angle interval, a second normalized angle interval, a first radius interval, and a second radius interval. When the angle interval satisfies the two-dimensional angle intersection determination rule and the radius interval satisfies the two-dimensional radius intersection determination rule, the task segment is determined to have a spatial conflict.

[0261] The conflict classification and early warning module 540, based on the judgment result of two-dimensional conflict determination, determines the conflict level of task segments with two-dimensional conflicts, highlights the corresponding operating arm in the two-dimensional simulation interface, marks the target workstation involved with flashing, and displays the conflict level prompt and arrival time or in progress prompt.

[0262] Figure 6 This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application. For example... Figure 6 As shown, the electronic device 600 provided in this embodiment includes: a processor 601 and a memory 602; wherein:

[0263] Memory 602 is used to store computer programs, and the memory may also be flash memory.

[0264] Processor 601 is used to execute the execution instructions stored in the memory to implement the various steps in the above method. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0265] Alternatively, the memory 602 can be either standalone or integrated with the processor 601.

[0266] When the memory 602 is a device independent of the processor 601, the electronic device 600 may further include:

[0267] Bus 603 is used to connect the memory 602 and the processor 601.

[0268] This embodiment also provides a readable storage medium storing a computer program, which, when executed by at least one processor of an electronic device, enables the electronic device to perform the methods provided in the various embodiments described above.

[0269] This embodiment also provides a program product including a computer program stored in a readable storage medium. At least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the methods provided in the various embodiments described above.

[0270] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0271] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for visual monitoring of semiconductor devices, characterized in that, include: Data modeling involves collecting multi-source data during the operation of semiconductor equipment, breaking down tasks into multiple sub-tasks based on action primitive types, calculating the effective execution window for each action primitive, and constructing task fragments accordingly to form a set of task fragments. A task fragment pair set is generated. In the task fragment set, task fragments that are being executed or have not yet been executed are initially selected to form a task fragment candidate set. Any two task fragments in the task fragment candidate set are compared, and task fragment pairs with overlapping effective execution windows are selected to generate a task fragment pair set. Two-dimensional conflict determination involves performing temporal and spatial conflict determinations for each task segment pair in the task segment pair set, wherein: The time conflict is determined when the target workstations of the two task segments in the task segment pair are the same, and the duration of the time overlap of their effective execution windows is greater than the global time safety interval. The spatial conflict determination is performed in a two-dimensional simulation parameter space. Based on the global angle safety interval, the global radius safety interval, and the operating arm buffer width, the angle interval and the radius interval are safely expanded to obtain a first normalized angle interval, a second normalized angle interval, a first radius interval, and a second radius interval. When the angle interval satisfies the two-dimensional angle intersection determination rule and the radius interval satisfies the two-dimensional radius intersection determination rule, the task segment is determined to have a spatial conflict.

2. The method for visual monitoring of semiconductor devices according to claim 1, characterized in that, The task segments that are being executed or have not yet been executed are determined by the forward monitoring time window. When the effective execution window of the task segment overlaps with the forward monitoring time window, the task segment is included in the task segment candidate set. The forward-looking monitoring time window is based on the current system time and consists of the de-jitter buffer time and the forward-looking duration. The lower bound of the forward-looking monitoring time window is the current system time minus the de-jitter buffer time, and the upper bound of the forward-looking monitoring time window is the current system time plus the forward-looking duration.

3. The method for visual monitoring of semiconductor devices according to claim 1, characterized in that, The spatial conflict determination includes: Angle interval calculation and safety expansion: For each pair of task segments, calculate the lower bound and upper bound of the angle interval of the two task segments respectively to obtain the first angle interval and the second angle interval, and perform safety expansion according to the global angle safety interval. Radius interval calculation and safety expansion: For each pair of task segments, the lower bound and upper bound of the radius interval of the two task segments are calculated respectively, and the expansion is performed according to the global radius safety interval and the operating arm buffer width to obtain the first radius interval and the second radius interval. Angle normalization: Normalize the first angle interval and the second angle interval to... This forms the first normalized angle interval and the second normalized angle interval; Two-dimensional intersection determination is performed based on the two-dimensional angle intersection determination rules and the two-dimensional radius intersection determination rules, respectively, for the first normalized angle interval, the second normalized angle interval, the first radius interval, and the second radius interval. Spatial conflict determination is based on the results of two-dimensional intersection determination.

4. The method for visual monitoring of semiconductor devices according to claim 1, characterized in that, The two-dimensional angle intervals are closed and intersecting under the following conditions: the lower bound of the first normalized angle interval is not greater than the upper bound of the second normalized angle interval, and the lower bound of the second normalized angle interval is not greater than the upper bound of the first normalized angle interval.

5. The method for visual monitoring of semiconductor devices according to claim 1, characterized in that, The two-dimensional angle intersection determination rule is as follows: If either the first normalized angle interval or the second normalized angle interval covers a complete circle, then the task segment is determined to have a two-dimensional angle intersection. When both the first normalized angle interval and the second normalized angle interval do not cross zero, if the angle interval closure intersection condition is met, it is determined that the task segment pair has a two-dimensional angle intersection. When the first normalized angle interval crosses zero and the second normalized angle interval does not cross zero, the first sub-interval and the second sub-interval of the first normalized angle interval are compared with the second normalized angle interval twice. If any comparison satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection. When the first normalized angle interval does not cross zero and the second normalized angle interval crosses zero, the first sub-interval and the second sub-interval of the second normalized angle interval are compared with the first normalized angle interval twice. If either comparison satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection. When both the first normalized angle interval and the second normalized angle interval cross zero, the first and second sub-intervals of the first normalized angle interval are compared one by one with the first and second sub-intervals of the second normalized angle interval. When any pair of sub-intervals satisfies the angle interval closure intersection condition, it is determined that the task segment pair has a two-dimensional angle intersection.

6. The method for visual monitoring of semiconductor devices according to claim 1, characterized in that, The two-dimensional radius intersection determination rule is that when the lower bound of the first radius interval is not greater than the upper bound of the second radius interval, and the lower bound of the second radius interval is not greater than the upper bound of the first radius interval, the task segment is determined to have a two-dimensional radius intersection.

7. The method for visual monitoring of semiconductor devices according to claim 1, characterized in that, The visualization monitoring method also includes: Conflict classification and early warning: Based on the judgment results of two-dimensional conflict determination, the conflict level of task segments with two-dimensional conflict is determined, and the corresponding operating arm is highlighted in the two-dimensional simulation interface, the target workstation involved is marked and flashed, and the conflict level prompt and arrival time or in progress prompt are displayed.

8. A visual monitoring system for semiconductor equipment, characterized in that, include: The data modeling module collects multi-source data during the operation of semiconductor equipment, breaks down tasks into multiple sub-tasks based on action primitive types, calculates the effective execution window for each action primitive, and constructs task fragments accordingly to form a set of task fragments. The task fragment pair generation module initially selects task fragments that are being executed or have not yet been executed from the task fragment set to form a task fragment candidate set. It compares any two task fragments in the task fragment candidate set, filters out task fragment pairs whose effective execution windows have time overlap, and generates a task fragment pair set. The two-dimensional conflict determination module performs temporal conflict determination and spatial conflict determination for each task segment pair in the task segment pair set, wherein: The time conflict is determined when the target workstations of the two task segments in the task segment pair are the same, and the duration of the time overlap of their effective execution windows is greater than the global time safety interval. The spatial conflict determination is based on the global angle safety interval, the global radius safety interval, and the operating arm buffer width in the two-dimensional simulation parameter space. The angle interval and the radius interval are safely expanded to obtain the first normalized angle interval, the second normalized angle interval, the first radius interval, and the second radius interval. When the angle interval satisfies the two-dimensional angle intersection determination rule and the radius interval satisfies the two-dimensional radius intersection determination rule, the task segment is determined to have a spatial conflict. The conflict classification and early warning module, based on the judgment results of two-dimensional conflict determination, determines the conflict level of task segments with two-dimensional conflicts, highlights the corresponding operating arm in the two-dimensional simulation interface, marks and flashes the target workstation involved, and displays the conflict level prompt and arrival time or in progress prompt.

9. An electronic device, characterized in that, include: processor; as well as, Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 7.

Images