Industrial collaborative control method and system based on unified task scheduling time scale

Abstract

The application discloses an industrial collaborative control method and system based on a unified task scheduling time scale, relates to the field of industrial collaborative control, and comprises the following steps: establishing a unified task scheduling time scale, unifying the time base of the execution units of multiple industrial agents, limiting the time resolution to the millisecond level, constraining the starting time sequence of each execution unit through a high-precision time service device, and forming a precise and synchronous task starting point; based on the unified task scheduling time scale, a dynamic priority buffer area is set for each execution unit, and control instruction conflicts generated in the handover stage are temporarily stored in the buffer area. Through the unified task scheduling time scale and the dynamic priority buffer area, the application realizes time synchronization and instruction coordination of multiple execution units in task handover, and ensures continuous and stable control logic. Through the generation of an avoidance pre-occupancy area based on a predicted trajectory and the combination of rolling scheduling optimization, an adaptive steady state of time and space is formed, path intersection and collision are avoided, and long-term safe and efficient operation of the production process is ensured.

Description

Technical Field

[0001] This invention relates to the field of industrial collaborative control technology, and specifically to an industrial collaborative control method and system based on a unified task scheduling time scale. Background Technology

[0002] Industrial collaborative control based on a unified task scheduling timescale refers to the process of systematically verifying the reliability, safety, controllability, and reusability of industrial artificial intelligence models in real industrial scenarios throughout their entire lifecycle. This is achieved through establishing a unified standard system, scientific evaluation methods, and a professional platform. Its core objective is not only to detect performance indicators at the algorithm level, but also to evaluate whether the model can operate safely, stably, and in accordance with engineering specifications in industrial production, equipment operation and maintenance, design simulation, and quality inspection. This process prevents misjudgments, miscontrols, or abnormal outputs in high-risk industrial tasks, ensuring the model possesses verifiable safety and reliability in actual industrial environments. It provides technical support and risk assurance for model registration, application deployment, and standard setting.

[0003] The existing technology has the following shortcomings: In industrial collaborative control based on a unified task scheduling timescale, when multiple industrial agents dynamically hand over tasks within the same production process, there are often temporal differences in model weight updates. Due to the inconsistent update frequencies of each agent during parameter synchronization and task switching phases, anomalies such as instantaneous reversal of instruction priorities can easily occur. At this point, some agents will output control commands based on the old weights, while others will have already generated motion plans based on the new weights, leading to temporary logical errors. If this error occurs during the collaborative operation phase of robotic arms, the trajectories of the robotic arms may intersect or even collide, affecting not only the stability of equipment operation but also potentially causing fixture damage, workpiece displacement, or false triggering of safety protection systems, posing a serious threat to the continuity of the production line and operational safety.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an industrial collaborative control method and system based on a unified task scheduling time scale to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an industrial collaborative control method based on a unified task scheduling time scale, comprising the following steps: Step 1: Establish a unified task scheduling time standard, unify the time base of the execution units of multiple industrial intelligent agents, limit the time resolution to the millisecond level, constrain the start-up sequence of each execution unit through a high-precision time synchronization device, form a precise and synchronized task starting point, and serve as the time reference for task handover among multiple execution units; Step 2: Based on a unified task scheduling time scale, a dynamic priority buffer is set up for each execution unit. Control instruction conflicts generated during the handover phase are temporarily stored in the buffer, and the control instructions are rearranged and output in chronological order to ensure that each execution unit maintains a consistent time base during the handover process and to prevent instantaneous reversal of instruction priorities. Step 3: Based on the rearranged output of the dynamic priority buffer, continuously predict the spatial actions between multiple execution units, generate corresponding avoidance pre-occupancy areas at the control end, and achieve continuous connection of the action rhythm of each execution unit through slight time difference adjustment to maintain the spatial complementarity of cooperative operation. Step 4: When the avoidance pre-occupied area is in a dynamic and stable state, collect real-time action feedback information of each execution unit, establish a time series mapping, compare the deviation between the predicted trajectory and the actual trajectory, and correct the start time of the next task handover based on the deviation data to maintain the synchronous evolution of action rhythm and time base. Step 5: Based on the corrected task start time, perform rolling scheduling optimization on the overall task flow to form an adaptive time steady state during continuous operation, ensuring that multiple execution units maintain a long-term stable operating state during the dynamic collaboration phase and preventing the occurrence of running trajectory intersections or collisions.

[0007] Preferably, the steps for establishing a unified task scheduling time scale include: The timing devices inside the execution units of multiple industrial intelligent agents are initialized. The central timing device sends out a unified reference time signal. All execution units access the reference time signal through physical connections or the clock synchronization channel of industrial Ethernet, and correct the oscillation frequency and timing phase of the internal timing circuit through time pulse signals. Based on the establishment of a unified time base, time constraints are set for the task scheduling logic of each execution unit. The unified task scheduling time scale is used as the sole trigger source for task startup behavior, and each execution unit performs precise startup according to the preset time offset. After setting the time constraints for the startup sequence of each execution unit, establish the timing definition for the task handover window of multiple execution units. The start time of the handover window is precisely aligned with the end time of the previous execution unit's action, and the end time is kept in time continuous with the start time of the next execution unit. Based on the definition of the startup sequence and handover window of each execution unit, a precise and synchronized task starting point is established. The unified task scheduling time scale, execution unit startup sequence and handover window are integrated into a global time frame, and a startup signal is sent through the time synchronization device to achieve the unification of the task starting point of multiple execution units.

[0008] Preferably, while outputting a unified reference time signal, the central time synchronization device periodically emits a reference pulse signal. When each execution unit receives the reference pulse signal, it adjusts its internal timing deviation to keep its internal counter updated synchronously with the reference time signal. This ensures that the clock phase of the execution units remains consistent and that the time of the task start point remains consistent under a unified task scheduling time scale.

[0009] Preferably, the step of setting a dynamic priority buffer for each execution unit based on a unified task scheduling timescale includes: The control command channels of multiple execution units are synchronously configured during the task handover phase. Each execution unit sets the command output time window according to the unified task scheduling time stamp and records the arrival time of the control command through the time stamp, forming a clear time sequence. Establish a dynamic priority buffer, receive control instructions generated during the handover phase under the control of a unified task scheduling time scale, write the control instructions into the corresponding time slice interval, and set priority flags for the control instructions according to the task order and handover window position; The control instructions in the dynamic priority buffer are rearranged and output in chronological order. Based on the time slice trigger signal of the unified task scheduling time scale, the control instructions are output in order from high priority to low priority, so that the instructions between adjacent execution units are smoothly connected in time. The action status of each execution unit is updated based on the rearranged output results of the control instructions. The output time information is fed back to the task scheduling stage of the next execution unit. Each execution unit adjusts its action rhythm according to the unified task scheduling time scale to maintain the consistency of the time base during the task handover process.

[0010] Preferably, when the dynamic priority buffer rearranges the output of control commands, it uses the time slice of the unified task scheduling time scale as the triggering basis. After the output is completed, it records the output time of the control command and feeds the time information back to the task scheduling stage of the next execution unit. Each execution unit adjusts its action rhythm according to the feedback time information to maintain the consistency between the output of control commands and the execution of actions.

[0011] Preferably, the step of continuously predicting spatial actions among multiple execution units based on the rearranged output of the dynamic priority buffer includes: Based on the rearranged output of the dynamic priority buffer, the action sequence of each execution unit under the unified task scheduling time scale is spatially mapped, the time information in the control instructions is mapped one-to-one with the spatial location information of the execution unit, and an association mapping table is established to form a spatial distribution structure. Based on the output order of the dynamic priority buffer, the spatial actions between multiple execution units are continuously predicted, the action trajectories are arranged continuously in time order, the areas where the space is close or the trajectories intersect are identified, and potential conflict intervals are marked on the time axis. Based on the continuous prediction results of spatial actions, an avoidance pre-occupancy zone is generated. The action trajectory of the execution unit is divided into an interaction area in the spatial coordinate system. The spatial boundary, time duration interval and action restriction conditions of the avoidance pre-occupancy zone are defined to prevent action conflicts. Based on the generation results of the avoidance pre-occupancy area, a slight time difference adjustment is made, and the start time of each execution unit action is adjusted at the millisecond level. By correcting the action delay through the time difference, the continuous connection and spatial complementarity of the action rhythm of the execution units are maintained.

[0012] Preferably, the spatial boundary of the avoidance pre-occupancy zone is dynamically adjusted according to the action trajectory of each execution unit, and the time duration interval corresponds precisely to the time slice of the unified task scheduling time scale. The control terminal updates the boundary and time difference adjustment value of the avoidance pre-occupancy zone in each time period, so that the action rhythm of multiple execution units in the continuous operation phase remains synchronized and the spatial complementary relationship is maintained.

[0013] Preferably, the steps for collecting real-time action feedback information of each execution unit when the avoidance pre-occupancy area is in a dynamically stable state include: When the pre-occupied area enters a dynamic stable state, the action feedback signals of each execution unit are collected in real time. Each execution unit outputs state information including execution position, movement speed, attitude angle and action start and end time, and records it to the unified task scheduling time scale through high-precision time stamp. A time series mapping is established based on a unified task scheduling time scale. The feedback data of each execution unit is arranged in chronological order and matched with the predicted trajectory to form a mapping table containing the action time and the time distribution of the avoidance pre-occupancy zone boundary. By comparing the deviations between the predicted trajectory and the actual trajectory based on time series mapping, time deviations and spatial deviations are identified, and the deviation data are recorded in chronological order to form a deviation sequence; Based on the deviation data, the start time of the next task handover is corrected, and the task cycle of each execution unit is adjusted at the millisecond level to ensure that the action rhythm evolves synchronously with the unified task scheduling time scale.

[0014] Preferably, the steps for optimizing the overall task flow through rolling scheduling based on the corrected task start time include: Based on the revised time base, the task flow of the current cycle is reconstructed in time, and the task start time of each execution unit is redistributed to the continuous time interval of the unified task scheduling time scale, and then sequentially allocated to the corresponding time segments to form a coordinated time distribution structure. The overall task flow is scheduled in a rolling manner based on a unified task scheduling time scale. The task execution time of each execution unit is periodically adjusted with the corrected start time as a reference, and the rhythm deviation of the previous cycle is distributed to subsequent tasks in a rolling manner. Based on the optimized time distribution of the rolling schedule, the spatial trajectory of each execution unit is coordinated and adjusted. The action connection interval is recalculated and the action execution order is allocated according to the new task start time, forming a joint coordination relationship between time and space. After completing the rolling coordination of time and space, an adaptive time steady state is formed. Under the continuous control of the unified task scheduling time scale, the control end gradually adjusts the time distribution according to the feedback data, so that the task rhythm tends to be balanced and maintains long-term stable operation.

[0015] An industrial collaborative control system based on a unified task scheduling time scale includes a time scheduling time scale module, a dynamic priority buffer module, a spatial motion prediction module, a time series correction module, and a rolling scheduling optimization module. The time scheduling timing module establishes a unified task scheduling timing, unifies the time base for the execution units of multiple industrial intelligent agents, limits the time resolution to the millisecond level, and constrains the start-up sequence of each execution unit through a high-precision time synchronization device to form a precise and synchronized task starting point. The dynamic priority buffer module, based on a unified task scheduling time scale, sets up a dynamic priority buffer for each execution unit, temporarily stores control instruction conflicts generated during the handover phase in the buffer, and rearranges the control instructions according to the time sequence for output. The spatial motion prediction module continuously predicts the spatial motions between multiple execution units based on the rearranged output of the dynamic priority buffer, and generates corresponding avoidance pre-occupancy zones at the control end. The time series correction module collects real-time action feedback information from each execution unit when the avoidance pre-occupancy area is in a dynamic stable state, establishes a time series mapping, compares the deviation between the predicted trajectory and the actual trajectory, and corrects the start time of the next task handover based on the deviation data. The rolling scheduling optimization module optimizes the overall task flow based on the corrected task start time, forming an adaptive time steady state during continuous operation.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention achieves high-precision time synchronization and instruction sequence coordination among multiple execution units during task handover by establishing a unified task scheduling timescale and dynamic priority buffer. Each execution unit aligns its action start point under a unified time base, and control instructions are smoothly connected after being rearranged in time sequence, avoiding logical errors caused by instantaneous reversal of instruction priorities. The action rhythm of the entire production process remains consistent within a millisecond-level precision range, ensuring control continuity and execution stability during collaborative operations, fundamentally eliminating control anomalies caused by timing differences during task switching.

[0017] This invention generates avoidance pre-occupancy zones based on predicted trajectories and combines this with rolling scheduling optimization, enabling a dynamic coordination between the spatial actions and temporal rhythms of multiple execution units. Action feedback information forms a time series mapping under a unified time reference, and deviation data is used to correct the starting point of the next cycle task, thereby achieving adaptive evolution of the action rhythm. During long-term operation, the execution units reach a stable temporal steady state, maintaining spatial complementarity and rhythmic continuity, preventing path intersections or collisions during the coordination phase, and ensuring the continuous, safe, and stable operation of the production line under high concurrency conditions. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of the industrial collaborative control method based on a unified task scheduling time scale according to the present invention.

[0020] Figure 2 This is a schematic diagram of the modules of the industrial collaborative control system based on a unified task scheduling time scale according to the present invention. Detailed Implementation

[0021] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0022] This invention provides, for example Figure 1 The industrial collaborative control method based on a unified task scheduling time scale, as shown, includes the following steps: Step 1: Establish a unified task scheduling time standard, unify the time base of the execution units of multiple industrial intelligent agents, limit the time resolution to the millisecond level, constrain the start-up sequence of each execution unit through a high-precision time synchronization device, form a precise and synchronized task starting point, and serve as the time reference for task handover among multiple execution units; The specific implementation method for this step is as follows: In industrial production environments, multiple execution units are often distributed across different work areas, with their internal clocks maintained independently by their respective controllers. Therefore, the internal timing devices of each execution unit must first be initialized. During the initialization phase, a central timing device sends a unified reference time signal, which all execution units access via physical connections or an industrial Ethernet clock synchronization channel. The timing device continuously outputs time pulse signals, with each pulse interval being a fixed period. These pulse signals are used to correct the oscillation frequency and timing phase of the internal timing circuits of each execution unit, bringing their timing errors to within milliseconds. To prevent clock drift caused by electromagnetic interference, temperature variations, and network jitter in industrial environments, the timing device outputs a reference pulse in each time period. Upon receiving the reference pulse, the execution unit immediately adjusts its own timing deviation to maintain synchronization between its internal counter and the reference time signal. In this way, the clock phase of each execution unit remains unified, ensuring complete physical synchronization of the time base for the entire production process and guaranteeing accurate execution of time commands in subsequent scheduling phases.

[0023] After establishing a unified time base, time constraints are set for the task scheduling logic of each execution unit. By embedding time triggering conditions into the task startup process of each execution unit, the unified task scheduling timescale becomes the sole trigger source for startup behavior. When the timing device sends a startup pulse signal, each execution unit starts according to its specific position and function in the production process, based on a pre-configured time offset. The time offset is set in milliseconds to distinguish the startup order of adjacent execution units, preventing simultaneous startup from causing instantaneous power superposition or overlapping of robotic arm movement paths. After receiving the unified task scheduling timescale signal, each execution unit first records the current timestamp, and then delays the startup execution command according to its offset parameter. The delay time is precisely counted down by the timing register in the controller. After the timer expires, the execution unit immediately sends an action signal and enters the predetermined running state. In this way, all execution units form a continuous startup sequence at millisecond-level resolution, and the task startup point of each execution unit is strictly limited to the same time base, thereby ensuring the temporal consistency of the task scheduling process.

[0024] After setting the constraints for the startup timing of each execution unit, a timing definition for the task handover window of multiple execution units is established to ensure seamless connection of the handover process on the same time base. Each handover window is allocated a fixed time interval on a unified task scheduling timescale for data transfer or action switching between different execution units. The start time of the handover window is precisely aligned with the end time of the action of the previous execution unit, and the end time is time-continuous with the start time of the next execution unit. Taking the collaborative operation of robotic arms as an example, when the first robotic arm completes the workpiece transport to the designated position, its controller issues an end signal at the same time as the action terminates. This signal is recorded on the millisecond-level coordinates of the unified task scheduling timescale. Subsequently, the second robotic arm starts the assembly operation according to the same time coordinates. The entire process has no gaps or overlaps on the timeline. To ensure the accuracy of the handover time, the timing device continuously outputs high-frequency time pulses, and each execution unit continuously corrects its own action rhythm during the execution phase to align the actual execution time with the center time of the handover window. In this way, the order of operations of different execution units during the handover phase is completely dependent on a unified task scheduling time scale, preventing actions from being performed ahead of schedule or behind schedule, and ensuring that the handover behavior is consistent in time.

[0025] After the startup sequence and handover window of each execution unit are defined, a precisely synchronized task starting point is established as the unified time reference for the entire industrial collaborative control process based on a unified task scheduling timescale. In this stage, the unified task scheduling timescale, the startup sequence of the execution units, and the time interval of the handover window are integrated into a global time frame. The timing device sends a startup signal at zero time of the global time frame, and all execution units simultaneously receive this signal and execute a startup countdown according to their respective time offsets. After the countdown ends, the execution units enter the task running state at a millisecond-level time resolution, and the action rhythm continuously evolves according to the unified task scheduling timescale. When the production process enters a continuous task cycle, the timing device repeatedly outputs a time synchronization signal at a fixed period to ensure that the operating clocks of each execution unit remain consistent throughout long-term operation. The entire process constitutes a closed-loop system from time base establishment, timing constraint setting, handover window definition to task starting point formation. In this way, the action rhythm of multiple execution units during the collaborative operation phase is always based on a unified time base. The start-up, operation, and handover of each execution unit are all constrained to the same millisecond time scale, avoiding priority reversal and control logic disorder caused by time drift in task scheduling, thereby ensuring the operational stability and task execution continuity of the industrial large model during the evaluation process.

[0026] Through the above implementation steps, the unified task scheduling timescale not only achieves millisecond-level synchronization control at the time level but also establishes a highly consistent execution starting point at the task level, providing a stable time reference for the collaborative process of multiple execution units. Throughout the entire industrial collaborative control process based on the unified task scheduling timescale, the time base and action starting point of each execution unit are completely consistent. The task handover process is smoothly connected within a unified time framework, ensuring the logical coherence and physical security of collaborative execution, and providing a solid time foundation for subsequent dynamic priority scheduling and continuous task operation.

[0027] Step 2: Based on a unified task scheduling time scale, a dynamic priority buffer is set up for each execution unit. Control instruction conflicts generated during the handover phase are temporarily stored in the buffer, and the control instructions are rearranged and output in chronological order to ensure that each execution unit maintains a consistent time base during the handover process and to prevent instantaneous reversal of instruction priorities. The specific implementation method for this step is as follows: After the unified task scheduling timescale is established, the control instruction channels of multiple execution units are synchronized during the task handover phase. Before entering the task handover phase, each execution unit sets its own instruction output time window according to the time resolution of the unified task scheduling timescale. This time window uses milliseconds as the smallest unit, strictly limiting the number of instructions allowed to be output within each instant. To ensure that instructions from different execution units can be correctly processed under a unified time base, all execution units synchronously start the buffer's time counter using the unified task scheduling timescale before entering the handover state. When the buffer receives control instructions from different execution units, it records them according to the arrival timestamp, with each instruction accompanied by time base information. Through this time identification mechanism, the control instructions of multiple execution units are explicitly ordered in the time dimension, preventing time misalignment due to network transmission delays or differences in processing speed, thus providing an accurate time base for subsequent instruction conflict temporary storage.

[0028] After synchronizing the instruction channels of each execution unit, a dynamic priority buffer is established and receives control instructions generated during the handover phase under the control of a unified task scheduling timescale. The dynamic priority buffer is divided into multiple consecutive time periods, each corresponding to a millisecond-level time slice in the unified task scheduling timescale. Upon entering the handover phase, each execution unit writes its output control instructions into the corresponding time slice interval in chronological order. The buffer does not immediately output instructions upon receiving them; instead, it prioritizes multiple instructions within the same time slice. Priority marking is determined based on the task order of the execution units under the unified task scheduling timescale and the position of the handover window. For example, in collaborative robotic arm operations, the robotic arm at the start of the handover has a higher priority than the robotic arm in the action waiting state, while the robotic arm at the end of the action has a lower priority than the robotic arm about to execute the next task. Through this dynamic priority allocation method, the buffer establishes a clear execution order for each instruction on a unified time base. During this phase, all handover instructions are temporarily stored in the buffer and do not enter the output phase to ensure that instruction reordering can be completed within a millisecond-level time frame.

[0029] After prioritizing control instructions, the control instructions in the buffer are rearranged and output in chronological order. The rearrangement process uses a unified task scheduling timescale as the dominant timeline. When the timescale pulse reaches a certain time slice, the buffer reads all marked instructions from the corresponding storage area and outputs them sequentially according to priority. The output order considers not only chronological order but also the physical location and task type of the execution unit during the handover process to ensure smooth transition of control instructions between adjacent execution units under a unified time base. During the output process, each instruction is issued precisely at a specific millisecond in the unified task scheduling timescale. Once the previous instruction is output, the buffer immediately begins reading the next time slice, thus forming a continuous output rhythm. Because the execution time of each instruction is referenced to the unified task scheduling timescale, the control behavior of each execution unit remains synchronized during the handover phase. Even with differences in the number of instructions or execution time offsets, there will be no instruction overlap or conflict. Through this chronological rearrangement method, the dynamic priority buffer achieves time-series unification of control instructions across multiple execution units, avoiding priority inversion issues caused by processing delays between different execution units.

[0030] After the control instructions are output in a time-sequenced manner, the action status of each execution unit is updated based on the reordering result to maintain consistency in the time base of the task handover process. After each set of instructions is output, the buffer immediately records the output time of that set of instructions according to the unified task scheduling timescale and feeds back the completion time information to the task scheduling stage of the next execution unit. Upon receiving the time feedback information, the next execution unit uses that time point as the starting reference for its own action, aligning the start of the action with the time of the control instruction output. Throughout the process, the unified task scheduling timescale continuously provides high-precision time pulses, and each execution unit adjusts its own action rhythm according to the pulse signals, ensuring that the output of control instructions and the execution of actual actions remain synchronized. When multiple execution units are in a continuous handover phase, the buffer continuously monitors the changes in the unified task scheduling timescale and updates the priority weights in each time slice, allowing the instruction scheduling of the next cycle to inherit the time sequence relationship of the previous cycle. Through this continuous time base update and instruction reordering mechanism, the dynamic priority buffer maintains a stable time reference throughout the handover process, enabling the control behavior of all execution units to run coherently under a unified time base. This fundamentally prevents the instantaneous reversal of instruction priorities and ensures the consistency of control logic and operational safety of the industrial large-scale model during the multi-execution unit interaction phase.

[0031] Through the above steps, the dynamic priority buffer not only achieves unified scheduling of control commands in the time dimension, but also forms a continuous chain of action outputs at the execution level, ensuring strict time synchronization and logical order among execution units during task handover. Throughout the entire industrial collaborative control process based on a unified task scheduling timescale, the generation, temporary storage, rearrangement, and output of control commands are all constrained by the unified task scheduling timescale, thereby establishing a stable time base environment, preventing priority inversion and command contention during the handover phase, and ensuring the continuous and stable operation of multiple execution units during the collaborative work phase.

[0032] Step 3: Based on the rearranged output of the dynamic priority buffer, continuously predict the spatial actions between multiple execution units, generate corresponding avoidance pre-occupancy areas at the control end, and achieve continuous connection of the action rhythm of each execution unit through slight time difference adjustment to maintain the spatial complementarity of cooperative operation. The specific implementation method for this step is as follows: After the control commands are rearranged and output in the dynamic priority buffer, the action sequence of each execution unit is spatially mapped under the unified task scheduling timescale. The control commands output by each execution unit in the buffer include information such as action type, start time, and duration. The control terminal uses the unified task scheduling timescale to map this time information to the spatial location information of the execution unit in the production line. Taking collaborative robotic arm operation as an example, when the first robotic arm completes its workpiece gripping and is about to enter the transfer action, the time window corresponding to its control command is mapped to a specific time interval on the unified task scheduling timescale. The spatial coordinates of this interval are adjacent to the working path area of ​​the second robotic arm. Based on the output of the dynamic priority buffer, the control terminal establishes an association mapping table between this time interval and the spatial coordinates for subsequent action prediction. Through this synchronous mapping of time and space, the action trajectory of each execution unit is continuously arranged on the time axis and forms a non-overlapping basic distribution structure in the spatial coordinates, providing spatial boundary conditions for the subsequent generation of avoidance pre-occupancy areas.

[0033] After completing the time-space mapping, the spatial actions of multiple execution units are continuously predicted based on the output order of the dynamic priority buffer. The control unit decomposes the actions of each execution unit into continuous time segments according to the time resolution of the unified task scheduling timescale, and records the spatial position change trend of the execution unit within each time segment. The spatial action trajectory of each execution unit is continuously arranged according to the rearranged time order in the buffer, forming a spatial path sequence from the current handover point to the next task node. By comparing the spatial path sequences of adjacent execution units, the control unit identifies areas where spatial proximity or trajectory intersection may occur, and marks potential conflict intervals in these areas. Each conflict interval corresponds to a specific time period in the unified task scheduling timescale. Within this time period, the control unit presets a spatial avoidance strategy based on the action prediction results, enabling subsequently generated control commands to adjust their action rhythm in advance, avoiding simultaneous actions by multiple execution units in the same spatial area. Through this continuous spatial action prediction, the control unit can establish time-space coordination relationships between multiple execution units in advance based on the unified task scheduling timescale, thus laying the foundation for the generation of avoidance pre-occupancy zones.

[0034] After continuously predicting spatial movements, the control unit generates corresponding avoidance pre-occupancy zones based on the prediction results. An avoidance pre-occupancy zone is a temporary spatial buffer area reserved within the unified task scheduling timeframe to prevent conflicts in the actions of execution units. Based on the rearranged output of the dynamic priority buffer, the control unit divides the movement trajectories of multiple execution units into several interaction regions in the spatial coordinate system, and defines avoidance pre-occupancy zones within interaction regions where conflicts may occur. Each avoidance pre-occupancy zone comprises three parts: spatial boundaries, a time duration interval, and action restrictions. The spatial boundaries limit the range of motion of the execution unit, the time duration interval corresponds to the corresponding time slice in the unified task scheduling timeframe, and the action restrictions define the types of actions prohibited within that interval. Taking a robotic arm collaborative assembly scenario as an example, when the first robotic arm completes its handling action and enters the pre-placement stage, its avoidance pre-occupancy zone covers the workpiece placement area, while the second robotic arm is confined to outside the avoidance pre-occupancy zone during this time slice to perform preparatory actions, thus avoiding intersection of the two robotic arm trajectories. The core role of the control unit in the whole process is to continuously generate new avoidance pre-occupancy areas according to the time sequence of the buffer zone, and dynamically superimpose these areas onto the timeline of the unified task scheduling time scale, thereby forming an avoidance system that can be dynamically adjusted in both time and space.

[0035] After the avoidance pre-occupancy zone is generated, a micro-time difference adjustment is used to ensure the continuous connection of the action rhythms of each execution unit, maintaining the spatial complementarity of coordinated operation. The control unit makes minute adjustments to the start times of all execution units based on high-precision pulse signals from a unified task scheduling timescale, with the adjustment range maintained within milliseconds. This micro-time difference is used to correct the action delays introduced by the spatial avoidance pre-occupancy zone, ensuring continuous action connection between adjacent execution units. When an execution unit completes its action within the avoidance pre-occupancy zone, its end time is recorded by the unified task scheduling timescale and transmitted to the next execution unit. The next execution unit automatically adjusts its start delay time based on the time difference between this end time and its predetermined start time, thus achieving seamless connection of action rhythms. Through this micro-time difference adjustment, the actions of multiple execution units form a continuous chain in time and a complementary relationship in space, with no action overlap or gaps. Taking the collaborative operation of two robotic arms on a production line as an example, the moment the first robotic arm completes its gripping and exits the avoidance pre-occupancy zone, the second robotic arm immediately enters the work area according to the time difference recorded in the unified task scheduling time scale. Its movement rhythm is perfectly matched with the first robotic arm, and the entire production cycle remains smooth and stable. As the production process continues to run, the control terminal continuously updates the boundary and time difference adjustment value of the avoidance pre-occupancy zone in each time cycle, ensuring that the spatial complementarity and temporal consistency of multiple execution units remain stable during continuous operation.

[0036] Through the above steps, the rearranged output based on the dynamic priority buffer not only achieves the continuity of actions of multiple execution units at the temporal level, but also forms a dynamic avoidance and pre-occupancy relationship at the spatial level. This allows the action rhythm of each execution unit to smoothly transition under the control of a unified task scheduling timescale. Throughout the entire industrial collaborative control process based on a unified task scheduling timescale, control commands, spatial trajectories, and time sequences remain consistent within the same logical framework. This ensures continuous action connection and spatial complementarity during task handover, avoiding collaboration anomalies caused by action delays, path conflicts, or time misalignments. Consequently, it enables stable, safe, and efficient collaborative operation of multiple execution units in complex industrial environments.

[0037] Step 4: When the avoidance pre-occupied area is in a dynamic and stable state, collect real-time action feedback information of each execution unit, establish a time series mapping, compare the deviation between the predicted trajectory and the actual trajectory, and correct the start time of the next task handover based on the deviation data to maintain the synchronous evolution of action rhythm and time base. The specific implementation method for this step is as follows: When the avoidance pre-occupancy zone enters a dynamically stable state after continuous operation by multiple execution units, the action feedback signals of each execution unit are collected in real time. Each execution unit, under the control of a unified task scheduling timescale, outputs action status signals at fixed time intervals, including execution position, speed, attitude angle, and start and end times. The control unit records these feedback signals using high-precision time stamps, ensuring that each data point corresponds to a specific millisecond in the unified task scheduling timescale. The acquisition process, with the avoidance pre-occupancy zone as the boundary, covers the entire action cycle of each execution unit within the transition area. When an execution unit within the avoidance pre-occupancy zone completes its action transition, its action end time and spatial position are simultaneously recorded to describe the temporal and spatial correspondence of the action. Through this precise time stamping and action data acquisition, the control unit can form a high-temporal-resolution action feedback sequence under a unified task scheduling timescale, providing foundational data for subsequent time series mapping.

[0038] After collecting real-time action feedback information, a time series mapping is established based on a unified task scheduling timescale. The time series mapping involves arranging the action feedback data collected by each execution unit within the avoidance pre-occupancy zone in chronological order and matching it with the corresponding predicted trajectory. The control unit establishes a continuous time series mapping table for the feedback data of each execution unit within the avoidance pre-occupancy zone, with time as the main axis. Each row of data represents a time segment, and each column of data corresponds to the spatial action state of the execution unit within that time segment. When multiple execution units have the same or adjacent time points under the unified task scheduling timescale, their corresponding feedback data form associated nodes in the time series mapping. In this way, the control unit constructs a collaborative operation sequence of multiple execution units on the time axis, thus clearly reflecting the action rhythm and spatial relationship of each execution unit under the same time reference. This time series mapping not only includes the action time of each execution unit but also the time distribution of the avoidance pre-occupancy zone boundary, enabling the control unit to accurately identify the offset trend between the action rhythm and the time reference.

[0039] After the time series mapping is established, the deviations between the predicted trajectory and the actual trajectory are compared to identify inconsistencies in the execution actions across time and space. The control unit uses the rearrangement control instructions output from the dynamic priority buffer as the basis for the predicted trajectory. These predicted trajectories contain the expected position and time interval of each execution unit under the unified task scheduling time scale. The actual trajectory is composed of feedback data from the time series mapping established in the previous sub-step. The control unit identifies deviation areas by comparing the spatial position differences between the predicted trajectory and the actual trajectory within the same time segment. When the time coordinate of the deviation differs from the reference time in the unified task scheduling time scale, it is recorded as a time deviation; when the spatial coordinate of the deviation overlaps with the boundary of the avoidance pre-occupancy area, it is recorded as a spatial deviation. To ensure consistency between time and space data, the control unit uses the unified task scheduling time scale as the main time axis during the comparison process, ensuring that all deviations are recorded based on the same time reference. The deviation data is stored in chronological order to form a deviation sequence for subsequent start time correction. This deviation data accurately reflects the dynamic error caused by equipment response lag, motion inertia, or task connection time difference during continuous operation of the execution unit, providing a precise basis for adjusting the task handover time in the next step.

[0040] After the deviation data is generated, the start time of the next task handover is adjusted based on the deviation data to maintain the synchronous evolution of the action rhythm and the time base. The control unit adjusts the time of the next task handover cycle for each execution unit according to the unified task scheduling time scale. For execution units that show early completion of actions in the time deviation sequence, the control unit delays the task start time in their next task cycle; for execution units that show delayed actions, the task start time is advanced. The adjustment range is determined based on the deviation amount of the previous task handover cycle and is limited to millisecond-level resolution to ensure the continuity of time fine-tuning. During the adjustment process, the control unit uses the time boundary of the avoidance pre-occupancy area as a reference to ensure that the adjusted handover start time does not cause spatial overlap. After the adjustment is completed, each execution unit resynchronizes its task start point under the unified task scheduling time scale, thus forming a new coordinated time rhythm. When the entire production process enters the next cycle, all execution units perform task handover based on the corrected start time, and the action rhythm evolves synchronously with the time base. Through this time correction mechanism based on deviation feedback, the control end can dynamically maintain the rhythm consistency between execution units in each cycle, avoid the gradual drift of the action rhythm, and thus ensure the long-term stability of coordinated operation.

[0041] Through the above implementation steps, when the avoidance pre-occupancy area is in a dynamically stable state, real-time action feedback information and a unified task scheduling timescale form a complete correspondence. The time series mapping truly reflects the rhythm changes of each execution unit during collaborative operation. By analyzing the deviation between the predicted trajectory and the actual trajectory, the control end can promptly detect the difference between the action rhythm and the time reference, and achieve synchronous evolution by correcting the start time in the next task cycle. This method, through the synergistic effect of time reference constraints and deviation feedback adjustments, enables multiple execution units to maintain the unity of action rhythm and time reference during long-term continuous operation, achieving temporal consistency and spatial continuity under dynamic handover.

[0042] Step 5: Based on the corrected task start time, perform rolling scheduling optimization on the overall task flow to form an adaptive time steady state during continuous operation, ensuring that multiple execution units maintain a long-term stable operating state during the dynamic collaboration phase and preventing the occurrence of intersection or collision of running trajectories. The specific implementation method for this step is as follows: After correcting the task start time based on deviation feedback in the previous stage, the control unit reconstructs the task flow for the current cycle based on the corrected time reference. The core objective of time reconstruction is to redistribute the task start times of all execution units into continuous time intervals within a unified task scheduling time scale, ensuring that the start time of each execution unit in the new cycle is consistent with the corrected time. In practice, the control unit re-divides the time segments of the task flow using time pulses of the unified task scheduling time scale, with each time segment corresponding to the execution time zone of a task unit. The start times of each execution unit are sequentially allocated to these time segments according to the corrected time order. Taking a multi-robotic arm collaborative assembly scenario as an example, when the corrected start time of the first robotic arm is earlier than the original rhythm, its subsequent tasks will be moved forward accordingly, while the task times of the second and third robotic arms are compressed or expanded sequentially with this as a reference. Through this time reconstruction process, the task start order and time interval of all execution units are reorganized with the corrected time reference as the core, ensuring that each action rhythm forms a continuous and coordinated distribution structure on the time axis, laying the foundation for subsequent rolling scheduling.

[0043] After the task time reconstruction is completed, the overall task flow is rolled over based on a unified task scheduling timescale, ensuring that task allocation and execution rhythm remain balanced in a dynamically changing operating environment. Rolling scheduling is a dynamic process that continuously updates task execution time and handover rhythm during continuous operation. Its core idea is to periodically adjust the time distribution of subsequent tasks for each execution unit, using the current corrected start time as a reference. The control terminal reads the time pulse signal of the unified task scheduling timescale and recalculates the time offset of task execution in each time cycle, distributing the rhythm deviation of the previous cycle to subsequent tasks through rolling, thereby preventing time drift caused by error accumulation. Taking continuous assembly on a production line as an example, when a robotic arm experiences a time delay due to motion inertia in the previous cycle, the control terminal slightly advances its start time during the next cycle's rolling scheduling, while simultaneously adjusting the task cycles of other robotic arms accordingly, maintaining the overall production rhythm. The rolling scheduling process continues, forming a cyclical optimization of the task flow on the time axis, ensuring that the entire task sequence maintains a stable rhythm and predictable time extension during long-term operation.

[0044] After the task flow undergoes rolling scheduling, to prevent overlapping of execution units' spatial movements, the control unit coordinates and adjusts the spatial trajectories of each execution unit according to a unified task scheduling timescale. The optimized time distribution after rolling scheduling directly affects the spatial movement trajectory of the execution units. Therefore, before the start of each new scheduling cycle, the control unit remaps the corrected time information to the spatial paths of the execution units. In this process, the control unit first determines the running order of each execution unit within the avoidance pre-occupancy area, and then recalculates the action connection interval based on the new task start time. For example, in collaborative welding or assembly, after the previous robotic arm completes its operation, its action end time is used as the release time of the avoidance pre-occupancy area, and the task start time of the next robotic arm is adjusted according to the rolling scheduling results to ensure that the two maintain a safe distance in space. When there is partial overlap in the movement paths of multiple execution units, the control unit will allocate action execution rights sequentially according to the time sequence optimized by rolling scheduling, so that the spatial movements form a complementary relationship in time. Through this time-space joint adjustment mechanism, rolling scheduling not only maintains rhythm continuity in the time dimension, but also forms a stable action distribution in the spatial dimension, preventing path intersections or physical collisions caused by task rhythm drift.

[0045] After completing the rolling coordination in time and space, the control unit achieves an adaptive time steady state during continuous operation, ensuring that multiple execution units maintain a stable operating state over the long term during the dynamic collaboration phase. Adaptive time steady state refers to the automatic convergence of the task start time, action rhythm, and handover rhythm of each execution unit under long-term rolling scheduling. Under continuous control with a unified task scheduling timescale, the control unit continuously collects operational feedback data from each execution unit, gradually accumulating the results of time deviation fine-tuning into the time distribution of the next cycle, thus forming a closed-loop system with time self-balancing characteristics for the entire task flow. When equipment status or operating conditions change in the industrial production environment, such as fluctuations in robotic arm load, adjustments to conveyor belt speed, or changes in workpiece size, the rolling scheduling mechanism automatically adjusts the task start time, causing the rhythm deviation to gradually converge over several cycles, eventually restoring a new balanced rhythm. At this point, the action start time of each execution unit is consistent with the release time of the avoidance pre-occupancy area, the connection interval of handover actions remains stable, and the time distribution of the task flow enters an adaptive steady state. Through this continuous time-based rolling and dynamic balance adjustment, industrial production lines avoid spatial interference caused by the gradual drift of task rhythm during long-term operation, achieving continuous coordination between time and action, thereby fundamentally preventing the occurrence of intersections or collisions in operating trajectories.

[0046] Through the above implementation steps, the rolling scheduling optimization based on the corrected task start time not only achieves dynamic self-balancing of the task flow in the time dimension, but also forms a sustainable and collaborative stable structure in the spatial dimension. The rolling scheduling optimization process organically integrates time adjustment, action coordination, and spatial avoidance, enabling multiple execution units to adaptively adjust their own rhythm during continuous operation, maintaining stability and collaboration over long periods, and ensuring that the entire industrial model maintains a highly efficient, safe, and conflict-free long-term operating state under dynamic collaborative conditions.

[0047] This invention achieves high-precision time synchronization and instruction sequence coordination among multiple execution units during task handover by establishing a unified task scheduling timescale and dynamic priority buffer. Each execution unit aligns its action start point under a unified time base, and control instructions are smoothly connected after being rearranged in time sequence, avoiding logical errors caused by instantaneous reversal of instruction priorities. The action rhythm of the entire production process remains consistent within a millisecond-level precision range, ensuring control continuity and execution stability during collaborative operations, fundamentally eliminating control anomalies caused by timing differences during task switching.

[0048] This invention generates avoidance pre-occupancy zones based on predicted trajectories and combines this with rolling scheduling optimization, enabling a dynamic coordination between the spatial actions and temporal rhythms of multiple execution units. Action feedback information forms a time series mapping under a unified time reference, and deviation data is used to correct the starting point of the next cycle task, thereby achieving adaptive evolution of the action rhythm. During long-term operation, the execution units reach a stable temporal steady state, maintaining spatial complementarity and rhythmic continuity, preventing path intersections or collisions during the coordination phase, and ensuring the continuous, safe, and stable operation of the production line under high concurrency conditions.

[0049] This invention provides, for example Figure 2 The industrial collaborative control system based on a unified task scheduling time scale shown includes a time scheduling time scale module, a dynamic priority buffer module, a spatial motion prediction module, a time series correction module, and a rolling scheduling optimization module. The time scheduling timing module establishes a unified task scheduling timing, unifies the time base for the execution units of multiple industrial intelligent agents, limits the time resolution to the millisecond level, and constrains the start-up sequence of each execution unit through a high-precision time synchronization device to form a precise and synchronized task starting point. The dynamic priority buffer module, based on a unified task scheduling time stamp, sets up a dynamic priority buffer for each execution unit, temporarily stores control instruction conflicts generated during the handover phase in the buffer, and rearranges the control instructions for output according to the time sequence. The spatial motion prediction module continuously predicts the spatial motions between multiple execution units based on the rearranged output of the dynamic priority buffer, and generates corresponding avoidance pre-occupancy zones at the control end. The time series correction module collects real-time action feedback information from each execution unit when the avoidance pre-occupancy area is in a dynamic stable state, establishes a time series mapping, compares the deviation between the predicted trajectory and the actual trajectory, and corrects the start time of the next task handover based on the deviation data. The rolling scheduling optimization module optimizes the overall task flow based on the corrected task start time, forming an adaptive time steady state during continuous operation.

[0050] The industrial collaborative control method based on a unified task scheduling time scale provided in this embodiment of the invention is implemented through the aforementioned industrial collaborative control system based on a unified task scheduling time scale. For details of the specific methods and processes of the industrial collaborative control system based on a unified task scheduling time scale, please refer to the embodiments of the industrial collaborative control method based on a unified task scheduling time scale, which will not be repeated here.

[0051] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An industrial collaborative control method based on a unified task scheduling time scale, characterized in that, Includes the following steps: Step 1: Establish a unified task scheduling time standard, unify the time base of the execution units of multiple industrial intelligent agents, limit the time resolution to the millisecond level, and constrain the start-up sequence of each execution unit through a high-precision time synchronization device to form a precise and synchronized task starting point. Step 2: Based on a unified task scheduling time scale, a dynamic priority buffer is set up for each execution unit to temporarily store control instruction conflicts generated during the handover phase in the buffer and rearrange the control instructions according to the time order. Step 3: Based on the rearranged output of the dynamic priority buffer, continuously predict the spatial actions between multiple execution units and generate corresponding avoidance pre-occupancy areas at the control end. Step 4: When the avoidance pre-occupied area is in a dynamic stable state of continuous operation and the action rhythm of each execution unit remains stable, collect the real-time action feedback information of each execution unit, establish a time series mapping, compare the deviation between the predicted trajectory and the actual trajectory, and correct the start time of the next task handover based on the deviation data. Step 5: Based on the corrected task start time, perform rolling scheduling optimization on the overall task flow to form an adaptive time steady state during continuous operation.

2. The industrial collaborative control method based on a unified task scheduling time scale according to claim 1, characterized in that, The steps to establish a unified task scheduling time scale include: The timing devices inside the execution units of multiple industrial intelligent agents are initialized. The central timing device sends out a unified reference time signal. All execution units access the reference time signal through physical connections or the clock synchronization channel of industrial Ethernet, and correct the oscillation frequency and timing phase of the internal timing circuit through time pulse signals. Based on the establishment of a unified time base, time constraints are set for the task scheduling logic of each execution unit. The unified task scheduling time scale is used as the sole trigger source for task startup behavior, and each execution unit performs precise startup according to the preset time offset. After setting the time constraints for the startup sequence of each execution unit, establish the timing definition for the task handover window of multiple execution units. The start time of the handover window is precisely aligned with the end time of the previous execution unit's action, and the end time is kept in time continuous with the start time of the next execution unit. Based on the definition of the startup sequence and handover window of each execution unit, a precise and synchronized task starting point is established. The unified task scheduling time scale, execution unit startup sequence and handover window are integrated into a global time frame, and a startup signal is sent through the time synchronization device to achieve the unification of the task starting point of multiple execution units.

3. The industrial collaborative control method based on a unified task scheduling time scale according to claim 2, characterized in that, While outputting a unified reference time signal, the central time synchronization device periodically emits a reference pulse signal. Each execution unit adjusts its internal timing deviation when receiving the reference pulse signal, so that the internal counter keeps updated in sync with the reference time signal.

4. The industrial collaborative control method based on a unified task scheduling time scale according to claim 2, characterized in that, The steps for setting dynamic priority buffers for each execution unit based on a unified task scheduling timescale include: The control command channels of multiple execution units are synchronously configured during the task handover phase. Each execution unit sets the command output time window according to the unified task scheduling time stamp and records the arrival time of the control command through the time stamp, forming a clear time sequence. Establish a dynamic priority buffer, receive control instructions generated during the handover phase under the control of a unified task scheduling time scale, write the control instructions into the corresponding time slice interval, and set priority flags for the control instructions according to the task order and handover window position; The control commands in the dynamic priority buffer are rearranged and output in chronological order, and the control commands are output sequentially from high priority to low priority according to the time slice trigger signal of the unified task scheduling time scale. The action status of each execution unit is updated based on the rearranged output results of the control instructions, and the output time information is fed back to the task scheduling stage of the next execution unit. Each execution unit adjusts its action rhythm according to the unified task scheduling time scale.

5. The industrial collaborative control method based on a unified task scheduling time scale according to claim 4, characterized in that, When the dynamic priority buffer rearranges the output of control instructions, it uses the time slice of the unified task scheduling time scale as the trigger. After the output is completed, it records the output time of the control instruction and feeds the time information back to the task scheduling stage of the next execution unit. Each execution unit adjusts its action rhythm based on the feedback time information.

6. The industrial collaborative control method based on a unified task scheduling time scale according to claim 4, characterized in that, The steps for continuously predicting spatial actions among multiple execution units based on the rearranged output of the dynamic priority buffer include: Based on the rearranged output of the dynamic priority buffer, the action sequence of each execution unit under the unified task scheduling time scale is spatially mapped, the time information in the control instructions is mapped one-to-one with the spatial location information of the execution unit, and an association mapping table is established to form a spatial distribution structure. Based on the output order of the dynamic priority buffer, the spatial actions between multiple execution units are continuously predicted, the action trajectories are arranged continuously in time order, the areas where the space is close or the trajectories intersect are identified, and potential conflict intervals are marked on the time axis. Based on the continuous prediction results of spatial actions, an avoidance pre-occupancy zone is generated. The action trajectory of the execution unit is divided into an interaction area in the spatial coordinate system, and the spatial boundary, time duration interval and action restriction conditions of the avoidance pre-occupancy zone are defined. Based on the generation results of the avoidance pre-occupancy area, a slight time difference adjustment is made, and the start time of each execution unit action is adjusted at the millisecond level. By correcting the action delay through the time difference, the continuous connection and spatial complementarity of the action rhythm of the execution units are maintained.

7. The industrial collaborative control method based on a unified task scheduling time scale according to claim 6, characterized in that, The spatial boundary of the avoidance pre-occupancy zone is dynamically adjusted according to the action trajectory of each execution unit. The duration interval corresponds precisely to the time slice of the unified task scheduling time scale. The control terminal updates the boundary and time difference adjustment value of the avoidance pre-occupancy zone in each time cycle, so that the action rhythm of multiple execution units in the continuous operation phase remains synchronized and the spatial complementary relationship is maintained.

8. The industrial collaborative control method based on a unified task scheduling time scale according to claim 6, characterized in that, The steps for collecting real-time action feedback information of each execution unit when the avoidance pre-occupied area is in a dynamically stable state include: The steps for collecting real-time action feedback information of each execution unit when the avoidance pre-occupied area is in a dynamically stable state include: When the pre-occupied area is entered into continuous operation and the action rhythm of each execution unit maintains a dynamic and stable state of temporal continuity and spatial coordination under the unified task scheduling time scale, the action feedback signals of each execution unit are collected in real time. Each execution unit outputs state information including execution position, movement speed, attitude angle and action start and end time, and records it to the unified task scheduling time scale through high-precision time stamp. A time series mapping is established based on a unified task scheduling time scale. The feedback data of each execution unit is arranged in chronological order and matched with the predicted trajectory to form a mapping table containing the action time and the time distribution of the avoidance pre-occupancy zone boundary. By comparing the deviations between the predicted trajectory and the actual trajectory based on time series mapping, time deviations and spatial deviations are identified, and the deviation data are recorded in chronological order to form a deviation sequence; Based on the deviation data, the start time of the next task handover is corrected, and the task cycle of each execution unit is adjusted at the millisecond level.

9. The industrial collaborative control method based on a unified task scheduling time scale according to claim 8, characterized in that, The steps for optimizing the overall task flow through rolling scheduling based on the revised task start time include: Based on the revised time base, the task flow of the current cycle is reconstructed in time, and the task start time of each execution unit is redistributed to the continuous time interval of the unified task scheduling time scale, and then sequentially allocated to the corresponding time segments to form a coordinated time distribution structure. The overall task flow is scheduled in a rolling manner based on a unified task scheduling time scale. The task execution time of each execution unit is periodically adjusted with the corrected start time as a reference, and the rhythm deviation of the previous cycle is distributed to subsequent tasks in a rolling manner. Based on the optimized time distribution of the rolling schedule, the spatial trajectory of each execution unit is coordinated and adjusted. The action connection interval is recalculated and the action execution order is allocated according to the new task start time, forming a joint coordination relationship between time and space. After completing the rolling coordination of time and space, an adaptive time steady state is formed. Under the continuous control of the unified task scheduling time scale, the control end gradually adjusts the time distribution based on the feedback data.

10. An industrial collaborative control system based on a unified task scheduling time scale, used to implement the industrial collaborative control method based on a unified task scheduling time scale as described in any one of claims 1-9, characterized in that, It includes a time scheduling time stamp module, a dynamic priority buffer module, a spatial action prediction module, a time series correction module, and a rolling scheduling optimization module; The time scheduling timing module establishes a unified task scheduling timing, unifies the time base for the execution units of multiple industrial intelligent agents, limits the time resolution to the millisecond level, and constrains the start-up sequence of each execution unit through a high-precision time synchronization device to form a precise and synchronized task starting point. The dynamic priority buffer module, based on a unified task scheduling time scale, sets up a dynamic priority buffer for each execution unit, temporarily stores control instruction conflicts generated during the handover phase in the buffer, and rearranges the control instructions according to the time sequence for output. The spatial motion prediction module continuously predicts the spatial motions between multiple execution units based on the rearranged output of the dynamic priority buffer, and generates corresponding avoidance pre-occupancy zones at the control end. The time series correction module collects real-time action feedback information from each execution unit when the avoidance pre-occupancy area is in a dynamic stable state, establishes a time series mapping, compares the deviation between the predicted trajectory and the actual trajectory, and corrects the start time of the next task handover based on the deviation data. The rolling scheduling optimization module optimizes the overall task flow based on the corrected task start time, forming an adaptive time steady state during continuous operation.

Images