A dynamic action priority scheduling and fault-tolerant method for end-side body cerebellum

Abstract

The application discloses a dynamic action priority scheduling and fault-tolerant method and system of an end-side body cerebellum. The method comprises the following steps: collecting an action task urgency parameter U and an end-side device real-time load parameter L; calculating an action priority according to a formula P=aU+b(1-L); executing the action according to the priority; monitoring an abnormality in the execution process, recording an execution step number and state data and pausing the action when an execution fault occurs; and resuming the action execution when the device load is lower than a preset threshold. The application realizes dynamic priority scheduling and breakpoint recovery of the end-side action, and improves system stability.

Description

Technical Field

[0001] This invention relates to the field of embodied intelligent control technology, and in particular to a dynamic action priority scheduling and fault-tolerant method and system for end-sided embodied cerebellum. Background Technology

[0002] Current edge-side action execution typically employs a fixed-priority scheduling mechanism. When device load increases or execution anomalies occur, low-priority actions cannot resume execution, while high-priority actions may be delayed due to insufficient resources, leading to reduced system stability.

[0003] Therefore, a scheduling method with dynamic priority calculation and breakpoint recovery capabilities is needed. Summary of the Invention

[0004] (I) Methodology This invention provides a dynamic action priority scheduling and fault-tolerant method for end-to-end cerebellum, comprising: Step S1: Parameter Acquisition collection: Action task urgency parameter U Real-time load parameters L of end-side equipment in: U is the preset level value; L represents the processor utilization rate or system load value.

[0005] Step S2: Priority Calculation Calculate the action priority P using the following formula: P = aU + b(1-L) in: a and b are preset weighting coefficients; a + b = 1.

[0006] Step S3: Action Sequencing The queue of actions to be executed is sorted according to priority P, with actions of higher priority values ​​being executed first.

[0007] Step S4: Perform monitoring Monitoring during action execution: Did an execution timeout occur? Should an error status be returned? Has an abnormal interruption occurred?

[0008] If any of the conditions are met, the process is considered a failure.

[0009] Step S5: Breakpoint caching When an execution failure occurs: Record the current execution step number; Store completed status data; Mark the current action as paused.

[0010] Step S6: Load assessment and recovery When the load parameter L of the end-side device is lower than the preset threshold T: Read the cached execution progress; Resume the paused action.

[0011] System Solution This invention also provides a lateral embodied cerebellar dynamic scheduling and fault-tolerant system, comprising: Parameter acquisition module; Priority calculation module; Action sorting module; Execute monitoring module; Breakpoint caching module; Resume execution of the module.

[0012] Each module is executed by the edge processor.

[0013] Computer-readable storage media A computer-readable storage medium having a computer program stored thereon, the program implementing the above method steps when executed by a processor. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the end-to-end cerebellar scheduling system of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram of the action breakpoint cache structure of the present invention.

[0015] Figure 1 Detailed description of structural diagram like Figure 1 As shown, the system includes: 100 edge-mounted smart devices Dispatch control module 110 Execution Unit 120 Status monitoring module 130 Breakpoint cache storage module 140 Structural relationship explanation: The action task enters the scheduling control module 110; The scheduling control module 110 calculates the priority; After sorting, the data is sent to execution unit 120; Status monitoring module 130 monitors the execution status; If a fault occurs, the breakpoint cache storage module 140 will be triggered.

[0016] Figure 2 Detailed explanation of the method and process like Figure 2 As shown, the method includes: S201 Collect urgency and load parameters; S202 Calculate action priority; S203 Sort and execute actions; S204 Monitor execution status; S205 Cache execution progress if a failure occurs; S206 Resume execution when the load decreases.

[0017] The process uses a closed-loop control structure.

[0018] Figure 3 Detailed explanation of breakpoint cache structure like Figure 3 As shown, the breakpoint cache structure includes: Action Number Area 301 Execution Step Record Area 302 Status data storage area 303 Pause flag 304 When an error occurs during action execution: Write the current execution step number into the execution step record area 302; Write the current state data to the state data storage area 303; Set the pause flag 304 to 1.

[0019] When the recovery conditions are met: Read the execution step record area 302; Revert to the corresponding step and continue execution; Clear pause flag 304.

Claims

1. A dynamic action priority scheduling and fault-tolerant method for end-to-end cerebellum, characterized in that, Includes the following steps: S1. Collect the urgency parameter U of the action task and the real-time load parameter L of the end device; S2. Calculate the priority P of the action task according to the formula P = aU + b(1-L), where a and b are preset weight coefficients, and a+b=1; S3. Sort the actions to be executed according to the priority P, and execute the actions in priority order; S4. Monitor the execution status during the execution of the action. When an execution timeout, error return, or abnormal interruption is detected, the action is determined to be an execution failure. S5. When an execution failure is detected, record the current execution step number and the completed status data, and mark the action as paused. S6. When the real-time load parameter L of the terminal device is lower than the preset threshold T, read the execution step number and status data, and resume the pause action.

2. The method according to claim 1, characterized in that: The urgency parameter U is a preset level value.

3. The method according to claim 1, characterized in that: The real-time load parameter L is the processor utilization rate or system load value.

4. The method according to claim 1, characterized in that: The execution step number is stored in the local cache storage area.

5. A dynamic action priority scheduling and fault-tolerant system for endosomal cerebellum, characterized in that, include: The parameter acquisition module is used to acquire the urgency parameter U and the real-time load parameter L; The priority calculation module is used to calculate the priority according to the formula P = aU + b(1-L); The action sorting module is used to sort and execute actions according to priority; The execution monitoring module is used to monitor the execution status of actions; The breakpoint caching module is used to record the execution step number and status data when an execution failure occurs; The execution recovery module is used to resume action execution when the load parameter is below the threshold T.

6. A computer-readable storage medium having a computer program stored thereon, the program, when executed by a processor, implementing the method of any one of claims 1-4.

Images