Inference task scheduling method for time delay optimization in vehicle-mounted edge computing

A technology of task scheduling and edge computing, which is applied in computing, program control design, electrical digital data processing, etc., and can solve problems such as weak research on single-vehicle reasoning task scheduling

Pending Publication Date: 2021-09-14
FUJIAN NORMAL UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the past scholars' research, the research results of multi-vehicle task coordination scheduling in vehicle edge computing are quite fruitful, but the research on single-vehicle reasoning task scheduling with data dependence is still weak.
Facing the delay-optimized reasoning task scheduling problem in vehicle-mounted edge computing, the current research work has not yet formed a complete and effective solution at the level of single-vehicle reasoning tasks

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Inference task scheduling method for time delay optimization in vehicle-mounted edge computing
  • Inference task scheduling method for time delay optimization in vehicle-mounted edge computing
  • Inference task scheduling method for time delay optimization in vehicle-mounted edge computing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0071] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0072] Such as Figures 1 to 5 As shown in one, the present invention discloses a reasoning task scheduling method for delay optimization in vehicle-mounted edge computing, which includes the following steps:

[0073] Step 1, initialize the experience pool with constant storage space capacity, and the action-value function Q with random weight θ θ (s t , a t ) and the corresponding target-Q θ (s t , a t ), determine the upper limit of rounds and the number of iterations per round;

[0074] Step 2, the action state s t restore to the original state;

[0075] Step 3, choose the action a with the largest historical reward value with the probability of ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an inference task scheduling method for time delay optimization in vehicle-mounted edge computing, and provides a priority judgment method for inference tasks, which is beneficial to efficiently utilizing edge nodes so as to shorten task execution time. In order to avoid the premature convergence problem of a traditional heuristic algorithm for solving the NP-hard problem, a scheduling algorithm based on a DQN algorithm is designed, and a better scheduling strategy in a complex edge environment is explored. According to the method, under the condition that the vehicle-mounted reasoning task is scheduled and executed within the tolerable time, the reasoning task scheduling time delay in the edge environment is optimized.

Description

technical field [0001] The invention relates to an inference task scheduling strategy technology in the field of parallel and distributed high-performance computing, and in particular to a delay-optimized inference task scheduling method in vehicle-mounted edge computing. Background technique [0002] As a key application scenario under the 5G scenario, the Internet of Vehicles has become a research hotspot, and its future development prospects are beyond doubt. The self-driving technology derived from it can not only improve the safety of driving, but also help solve the increasingly serious problems of low traffic efficiency and lane congestion in cities. However, the exploration in the field of autonomous driving also brings severe performance challenges such as shorter execution time requirements. Previously, for autonomous driving, most research focused on how to design specific functions such as traffic condition recognition during vehicle driving as inference tasks. ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F9/48
CPCG06F9/4881G06F2209/484
Inventor 林兵陈乔鑫卢宇林凯
Owner FUJIAN NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap