一种基于Agent的任务自生成方法  

A Method of Task Self-generation Based on Agent

作　　者：张可佳[1,2] 姜良涛 ZHANG Kejia;JIANG Liangtao(School of Computer&Information Technology,Northeast Petroleum University,Daqing 163318,China;Key Laboratory of Petroleum Big Data and Intelligent Analysis in Heilongjiang Province,Daqing 163318,China)

机构地区：[1]东北石油大学,计算机与信息技术学院,黑龙江大庆163318 [2]黑龙江省石油大数据与智能分析重点实验室,黑龙江大庆163318

出　　处：《微型电脑应用》2024年第8期15-19,23,共6页Microcomputer Applications

基　　金：国家自然科学基金项目(42172161);黑龙江省创新科研人才培养计划项目(UNPYSCT-2020144);东北石油大学基本科研基金项目(15071202202)。

摘　　要：任务自生成是基于预先设定的业务工作流完成任务执行流程制定的一个过程,任务执行流程制定主要包括任务执行步骤选取与任务执行步骤相应的解决方案选取。提出一种基于Agent的任务自生成方法来解决基于传统的业务工作流的流程制定机制,在面对有差异性需求的不同任务时,因其不能做出任务执行流程制定的自适应性改变而在处理任务时出现处理速度慢、结果不准确等问题。利用Agent的环境感知能力实现对任务需求信息和任务自生成系统环境的感知,利用Agent的自主行为决策能力结合感知的信息实现对任务执行流程制定工作的自主管理,在真实应用场景测试中,通过一系列比较实验验证了这种方法在处理速度和准确度方面具有明显优势。Task self-generation is a process to complete task execution process formulation based on preset business workflow.Task execution process formulation mainly includes task execution step selection and task execution step corresponding solution selection.This paper proposes a task self-generation method based on Agent to solve the process formulation mechanism based on traditional business workflow.When faced on tasks with different requirements,because it can not make adaptive changes to task execution process formulation,problems such as slow processing speed and inaccurate results appear in processing tasks.The Agent’s environment awareness ability is used to realize the awareness of task demand information and task self-generated system environment.The Agent’s autonomous behavior decision-making ability is combined with the perceived information to realize the autonomous management of task execution process development.In the real application scenario test,a series of comparative experiments are designed to verify that this method has obvious advantages in processing speed and accuracy.

关 键 词：任务自生成 AGENT 业务算法库 知识库与推理机 

分 类 号：TP391.7[自动化与计算机技术—计算机应用技术]