SIA OpenIR  > 智能检测与装备研究室
面向生产车间的多AGV任务调度与路径规划研究
Alternative TitleResearch on multi-AGV task scheduling and path planning for production workshops
刘生伟
Department智能检测与装备研究室
Thesis Advisor马钺
KeywordA*算法 拍卖算法 死锁与避碰 任务调度 多AGV路径规划
Pages59页
Degree Discipline控制工程
Degree Name专业学位硕士
2020-05-26
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract近年来,随着工业4.0的推进和电商的发展,多AGV任务运输系统越来越多的应用在智能仓储中。多AGV路径规划是保证任务调度和运输的基本条件,所以多AGV路径规划研究是十分重要的。本文研究的主要内容是多AGV任务调度和多AGV路径规划两个方面。根据课题来源,分析了本文研究内容对于课题适应的背景和研究意义,总结了当前路径规划研究算法的不同种类,根据不同算法的种类,说明了各个算法的优缺点。介绍了单AGV路径规划算法,通过研究A*算法的原理和发展现状,改进A*算法的启发函数和搜索方式,并完成了针对A*算法的双向搜索的探究,验证了单路径规划的双向搜索算法的可行性,为多AGV路径规划奠定了基础。其次研究了多AGV任务分配策略。提出一种改进的拍卖法的任务调度模型。通过当前各个AGV的实际工作状态,当前时刻产生任务的实际信息以及当前AGV对应的调度表,对任务进行竞拍,然后把产生的任务插入的相应AGV的调度表中。该部分根据火腿肠生产车间的实际分配状态,探究了拍卖法的实际竞拍规则,快速完成了对多个任务的调度分配。最后,本文将改进拍卖法的任务调度引入多AGV路径规划系统中。针对多AGV路径规划及避免死锁和碰撞问题,本文深入探究了各种碰撞和产生死锁的各种状态,并对各种碰撞和死锁进行分类,针对不同的种类,设定了采用的避碰策略,并提出一种基于碰撞分类和A*算法的占位多AGV路径规划算法,通过实验,证明了该算法的有效性。
Other AbstractIn recent years, with the promotion of industry 4.0 and the development of e-commerce, multi-AGV task transportation systems have been increasingly used in intelligent warehousing. Multi-AGV path planning is the basic condition to ensure task scheduling and transportation, so it is very important to study multi-AGV path planning. The main contents of this paper are multi-AGV task scheduling and multi-AGV path planning. According to the source of the subject, the background and research significance of the research content of this paper for the subject adaptation are analyzed, and the different types of current path planning research algorithms are summarized. According to the types of different algorithms, the advantages and disadvantages of each algorithm are explained. This paper introduces the single AGV path planning algorithm, improves the heuristic function and search mode of A* algorithm by studying the principle and development status of A* algorithm, and completes the research on the bidirectional search of A* algorithm, verifies the feasibility of the bidirectional search algorithm of single path planning, and lays A foundation for multi-AGV path planning. Next, the multi-AGV task assignment strategy is studied. An improved task scheduling model of auction method is proposed. According to the actual working state of each AGV, the actual information of the task generated at the current moment and the scheduling table corresponding to the current AGV, the task is auctioned, and then the generated task is inserted into the scheduling table of the corresponding AGV. According to the actual allocation status of ham sausage production workshop, this part explores the actual bidding rules of auction method, and quickly completes the scheduling and allocation of multiple tasks. Finally, this paper introduces the improved auction task scheduling into the multi- AGC path planning system. For AGV path planning and avoid deadlock and collision problem, this article delves into the state of all sorts of all sorts of collision and deadlocks, and classifying all kinds of collision and deadlock, according to different types, set the collision avoidance strategy of want to use, and put forward A classification and A * algorithm based on collision placeholder multiple AGV path planning algorithm, through the experiment, proved the effectiveness of the proposed algorithm.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/27143
Collection智能检测与装备研究室
Affiliation中国科学院沈阳自动化研究所
Recommended Citation
GB/T 7714
刘生伟. 面向生产车间的多AGV任务调度与路径规划研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2020.
Files in This Item:
File Name/Size DocType Version Access License
面向生产车间的多AGV任务调度与路径规划(3491KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘生伟]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘生伟]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘生伟]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.