摘要
多机器人动态任务分配的研究对于智能仓储系统的高效运行具有重要意义。针对现有研究面临的难以在保证实时性的前提下获得高质量解的问题,提出了一种基于平行单项拍卖的DYPSI算法。首先,引入时间窗策略以增强算法对动态环境的反应能力;其次,提出了一种新的任务分配和路径规划的耦合方法,拍卖投标估值根据迭代情况和路径动态变化;最后,设计了反向拍卖机制加速算法收敛。仿真结果表明,所提出的算法与几种先进算法相比,在性能和效率方面具有竞争力。
The research on multi-robot dynamic task allocation is of great significance for the efficient operation of intelligent warehousing systems.Aiming at the problem that existing research faces difficulty in obtaining highquality solutions while ensuring real-time performance,a DYPSI algorithm based on parallel one-sided auctions is proposed.Firstly,a time window strategy is introduced to enhance the algorithm's adaptability to dynamic environ⁃ments.Secondly,a novel coupling approach between task allocation and path planning is proposed,where auction bid values are adjusted based on iteration progress and dynamic path changes.Lastly,a reverse auction mechanism is de⁃signed to expedite algorithm convergence.The simulation results show that the proposed algorithm is competitive in performance and efficiency compared with several advanced algorithms.
作者
严翠翠
韩志敏
王博
林志赟
YAN Cui-cui;HAN Zhi-min;WANG Bo;LIN Zhi-yun(School of Automation,Hangzhou Dianzi University,Hangzhou Zhejiang 310018,China)
出处
《计算机仿真》
2025年第7期28-33,共6页
Computer Simulation
基金
国家自然科学基金(62173118)
浙江省自然科学基金(LY24F030009)。
关键词
任务分配
路径规划
时间窗策略
反向拍卖
Task allocation
Path planning
Time window strategy
Reverse auction
