摘要
目的:验证基于实际订单数据的搬运机器人的任务调度和算法测试平台可行性。方法:通过自主搭建的仿真实验验证平台,以物流中心中典型的仿真场景——存在任务分级的多拣选机器人任务分配策略场景为例,设计基于记忆精英种群的灾变自适应大邻域搜索算法。结果:相比ALNS算法,MEPCALNS算法效率平均提升3.16%。结论:通过算法优化研究和各种场景的模拟运行比较,证明了平台在算法验证方面的价值,说明MEPCALNS算法在典型场景下的有效性。
Objective:Verify the feasibility of the task scheduling and algorithm test platform of the materials transportation robots system based on the actual order data.Methods:This paper discusses the design of simulation experiment through the independently developed platform,focusing on the typical scenarios of pharmaceutical logistics.Aiming at the logistics scenario of task allocation strategy of multi robot with task classification,the Memory Elite Population based Catastrophe Adaptive Large Neighborhood Search(MEPCALNS)algorithm is designed.Results:Compared with ALNS algorithm,the efficiency of MEPCALNS algorithm is increased by 3.16%on average.Conclusion:By means of algorithm optimization research and simulation operation comparison of various scenarios.The value of the platform in algorithm verification is proved,which shows the effectiveness of MEPCALNS algorithm in typical scenarios.
作者
翁迅
张经天
胡晓
WENG Xun;ZHANG Jing-tian;HU Xiao(School of Modern Post,Beijing University of Posts and Telecommunications,Beijing 100876,China;School of Artificial Intelligence,Beijing University of Posts and Telecommunications,Beijing 100876,China)
出处
《食品与机械》
北大核心
2022年第7期122-127,共6页
Food and Machinery
基金
科技部科技创新2030“新一代人工智能”重大项目(编号:2021ZD0114204)
中央高校基本科研业务费专项资金资助项目(编号:2020RC15)
邮政行业人才发展研究课题(编号:202101)。
关键词
搬运机器人
任务调度
仿真验证平台
算法优化
materials transportation robots system
task scheduling
simulation experiment platform
algorithm optimization
