面向任务的云、边、端分布式异构资源优选技术  

Task-oriented cloud-edge-end distributed heterogeneous resource optimization technology

作　　者：李大伟 李丹[1] 刘博文 刘赛赛 王栋 LI Da-wei;LI Dan;LIU Bo-wen;LIU Sai-sai;WANG Dong(China Academy of Electronics and Information Technology,Beijing 100041,China)

机构地区：[1]中国电子科学研究院,北京100041

出　　处：《计算机工程与设计》2024年第2期618-625,共8页Computer Engineering and Design

摘　　要：聚焦未来复杂、高动态环境下异构无人系统跨域联合、超视距分布式联合等应用需求,构建包含任务要素、多智能体要素的任务空间,提出一种混合知识驱动的“云-边-端”跨层跨域任务分配和资源优选模式,分析智能体任务包构建、冲突消解等关键技术,研究特定时间窗口下的分布式任务分配效果,考虑到复杂动态对抗环境因素影响,分析面向扰动的柔性适变方法。通过结合典型空海场景对任务分配算法进行仿真,验证跨域分布式任务分配、面向扰动的适变调整等设计的可行性和有效性。Focusing on the application requirements of heterogeneous unmanned platforms cross-domain joint,over-the-horizon distributed joint under the conditions of high dynamics and strong confrontation in the future,the task space that contained task elements and multi-agents was constructed.A hybrid-knowledge driven cloud-edge-node cross-layer and cross-domain task allocation and resource optimization mode was proposed.Key technologies such as agent task package construction and conflict resolution were analyzed,and the effects of distributed task allocation under specific time windows were studied.Considering the highly complex,dynamic and confrontational environment factors,the flexible adaptive method for disturbance was analyzed.By simulating the task allocation algorithm in combination with typical air-sea scenarios,the feasibility and effectiveness of cross-domain distributed task allocation and disturbance-oriented adaptive adjustment are verified.

关 键 词：异构无人系统 分布式动态任务分配 云-边-端 多智能体 任务空间 全局共识 冲突消解 柔性适变 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]