基于群体角色的无人机集群任务分配  

Task allocation of unmanned aerial vehicle cluster based on group roles

作　　者：王传云[1] 曾跨省 郑会龙[2,3,4] 高骞 Wang Chuanyun;Zeng Kuasheng;Zheng Huilong;Gao Qian(Shenyang Aerospace University,Shenyang 110136,China;Institute of Engineering Thermophysics,Beijing 100190,China;National Key Laboratory of Science and Technology on Advanced Light-duty Gas-turbine,Beijing 100190,China;Binzhou Weiqiao National Science and Technology Advanced Technology Research Institute,Binzhou 256212,China)

机构地区：[1]沈阳航空航天大学,沈阳110136 [2]中国科学院工程热物理研究所,北京100190 [3]轻型涡轮动力全国重点实验室,北京100190 [4]滨州魏桥国科高等技术研究院,滨州256212

出　　处：《空天技术》2024年第6期96-104,共9页Aerospace Technology

基　　金：国家自然科学基金资助项目(61703287);辽宁省自然科学基金计划项目(2024-MS-137);辽宁省教育厅科学研究项目(LJKMZ20220556);中国国家铁路集团有限公司项目(N2023X005)。

摘　　要：为提高无人机集群执行复杂任务的能力,解决无人机集群面对多个攻击目标时,无法快速根据所要打击目标的实际情况进行合理任务分配的问题,提出了采用基于E-CARGO模型的智能任务分配方法对现实环境进行分角色建模,按照无人机到目标的欧式距离计算代价矩阵,并将得到的代价矩阵输入模型,进而快速得到分配矩阵。解决了使用传统方法进行无人机集群任务分配时主机负载较大、分配耗时较长等问题,并通过对比实验验证了算法的精确性和实时性,最终,通过搭建半物理仿真平台,将算法部署到无人机集群上进行验证,确定算法无误后,将其部署到真机上进行下一步实验。结果表明,基于E-CARGO模型的任务分配算法在精确性、实时性方面表现良好,且满足大规模无人机集群的任务分配需求。To enhance the capabilities of unmanned aerial vehicle(UAV)clusters in executing complex tasks and address the challenge of task allocation based on factors such as strike capabilities and locations of different subclusters upon reaching the destination,an intelligent task allocation method utilizes the ECARGO model for role-based modeling of real-world environments is proposed.The method involves calculating the cost based on the euclidean distance from UAVs to targets and inputting the obtained cost matrix into the model to rapidly derive an allocation matrix.The issues such as heavy computational loads and lengthy allocation times associated with traditional methods in UAV clusters task allocation are resolved.Comparative experiments are conducted to validate the accuracy and real-time performance of the algorithm.The algorithm is deployed to a UAV cluster through a semi-physical simulation platform for further verification before real-world experiments.The results indicate that the task allocation algorithm based on the E-CARGO model performs well in terms of accuracy and real-time responsiveness and is suitable for large-scale UAV clusters task allocation requirements.

关 键 词：无人机 集群任务分配 E-CARGO 组角色分配 多智能体系统 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TP212.9[自动化与计算机技术—检测技术与自动化装置]