摘要
多无人机协同执行巡视任务是实时精确情报获取的重要手段,合理的任务规划将大幅度提升巡视效能。首先分析了国内外关于多无人机协同控制、任务规划的研究现状。然后介绍了协同巡视任务规划的概念、常用模型及其不足之处,分析了集中式、分布式这两种决策控制系统结构的概念及优缺点,介绍了最优化方法、启发式方法和合同竞拍这三种常用求解方法的特点及应用场景。最后总结了现阶段在建模的精确性、不确定性环境中任务分配的自主性和算法快速收敛性等方面研究中存在的问题及未来发展方向。
Multi-UAV cooperative patrol is an important means of obtaining accurate and real-time intelligence,and the reasonable patrol task planning will significantly enhance the performance of multi-UAV.First of all,the research status of cooperative control of multi-UAV and task planning,both home and abroad,are analyzed. Then,the concept and common models of cooperative patrol task planning as well as shortcomings are introduced. Afterwards,the concept,advantages and disadvantages of both centralized and distributed control systems are analyzed. The characteristics and application scenarios of three common solving methods are introduced: the optimization methods,heuristic methods and contract auction. Finally,the problems existing in the research on the accuracy of the modeling,the autonomy of task allocation in the uncertain environment,and the fast convergence of the algorithm are summarized.
作者
杨杰
席建祥
王成
谢学立
YANG Jie;XI Jian-xiang;WANG Cheng;XIE Xue-li(Department of Automatic Control Engineering,Rocket Force Engineering University,Xi'an 710025,China)
出处
《飞行力学》
CSCD
北大核心
2018年第5期1-6,共6页
Flight Dynamics
基金
国家自然科学基金资助(61374054)
关键词
无人机
协同巡视
任务规划
分布式
UAV
cooperative patrol
task planning
distributed