小天体探测自主绕飞智能规划建模

Modeling of Autonomous Flight Mission Intelligent Planning for Small Body Exploration

针对小天体探测存在显著通讯延迟、任务执行效率低等问题,梳理了小天体探测智能规划需求,面向自主绕飞任务开展了智能规划研究。首先将该问题分解为平台任务智能规划和载荷任务智能规划两部分。针对平台任务智能规划问题,基于PDDL语言设计了探测器自主管理知识模型,提出了基于状态时间线扩展的求解算法;针对任务智能规划问题,建立了基于CSP问题的智能规划数学模型,提出了基于遗传策略的求解算法。最后开发了仿真系统进行算法验证。仿真结果表明:该方法可综合平台与载荷需求,在存储、能源、通信等多种约束条件下,对绕飞探测任务进行统一的任务规划,并得到指令序列和动作序列,能够提高任务管控的智能化程度,降低任务操作的复杂性。

The needs for samll body intelligent planning are analuzed due to the significant communication delay and low efficiency of mission execution.The intelligent planning of autonomous flight aroundfor small body exploration mission is studied.Firstly,the problem is decomposed into two parts:platform task intelligent planning and payload task intelligent planning.A knowledge model of detector autonomous management is designed based on PDDL language,and a solution algorithm of specific state time line extension is proposed.The mathematical model of intelligent planning based on CSP is established,and the solving algorithm based on genetic strategy is proposed.Finally,a simulation system is developed to verify the algorithm.The simulation results show that the method can integrate the platform and payload requirements,making unified mission planning under the constraints of storage,energy,communication and other constraints,and obtaining command sequence and action sequence.It can improve the intelligence of task management and control,and reduce the complexity of task operation.