基于层级任务网络的卫星自主规划模型与算法

HTN-based Model and Algorithm for Autonomous Planning of Satellite

针对面向任务的卫星操作计划自动生成(即卫星的自主规划)问题,采用层级任务网络规划的任务分解思想,在分析卫星系统结构的基础上,建立了光学成像侦察卫星的规划模型,定义了2种常规任务的任务分解方法。根据模型的特点,开发了一个基于任务分解方法的导引式求解算法。通过一个实例进行了验证,结果表明模型和算法能有效解决卫星的自主规划问题。

The problem of operation planning for satellite is investigated. Task decomposition approach of hierarchical task network（HTN） is adopted. Based on the analysis of system structure for satellite, the planning domain model is built and decomposition methods of two main kinds of tasks of satellite are defined. Subsequently an algorithm for the model is developed, which is based on the forward search algorithm. An example is given to evaluate the model and corresponding algorithm and the results show that the autonomous planning problem can be solved successfully using the approach.