Scholarlink

<p>This paper presents a hierarchical tree-based approach to automatically generating spacecraft trajectories in the Earth-Moon circular restricted three-body problem. Inspired by the use of rapidly-exploring random trees in robotic motion planning, a tree is constructed to explore a local region of the phase space. This process is repeated to produce a forest of trees that collectively covers a broader region of the phase space. Then, the initial guess construction problem is reformulated as a hierarchical graph search problem. First, a tree sequence graph is searched to produce sequences of connected trees. A localized path graph is then formed for each tree sequence and searched to identify a sequence of branches that form an initial guess for a path between two boundary conditions. To demonstrate the utility of these initial guesses, correction and continuation schemes are leveraged to recover continuous trajectories. Then, the paths are analyzed for collisions with dynamic obstacle corridors. This procedure is demonstrated by automatically generating an array of geometrically distinct initial guesses between selected Lyapunov and halo orbits.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Tree-Based Approach to Spacecraft Trajectory Design in Multi-Body Systems

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月18日

查看全文
  • Renee L. Spear,
  • Giuliana E. Miceli,
  • Natasha Bosanac

摘要

This paper presents a hierarchical tree-based approach to automatically generating spacecraft trajectories in the Earth-Moon circular restricted three-body problem. Inspired by the use of rapidly-exploring random trees in robotic motion planning, a tree is constructed to explore a local region of the phase space. This process is repeated to produce a forest of trees that collectively covers a broader region of the phase space. Then, the initial guess construction problem is reformulated as a hierarchical graph search problem. First, a tree sequence graph is searched to produce sequences of connected trees. A localized path graph is then formed for each tree sequence and searched to identify a sequence of branches that form an initial guess for a path between two boundary conditions. To demonstrate the utility of these initial guesses, correction and continuation schemes are leveraged to recover continuous trajectories. Then, the paths are analyzed for collisions with dynamic obstacle corridors. This procedure is demonstrated by automatically generating an array of geometrically distinct initial guesses between selected Lyapunov and halo orbits.