Reconfigurable mobile planetary rovers are versatile platforms that may safely traverse cluttered environments by morphing their physical geometry. Planning paths for these adaptive robots is challenging due to their many degrees of freedom, and the need to consider potentially continuous platform reconfiguration along the length of the path. We propose a novel hierarchical structure for asymptotically optimal (AO) sampling‐based planners and specifically apply it to the state‐of‐the‐art Fast Marching Tree (FMT*) AO planner. Our algorithm assumes a decomposition of the full configuration space into multiple subspaces, and begins by rapidly finding a set of paths through one such subspace. This set of solutions is used to generate a biased sampling distribution, which is then explored to find a solution in the full configuration space. This technique provides a novel way to incorporate prior knowledge of subspaces to efficiently bias search within existing AO sampling‐based planners. Importantly, probabilistic completeness and asymptotic optimality are preserved. Experimental results in simulation are provided that benchmark the algorithm against state‐of‐the‐art sampling‐based planners without the hierarchical variation. Additional experimental results performed with a physical wheel‐on‐leg platform demonstrate application to planetary rover mobility and showcase how constraints such as actuator failures and sensor pointing may be easily incorporated into the planning problem. In minimizing an energy objective that combines an approximation of the mechanical work required for platform locomotion with that required for reconfiguration, the planner produces intuitive behaviors where the robot dynamically adjusts its footprint, varies its height, and clambers over obstacles using legged locomotion. These results illustrate the generality of the planner in exploiting the platform's mechanical ability to fluidly transition between various physical geometric configurations, and wheeled/legged locomotion modes, without the need for predefined configurations.

中文翻译：

---

基于采样的分层运动计划，用于可重新配置的轮对腿行星模拟探索漫游车

可重新配置的移动行星漫游车是通用平台，可以通过变形其物理几何形状来安全地穿越混乱的环境。由于这些自适应机器人的许多自由度，以及为沿路径长度考虑潜在的连续平台重新配置的需求，为它们规划路径非常具有挑战性。我们为基于渐进最优（AO）抽样的规划人员提出了一种新颖的层次结构，并将其专门应用于最新的快速行进树（FMT *）AO规划人员。我们的算法假设将整个配置空间分解为多个子空间，并从快速找到通过一个这样的子空间的一组路径开始。这套解决方案用于生成有偏差的采样分布，然后对其进行探索以在整个配置空间中找到解决方案。这项技术提供了一种新颖的方式来合并子空间的先验知识，以在现有基于AO采样的规划人员中有效地偏向搜索。重要的是，保留了概率完整性和渐近最优性。提供了仿真实验结果，以针对最新的基于抽样的计划者对算法进行基准测试，而没有分层的变化。使用实际的腿上轮式平台进行的其他实验结果证明了其在行星漫游车机动性中的应用，并展示了如何将执行器故障和传感器指向等约束轻松纳入规划问题。在将平台移动所需的机械功与重新配置所需的近似功相结合的能量目标降至最低时，规划器会产生直观的行为，在这种行为中，机器人会通过有腿的运动来动态调整其足迹，改变其高度并爬上障碍物。这些结果说明了计划者利用平台的机械能力在各种物理几何配置和轮式/腿式运动模式之间进行流体转换的通用性，而无需预定义的配置。