Coverage path planning technique is an essential ingredient in every floor cleaning robotic systems. Even though numerous approaches demonstrate the benefits of conventional coverage motion planning techniques, they are mostly limited to fixed morphological platforms. In this article, we put forward a novel motion planning technique for a Tetris-inspired reconfigurable floor cleaning robot named “hTetro” that can reconfigure its morphology to any of the seven one-sided Tetris pieces. The proposed motion planning technique adapts polyomino tiling theory to tile a defined space, generates reference coordinates, and produces a navigation path to traverse on the generated tile-set with an objective of maximizing the area coverage. We have summarized all these aspects and concluded with experiments in a simulated environment that benchmarks the proposed technique with conventional approaches. The results show that the proposed motion planning technique achieves significantly higher performance in terms of area recovered than the traditional methods.

中文翻译：

---

受俄罗斯方块启发的可重构地板清洁机器人的运动规划器

覆盖路径规划技术是每个地板清洁机器人系统中必不可少的组成部分。尽管许多方法证明了传统覆盖运动规划技术的好处，但它们大多仅限于固定的形态学平台。在本文中，我们为受俄罗斯方块启发的可重构地板清洁机器人“hTetro”提出了一种新颖的运动规划技术，该技术可以将其形态重新配置为七个单面俄罗斯方块中的任何一个。所提出的运动规划技术采用多米诺平铺理论来平铺定义的空间，生成参考坐标，并生成导航路径以在生成的平铺集上遍历，目标是最大化区域覆盖。我们总结了所有这些方面，并在模拟环境中通过实验得出结论，该模拟环境用传统方法对所提出的技术进行了基准测试。结果表明，与传统方法相比，所提出的运动规划技术在恢复区域方面取得了显着更高的性能。