Robot navigation in unknown terrains: Introductory survey of non-heuristic algorithms
Robot navigation in unknown terrains: Introductory survey of non-heuristic algorithms
复制标题
DOI：
10.2172/10180101
复制
发表时间：
1993-07
期刊：
影响因子：
--
通讯作者：
N. Rao;Srikumar Kareti;W. Shi;S. Iyengar;Reinhold Mann;E. Oblow;D. Reister;Judd P Jones
中科院分区：
文献类型：
--
作者：
N. Rao;Srikumar Kareti;W. Shi;S. Iyengar;Reinhold Mann;E. Oblow;D. Reister;Judd P Jones
考虑了一个正式的框架，用于通过一组未知的障碍在几何地形中导航机器人。这里的地形模型不是先验的，但机器人配备了用于导航目的的传感器系统（视觉或触摸）。焦点仅限于非效法算法，理论上可以证明在机器人，地形和传感器系统的给定模型框架内是正确的。这些配方虽然与现实生活中的情景相比抽象和简化，但通过突出潜在的关键问题为实用系统提供了基础。首先，作者考虑显示出正确导航的算法，而没有太多考虑性能参数（例如距离横穿等性能等）。其次，他们考虑了保证在距离横穿距离或距离距离之比的范围界限的非武器算法到最短路径长度（如果已知地形模型，则计算）。然后，他们考虑具有非常有限的计算功能（例如有限自动机）的机器人的导航。
A formal framework for navigating a robot in a geometric terrain by an unknown set of obstacles is considered. Here the terrain model is not a priori known, but the robot is equipped with a sensor system (vision or touch) employed for the purpose of navigation. The focus is restricted to the non-heuristic algorithms which can be theoretically shown to be correct within a given framework of models for the robot, terrain and sensor system. These formulations, although abstract and simplified compared to real-life scenarios, provide foundations for practical systems by highlighting the underlying critical issues. First, the authors consider the algorithms that are shown to navigate correctly without much consideration given to the performance parameters such as distance traversed, etc. Second, they consider non-heuristic algorithms that guarantee bounds on the distance traversed or the ratio of the distance traversed to the shortest path length (computed if the terrain model is known). Then they consider the navigation of robots with very limited computational capabilities such as finite automata, etc.