吸尘机器人非结构环境下自主路径规划方法的研究

【作者】 冯申珅；

【导师】 朱世强； 金波；

【作者基本信息】 浙江大学 ， 机械电子工程， 2004， 硕士

【摘要】 吸尘机器人遍历路径规划是实现机器人智能的一个关键部分，本文主要以吸尘机器人在非结构化环境中的自主路径规划设计为中心，通过对吸尘机器人的一些关键技术—智能控制技术、路径规划技术等的研究，提出一种适应性较强的路径规划方法，以达到最终实现吸尘机器人家居环境中自主、独立工作的目的。 本文共由五章组成：第一章首先概述了吸尘机器人技术和机器人路径规划的主要算法。第二章则介绍了遍历路径规划实现的两种方案，以此作为后续研究的指导。第三章针对环境中障碍静止且室内环境局部已知的情况，提出一种将环境边界确定性考虑在内的遍历路径规划方法，并且对算法进行仿真并且给出模拟非结构化环境中的实验结果。第四章针对第三章提出算法的缺点，排出不确定传感器信息的影响，试验一种随机规划与局部路径规划相结合的混合路径规划方法。该算法不考虑全局定位，只利用局部有效定位信息，对复杂环境的适应性增强。同时，遍历路径规划过程中也考虑运动障碍的影响，设计相应“逃离”算法。另外，在遍历规划结束后，设计回归充电座的路径规划方法，并且论证各种定位误差下的回归必然性。最后，给出各个算法的实验结果并给予分析。更多还原

【Abstract】 Coverage path planning is a pivotal problem for Cleaning Robot to achieve intelligence, In this dissertation, the autonomous path planning in unstructured environment is described, along with the related key techniques such as intelligent control and path planning etc. An idea about coverage path planning is put forward to acclimate all kinds of complex conditions in order to improve the application of Cleaning Robot in the family life.First, chapter 1 gives a survey of algorithms about intelligent robot and path planning techniques. Chapter 2 describes the aim and detailed contents of this research project, and finally the implementations of algorithms are also theoretically summarized in this chapter. Chapter 3, based on the actionless obstacle and known indoor environment, a way of coverage path planning is given considering confirmability of environment borderline, then the algorithm is emulated and the test result are also provided. Chapter 4, according to the shortage of the algorithm provided in the chapter 3 and removing the influence of unsureness sensors information, a feasible mixed approach of path planning is discussed. This algorithm mixes the advantages of random planning and partial path planning, and uses the valid information of location in order to improve the applicability in the unstructured room environment. At the same time, in view of the moving obstacle, a escape algorithm is discussed. Then a path planning method to return to charger is given, and the way back inevitability is proved taking into account different locations errors. At last the test results of algorithm are provided and analyzed.更多还原