论文部分内容阅读
提出移动机器人的一种新的完全遍历算法:矩形分解算法。首先通过机器人环境学习建立栅格地图,对环境中的障碍物实行矩形化建模。而后应用矩形化模型中的关键点将环境分解成为矩形块,最后在这个分块环境的拓扑图中寻找到一条Hamilton路径,机器人沿此路径即可实现对环境的完全遍历。为处理复杂的局部情况,又提出基于模板的局部环境处理算法。矩形算法的优点在于机器人可以实现完全自主的复杂环境遍历,并且可以处理未知障碍,从而使算法适合于任意非结构化的工作环境。
A new complete traversal algorithm of mobile robot: rectangular decomposition algorithm is proposed. First through the robot environment to create a grid map, rectangular obstruction of the implementation of the environment modeling. Then, the key points in the rectangular model are used to decompose the environment into rectangular blocks. Finally, a Hamilton path is found in the topological map of this block environment, and the robot can complete the environment traversal along this path. In order to deal with complex local conditions, a template-based local environment processing algorithm is proposed. The advantage of the rectangular algorithm is that the robot can achieve completely autonomous and complex environment traversal, and can handle unknown obstacles so as to make the algorithm suitable for any unstructured working environment.