摘要
提出了一种移动机器人路径规划和避障的系统设计方案,实现了移动机器人自主行进的路径规划和自动避障功能。详细说明了如何采用立体视觉实现对环境的探测,利用图像处理算法的组合分离出地面、背景、障碍物和目标物,采用边界不变矩实现障碍物和目标物的区分,改进了经典的人工势场法进行路径的规划,根据模糊控制原理设计了避障控制器和避障规则。实际的运行结果表明了该系统的可行性和有效性,该系统实现了移动机器人利用自身传感器感知环境信息,动态规划行进路径,成功躲避障碍物等功能。
A system design scheme of mobile robot is proposed.The mobile robot implements the function of path-planning and obstacle avoidance.The paper explains how to percept the environment by binocular vision, how to separate the ground, background and objects by using a combination of several image processing algorithm, and how to distinguish the obstacles and target by image statistical characteristics.At the same time we plan the path by improving artificial potential field, and design a fuzzy controller and formulating a sequence of obstacle avoidance rules based on fuzzy logic.The result of the system proves that the design is feasibility and validity.The mobile robot system realize the request of the project:in the help of binocular vision sensors, planning a path dynamically and avoiding the obstacles.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第23期5462-5466,5470,共6页
Computer Engineering and Design
基金
上海市重点学科建设基金项目(J50103)
关键词
移动机器人
双目视觉
图像处理
不变矩
人工势场
模糊控制
mobile robot
binocular vision
image processing
moment invariants
artificial potential field
fuzzy control