摘要
针对室内未知环境下的避障和局部路径规划,提出了一种单目移动机器人路径规划算法,该算法通过对环境图像的自适应阈值分割,获取障碍物与地面交线轮廓点集。通过对现有几种单目测距方法的分析比较,提出一种改进的空间几何约束单目视觉测距计算方法,并依据单目测距的几何关系建立了图像坐标系与机器人坐标系的映射,绘建了一定比例的局部地图。在局部地图上通过改进的人工势场算法为机器人规划路径,改进的人工势场算法解决了传统算法目标点不可到达的问题。通过MATLAB进行仿真实验,结果表明该方法可以规划出有效合理的路径。
Aiming to the obstacle avoidance and local path planning in unknown indoor environments, a monocular vision-based mobile robot path planning algorithm is proposed. The contour points of the obstacles and the ground are obtained by the algorithm adaptive threshold segmented to the environment images. Analyzing and comparing the present several monocular measuring methods, a spatial geometric constrains with monocular visual measuring calculation method is improved. the mapping between the image coordinate system and the robot coordinate system is established according to the geometric relationship of monocular measuring method, and a certain scale of the local map is built. The improved artificial potential field algorithm is used for robot to plan a path in a local map. The unreachable target point problem of the traditional algorithm has been solved by the improved artificial potential field algorithm. Through MATLAB simulation experiments, the result shows that the method can plan the effective and reasonable path.
出处
《计算机工程与应用》
CSCD
北大核心
2018年第3期255-258,264,共5页
Computer Engineering and Applications
基金
智能信号与信息处理四川省青年科技创新研究团队(No.2015TD0022)
2017年四川省第一批科技计划重点研发项目(No.2017GZ0068)
四川理工学院研究生创新基金(No.y2016040)
四川理工学院学科建设项目(No.2014JC02)
关键词
未知环境
单目视觉
障碍物测距
地图创建
路径规划
unknown environment
monocular vision
obstacle distance measurement
map building
path planning
