克服V型障碍陷阱的激光雷达机器人分层避撞方法

发布时间：2018-03-16 21:42

  本文选题：移动机器人　切入点：激光雷达　出处：《控制与决策》2017年08期 　论文类型：期刊论文

【摘要】：激光雷达以其高的角度和距离分辨率被广泛地应用于移动机器人的障碍规避,但其固有的探测特性易使机器人陷入V型障碍陷阱,并增大路径代价.为此,通过深入分析激光雷达精度与距离的特性关系,提出按照探测精度将探测区域划分为模糊规避区、精确规避区和应急规避区,进而建立一种分层障碍规避方法.为模糊规避区设计神经网络障碍规避算法,同时在精确规避区采用边界点追踪法避撞.仿真实验表明,所提出方法相较于传统的激光雷达避撞方法,不仅能够使机器人避免误入V型陷阱,而且可以产生更小的路径代价.
[Abstract]:Lidar is widely used in mobile robot for its high angle and range resolution, but its inherent detection characteristics easily make the robot fall into V-type obstacle trap and increase the path cost. Based on the analysis of the characteristic relationship between laser radar precision and range, it is proposed that the detection area be divided into fuzzy circumvention region, precise circumvention area and emergency circumvention area according to the detection accuracy. Then a layered obstacle avoidance method is established. The neural network obstacle avoidance algorithm is designed for the fuzzy circumvention area. At the same time, the boundary point tracking method is used to avoid collision in the precise circumvention area. The simulation results show that, Compared with the traditional laser radar collision avoidance method, the proposed method can not only avoid the robot from falling into V trap, but also produce a lower path cost.
【作者单位】： 空军工程大学航空航天工程学院;
【基金】：国家自然科学基金项目(61503405,61573373)
【分类号】：TP242
，

本文编号：1621765