论文部分内容阅读
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.
Lidar has been widely used in obstacle avoidance of mobile robots because of its high angle and distance resolution, but its inherent probing characteristics make it easy for the robot to fall into the V-type obstacle trap and increase the cost of the path.Therefore, According to the relationship between the accuracy of radar and the distance, this paper proposes that the detection area is divided into fuzzy avoidance area, precise avoidance area and emergency avoidance area according to the detection accuracy, and then a hierarchical obstacle avoidance method is established.Aiming at the design of fuzzy avoidance area, At the same time, the collision avoidance of the boundary point tracking method is used in the precise avoidance area.The simulation results show that the proposed method can not only avoid the V-type trap but also lead to a smaller path cost than the traditional lidar collision avoidance method .