摘要
如何滤除移动物体对实时建图的干扰是SLAM在复杂动态环境中需要解决的关键问题之一。论文提出一种新的面向动态3D场景的激光雷达SLAM算法。论文算法首先提取特征点,通过配准出估计机器人的位姿,然后设计了一种移动物体的滤除方法,通过使用卡尔曼目标跟踪算法滤除移动物体,最终构建出环境的静态地图。通过在真实场景中的移动机器人上实验表明,该方法提高了SLAM系统在动态环境中构建静态地图的鲁棒性,满足实时性要求,具有较好的实际使用价值。
How to filter out the interference of moving objects on real-time mapping is one of the key problems that SLAM needs to solve in a complex dynamic environment.A new lidar SLAM algorithm for dynamic 3D scenes is proposed.The algorithm first extracts feature points and registrates to estimate the pose of the robot,and then designs a method to filter moving objects which remove moving objects with the Kalman target tracking algorithm,and finally builds a static environment map.Experiments on mobile robot in real scenes show that this method improves the robustness of the SLAM system in building static maps in dynamic environments,meets real-time requirements,and has good practical value.
作者
唐印
TANG Yin(School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210094)
出处
《计算机与数字工程》
2022年第11期2449-2453,共5页
Computer & Digital Engineering
