Turtlebot机器人自主导航及避障实验设计研究

Experimental Design of Autonomous Navigation and Obstacle Avoidance for Turtlebot Robot

摘要: 随着机器人领域研究的深入，设计一款能够自主移动的家庭服务机器人具有重要的现实意义。本文基于自主导航和避障功能的实现，设计了一款基于ROS平台的Turtlebot机器人。为了实现自主导航和避障功能，采用Dijkstra算法进行全局路径规划，采用动态窗口法进行局部路径规划。通过搭载深度相机Astra和激光雷达EAI F4两款传感器，进行SLAM和自主导航功能效果的比较。实验过程中使用Gmapping算法来实现SLAM建图。此外，通过编写脚本，原本需要人工控制的SLAM建图过程也被改进成了无需人工控制的自主SLAM。实验过程也完成了这款机器人自主导航和避障的功能。

Abstract: With the research of robotics advancing, it is important to design a home service robot that can move autonomously. The Turtlebot robot based on ROS platform is used to achieve the autonomous navigation and obstacle avoidance functions. In order to realize the autonomous navigation function, the Dijkstra algorithm is used for global path planning, and the dynamic window algorithm is used for local path planning. A comparison between the results of SLAM and autonomous navigation is carried out by using two sensors, the depth camera Astra and the lidar EAI F4. During the experi-ment, the Gmapping algorithm was used to implement SLAM mapping. In addition, by writing scripts, the SLAM mapping process, which originally required manual control, has also been im-proved to autonomous SLAM without manual control, making it more convenient in practical appli-cations. The results of simulation and experiments show that autonomous navigation and obstacle avoidance functions have been implemented.