基于多传感器的长廊环境全局路径规划研究  

Research on Global Path Planning of Corridor Environment Based on Multi-sensor

作　　者：王霏 赵庆帅 冯志权 邵海燕[1] WANG Fei;ZHAO Qing-shuai;FENG Zhi-quan;SHAO Hai-yan(School of Mechanical Engineering,University of Jinan,Jinan 250022,China;School of Information Science and Engineering,University of Jinan,Jinan 250022,China)

机构地区：[1]济南大学机械工程学院,山东济南250022 [2]济南大学信息科学与工程学院,山东济南250022

出　　处：《山东工业技术》2021年第2期13-19,共7页Journal of Shandong Industrial Technology

基　　金：企业委托开发项目(W2020232);济南市自主创新团队项目2019GXRC013。

摘　　要：针对在长距离、等宽度、环境特征相似的长廊环境中规划路径时存在特征匹配错误、误差累积等问题,对现阶段全局路径规划算法进行了分析研究。基于机器人操作系统(Robot Operating System,ROS)设计搭建了以四轮驱动小车为载体,以单线激光雷达、深度相机和超声波传感器为环境检测传感器的轮式机器人平台。该平台在室内长廊环境下对现阶段四种改进的全局路径规划算法进行了仿真研究,仿真结果表明,相比其他三种算法,基于静态网络改进的A*算法规划的轨迹含7个拐点轨迹相对平滑、规划时间为49.8 s耗时最短。最后静态网络的A*算法的路径规划实验和对于动态障碍物与静态障碍物的自主避障实验,实验结果表明长廊环境中路径规划总耗时178.6 s含8个拐点路径平滑且能够有效地实现动态及静态障碍物的自主避障,躲避静态障碍物耗时7.18 s且无拐点,躲避动态障碍物的平均反应时间为52 ms,相比原A*算法缩短了22.3%。Aiming at the problem of feature matching errors and error accumulation when planning paths in long-distance, constant-width, and similar environmental characteristics, the current global path planning algorithm is analyzed and researched. Based on the Robot Operating System(ROS), a wheeled robot platform is designed and built with a four-wheel drive car as a carrier and a single-line lidar, depth camera and ultrasonic sensor as environmental detection sensors. The platform simulates the four improved global path planning algorithms at this stage in the indoor corridor environment. The simulation results show that compared with the other three algorithms, the trajectory planned by the improved A* algorithm based on the static network contains 7 inflection points. The trajectory is relatively smooth, and the planning time is 49.8 s, which is the shortest. Finally, the path planning experiment of the static network A* algorithm and the autonomous obstacle avoidance experiment for dynamic obstacles and static obstacles. The experimental results show that the path planning in the promenade environment takes 178.6 s and contains 8 inflection points. The path is smooth and effective Realize autonomous obstacle avoidance of dynamic and static obstacles. It takes 7.18 s to avoid static obstacles without inflection points. The average reaction time to avoid dynamic obstacles is 52 ms, which is 22.3% shorter than the original A* algorithm.

关 键 词：ROS 轮式机器人 多传感器融合 路径规划 室内长廊环境 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]