Abstract: Indoor mobile robots are a hot issue in recent years, and synchronous positioning and mapping technology is one of the key technologies of mobile robot. To this end, a synchronous positioning and the mapping system is designed in this paper, based on robot operating system (ROS) intelligent car, which takes Gmapping as the core algorithm. Firstly, to compensate for the motion distortion per frame of the data of the single-line lidar with the odometer constructed by the motor encoder, lidar data and odometer data after motion distortion compensation is used as the input of the Gmapping algorithm; Then particle initialized via the Gmapping algorithm, construct a proposed distribution considering the observed quantity and sampling the particles to estimate the intelligent parking poses, resampling the particles with the resampling importance coefficient and setting the resampling threshold, update the map state of each particle with a binary bayesian filter; At last, the pose estimation and map data of intelligent car based on the Gmapping algorithm. The test results show that the designed synchronous positioning and the mapping system are more accurate in the intelligent parking pose estimation and construction map in a small size environment, which meets the design target requirements.

Key words: Indoor ROS intelligent car; Gmapping algorithm; Synchronous localization and drawing mapping system; Lidar

摘要： 室内移动机器人是近年来研究的热点问题，同步定位与建图技术更是移动机器人的关 键技术之一。为此文章设计了基于机器人操作系统（ROS）智能车的同步定位与建图系统， 以 Gmapping 为核心算法，首先采用电机编码器构建的里程计对单线激光雷达的每帧数据进行 运动畸变补偿，以运动畸变补偿后的激光雷达数据和里程计数据作为 Gmapping 算法输入；然 后通过 Gmapping 算法粒子初始化，构建考虑观测量的提议分布并对粒子进行采样以估计智能 车位姿，以重采样重要性系数对粒子进行重采样并设置重采样阀值，利用二值贝叶斯滤波器 对每个粒子的地图状态更新；最后基于 Gmapping 算法输出智能车的位姿估计和地图数据。试 验结果表明，所设计的同步定位与建图系统，在小尺寸环境下对智能车位姿估计和构建的地 图较为准确，满足设计目标要求。

关键词: 室内 ROS 智能车；Gmapping 算法；同步定位与建图系统；激光雷达