高技术通讯2017，Vol.27Issue(9)：848-855,8.DOI:10.3772/j.issn.1002-0470.2017.09-10.010
基于自适应观测模型的移动机器人室内蒙特卡罗动态定位系统研究
Research on an indoor Monte Carlo dynamic positioning system for mobile robots based on adaptive observation model
摘要
Abstract
Aiming at the problem that the positioning performance of traditional Monte Carle localization ( MCL ) algo-rithms for mobile robots delines when working in the environment with glass doors and pedestrians because environ -mental noises cause dramatic changes in observation information , a new Monte Carlo localization algorithm based on an improved observation model is presented .The algorithm introduces the measurement failure error and dynamic error into the observation model to improve the laser sensor ' s effectiveness of data measuring , and the data meas-ured by the laser is used to match the created grid map to detect whether known environment changes , so the corre-sponding weight of the random error is changed , and the failure error and the dynamic error are measured , and then the noise impact on the measured value is reduced , which improves the location accuracy of the improved algorithm in complex environment .The new algorithm is tested based on the robot operating system ( ROS) , and the experi-mental results prove its effectiveness .关键词
蒙特卡罗定位(MCL)算法/观测信息/激光传感器/占用栅格地图/机器人操作系统(ROS)Key words
Monte Carle localization(MCL) algorithm/observation information/laser sensor/occupancy grid mapping/robot operating system ( ROS)引用本文复制引用
郑文磊,程磊,余秋月,陈泓宇,吴秋轩..基于自适应观测模型的移动机器人室内蒙特卡罗动态定位系统研究[J].高技术通讯,2017,27(9):848-855,8.基金项目
国家自然科学基金(60705035,61203331,61573263),湖北省自然科学基金(2014CFB813),湖北省科技支撑计划(2015BAA018),湖北省教育厅科研计划重点项目(D20131105),国家级大学生创新创业训练计划(201610488009)和杭州电子科技大学重中之重学科开放基金和浙江省自然科学基金(LY16F030007)资助项目. （60705035,61203331,61573263）
