摘要
为了满足机器人精确定位的要求,提出一种基于多传感器信息融合的自定位算法并介绍了如何利用Matlab搭建移动机器人定位系统的仿真模型.仿真程序建立了移动机器人活动的虚拟环境,模拟了机器人的运动模型、里程计、激光雷达观测模型,利用扩展卡尔曼滤波算法将里程计和激光传感器采集的数据进行融合;最后,由匹配的环境特征对机器人的位置进行修正,得到精确的位置估计.仿真试验的结果表明该定位系统具有较高的定位精度.模块化的仿真系统设计有利于其他定位算法的验证,对机器人系统的理论研究以及实用化具有重要的作用.
Robot self-localization is one of the most important issues to build map and plan a route.Always an accurate self-localization algorithm is tested by simulation system,then inspected in actual system.In order to meet the requirement of accurate localization,this paper proposes a self-localization algorithm based on multi-sensor information and introduces how to establish the mobile robot localization simulation system by Matlab.Simulation program establishes the virtual environment and the kinematics model of a mobile robot and the measurement model of odometric and laser radar sensors.The data provided by odometric and laser radar sensors are fused together by means of an extended Kalman filter technique.Finally,the position of robot is reset by matched environment feature; and the position estimation of robot is given accurately.Simulation results show that the positioning system has a higher position precision.The modular design of simulation system is for the benefit of other localization algorithm verification,which play an important role in the theoretical and practical study of the robot system.
出处
《武汉大学学报(工学版)》
CAS
CSCD
北大核心
2010年第6期784-787,791,共5页
Engineering Journal of Wuhan University
基金
国家自然科学基金项目(编号:60772107)
湖北第二师范学院重点项目(多机器人协作系统研究)