基于声纳的移动机器人环境建图的设计与实现*

针对未知环境中声纳传感器定位与地图创建时传感数据不确定性高、可靠性低的问题，提出了一种新的室内环境建图方法。该方法建立容忍函数以判断噪声和镜面反射，同时借鉴了ArcTransversal Median Algorithm的思想和栅格概率估计并采用贝叶斯法则进行两次数据融合以减小声纳传感器信息的不确定性。在MORCS2机器人平台上实时创建地图实验表明，这种方法能快速实现从局部地图到全局地图的更新且有较好的精确性与鲁棒性。

Sonar based map building design and implementation for mobile robot

This paper presented a new method of the indoor environment ,which was used to solve the problem that sensor data are with high uncertainty and low reliability when simultaneously localizating and mapping (SLAM) in unknown environment with sonar sensors.This method constructed tolerance function to judge the noise and mirror reflection,referenced the ATM method and grid probability estimation and used Bayes rule in two sonar information's fusion to reduce the uncertainty in sonar sensors.Then real-time sett...