基于计算机视觉/GPS/MG姿态测量的盲人行走组合导航方法研究

电子导盲辅助装置(ETA)是解决盲人出行困难的重要手段,而导航是ETA的关键技术.现有的ETA主要用GPS来定位定向,但在城市环境中经常存在GPS信号遮挡导致导航信息丢失的问题.针对该问题,利用视觉导航短时间内定位精度高,输出连续的优点以及 MG(Magnetic Gravity)姿态测量可补偿姿态积累误差的优点,提出一种基于视觉、GPS和MG姿态测量的盲人行走组合导航算法.该方法构建系统误差模型并以Kalman滤波为框架.仿真和实验结果表明,提出的组合导航算法准确度优于单独的导航算法,满足盲人户外安全出行导航的需求.

Integrated Navigation Method of the Blind Walking Based on Computer Vision/GPS/MG Attitude Measurement

ETA(electronic travel aids)is an important means to solve the difficulties of the blind travel,in which navigation is the crucial technology of ETA.The existing ETA mainly uses the GPS to get the position and the orientation,however,the navigation information is often lost because of occlusion in the urban.Aiming at the problem,this paper propo-ses an integrated navigation method of the blind walking based on vision,GPS and MG attitude measurement.The method takes advantage of the vision navigation which has high positioning accuracy in short time and continuous output,and MG attitude measurement compensated the attitude accumulation errors.The method constructs the system error model and uses Kalman filter as the framework.Simulation and experimental results show that the proposed integrated navigation algorithm is superior to the single navigation algorithm.