基于路标的机器人自定位方法

介绍了一种基于路标的移动机器人自定位方法。为了解决仅仅利用视觉系统得到的路标信息使用蒙特卡洛自定位(MCL)算法进行机器人自定位时存在较大误差的问题,提出了一种将MCL算法与unscented卡尔曼滤波器相结合进行自定位的方法,将从机器人视觉系统获取的路标信息,从驱动轮码盘获取的位置信息,以及从电子罗盘获取的机器人方位信息进行有效融合,从而提高了机器人的自定位精度。实验结果表明,该方法不但可以提高机器人的自定位精度,而且具有较好的鲁棒性。

An Approach for Robot Self-localization Based on Landmark

A method for robot self-localization based on landmark is presented. To solve the problem that a MCL algorithm can not give good self-localization results using the information of the land marks obtained from the robot vision system. A new robot self-localization method which combines the MCL algorithm and the UKF is proposed. In this new method, the information of landmarks, the position and azimuth of the robot, which are obtained from robot vision system, the encoder of the driving motors, the e-lectronic compass of the robot, respectively, are used to improve the precision of the self-localization of the robot. It has been shown by experimental results that this new method can not only improve the precision of the robot self-localization, but also is quite robust.