论文部分内容阅读
在室外环境中,由于车轮打滑、里程计存在严重的累积误差,同时由于二维激光测距仪扫描到环境信息量少,在进行地图匹配时,机器人常常陷入相似或对称环境,导致定位失败。针对以上问题,在二维激光地图匹配的基础上,提出了利用单目视觉信息补偿激光信息与里程计信息的定位方法。该方法利用单目摄像机获取环境中的色块信息补偿二维激光信息;摄像机同时作为视觉里程计实时辨识车轮打滑,校正里程计的误差;最后利用粒子滤波算法融合多传感器信息。该实验利用Pioneer3-AT机器人平台在大学校园进行,实验证明该方法具有较高的定位精度。
In the outdoor environment, there is a serious cumulative error of the odometer due to wheel slippage. At the same time, due to the small amount of environmental information scanned by the two-dimensional laser range finder, robots often fall into a similar or symmetrical environment during the map matching, resulting in positioning failure. In view of the above problems, based on the two-dimensional laser map matching, a method of locating the information of laser information and odometer using monocular visual information is proposed. The method compensates the two-dimensional laser information by using the monocular camera to obtain the color patches in the environment. The camera also realizes the wheel slip as a visual odometer and corrects the error of the odometer. Finally, the particle filter algorithm is used to fuse the multi-sensor information. The experiments were carried out on a university campus using the Pioneer3-AT robot platform. Experimental results show that the proposed method has high positioning accuracy.