论文部分内容阅读
针对基于视觉定位的移动机器人自主定位过程中容易产生多种候选位姿.提出一种基于多假设跟踪的同时定位和地图创建方法.该方法通过提取图像SIFT特征进行视觉量测,修正里程计累计误差,再利用多假设跟踪方法获得当前时刻的最优位姿,实时产生环境的栅格地图,并且随着机器人的位姿变化实时更新.研究结果表明:此算法不仅能够较好地创建环境地图,而且能够有效解决机器人“绑架”问题,提高自主定位精度.
Aiming at the problem of multiple candidate pose in the process of autonomous localization of mobile robot based on visual positioning, a simultaneous positioning and map creation method based on multi-hypothesis tracking is proposed in this paper. The method takes the SIFT feature of the image for visual measurement to correct the odometer totalizer Error, and then use the multi-hypothesis tracking method to obtain the optimal pose and pose of the current time, and generate a real-time raster map of the environment, and update the robot’s position and pose in real time.The results show that this algorithm not only can create a good environment map , But also can effectively solve the robot “kidnapping ” problem, improve the accuracy of autonomous positioning.