CAJ | 학술논문

针对移动机器人在单目视觉导航方面实时性与鲁棒性较差的问题，提出一种基于 FAST 算法和一点RANSAC算法的移动机器人导航新方法，该方法主要由3个步骤组成：首先，机器人通过自身摄像头采用改进FAST角点提取算法采集周围的环境信息；其次，在匹配过程中采用结合扩展卡尔曼滤波的一点RANSAC算法进行误匹配剔除；最后，用得到的匹配点进行三维环境重建和地图创建。实验结果表明，改进方法能有效提高移动机器人定位与导航的实时性与鲁棒性。
침대이동궤기인재단목시각도항방면실시성여로봉성교차적문제，제출일충기우 FAST 산법화일점RANSAC산법적이동궤기인도항신방법，해방법주요유3개보취조성：수선，궤기인통과자신섭상두채용개진FAST각점제취산법채집주위적배경신식；기차，재필배과정중채용결합확전잡이만려파적일점RANSAC산법진행오필배척제；최후，용득도적필배점진행삼유배경중건화지도창건。실험결과표명，개진방법능유효제고이동궤기인정위여도항적실시성여로봉성。
In order to solve the problem of poor real-time performance and weak robustness of mobile robots during monocular vision navigation , a new improved navigational algorithm is put forward on the basis of the FAST algo-rithm and the one point RANSAC algorithm for mobile robots .This improved new algorithm is made up of three main steps.Firstly, the mobile robots use the FAST algorithm to collect the information by camera .Secondly,in the process of matching , we use the improved RANSAC algorithm to reject the mismatching .Finally, a 3-D map of the environment is built .Experiments show that this method can improve the real-time performance and robustness ef-fectively regarding the aspects of positioning and navigation of the mobile robots .