基于Kinect传感器的室内机器人自定位研究

针对同步定位地图创建（SLAM）中的定位问题，提出一种基于Kinect传感器的室内机器人自定位方法。利用Kinect传感器采集图像信息和深度信息，获取场景图像的SIFT匹配特征点集，并将特征点集的像素和深度信息转为三维数据，利用ICP算法计算机器人相邻位姿的旋转矩阵与平移向量。采用RANSAC方法去除SIFT匹配外点，提高了运动参数的计算精度。通过实验验证该方法的可行性。

Study of Self-localization of Indoor Robot Based on Kinect Sensor

In order to solve the localization problem in the simultaneous localization and mapping （ SLAM）, this paper proposes a method of indoor robot＇ s self-localization based on Kinect sensor. Vision data and depth data captured by Kinect sensor are used to get SIFT matching point set from the environment pictures and reconstruct, the matching points＇ 3D data then the robot＇ s movement parameters of rotation matrix and transport vector are estimated by ICP algorithm and the outliers are removed by RANSAC method to improve the accuracy of the measurement.Experimental results indicates the feasibility of the proposed method.