论文部分内容阅读
针对室内机器人移动导航定位存在的问题,该文研究了基于Kinect传感器的定位算法。该方法应用Kinect传感器采集了移动机器人运动过程中连续帧的彩色和深度信息;通过尺度不变特征变换算法匹配连续场景图像中的特征点对,经随机抽样一致性算法剔除点集中的误匹配点;利用Kinect的深度信息将可用点集的二维坐标转换到相机坐标系下的三维坐标;并使用绝对定向算法计算机器人在相邻位置上的姿态信息和平移量,从而提高了机器人的移动参数的计算精度。通过实验验证了该方法的可行性,且能够满足实时要求,可以同时进行定位与地图创建。
In order to solve the problems existing in the mobile navigation and positioning of indoor robots, this paper studies the positioning algorithm based on Kinect sensor. The Kinect sensor is used to acquire the color and depth information of continuous frames during the motion of mobile robot. Matching feature pairs in continuous scene images is carried out by scaling invariant feature transform algorithm, and the false matching points in the point set are removed by random sampling consistency algorithm ; Using Kinect’s depth information to convert the two-dimensional coordinates of the set of available points to the three-dimensional coordinates under the camera coordinate system; and using the absolute orientation algorithm to calculate the attitude information and the translation amount of the robot at the adjacent positions, thereby improving the movement parameters of the robot The calculation accuracy. The feasibility of this method is verified through experiments, and it can meet the real-time requirements and can locate and create maps at the same time.