针对移动机器人自主导航过程实时准确进行三维重建问题,提出了一种基于人工势场法和三维NDT算法融合进行三维重建。移动机器人使用人工势场法实时规划行动路径,实时获取机器人初始变换矩阵。通过Kinect获取环境点云图,将三维NDT算法和初始变换矩阵进行融合完成对点云三维重建。实验结果表明,该算法相比较ICP和传统三维NDT算法具有更高的效率。
A 3D reconstruction method based on the fusion NDT and artificial potential field algorithm is pro- posed for real-time and accurate three-dimensional reconstruction during autonomous mobile robot navigation. The mobile robot uses the artificial potential field method in real time plan paths, and obtains the environment point cloud image by Kinect to complete the three-dimensional reconstruction by the fusion of NDT method and initial transforma- tion matrix. The experimental results show that the proposed algorithm is more efficient than the ICP and the tradi- tional three-dimensional NDT method.