迭代最近点
里程计
兰萨克
机器人
移动机器人
遥感
图像配准
里程表
扩展卡尔曼滤波器
作者
Marina Aguilar-Moreno,Manuel Graña
标识
DOI:10.1007/978-3-030-61705-9_3
摘要
Simultaneous localization and mapping (SLAM) is process highly relevant for autonomous systems. Accurate sensing provided by range sensors such as the M8 Quanergy LiDAR improves the speed and accuracy of SLAM, which can become an integral part of the control of innovative autonomous cars. In this paper we propose a hybrid point cloud registration method that profits from the high accuracy of classic iterated closest points (ICP) algorithm, and the robustness of the Normal Distributions Transform (NDT) registration method. We report positive results in an in-house experiment encouraging further research and experimentation.
科研通智能强力驱动
Strongly Powered by AbleSci AI