• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Saint Petersburg International Conference on Integrated Navigation Systems; 20070528-30; Saint Petersburg(RU) >SIMULTANEOUS LOCALIZATION AND MAPPING USING SONAR AND VISION SENSORS IN DYNAMIC ENVIRONMENTS
【24h】

SIMULTANEOUS LOCALIZATION AND MAPPING USING SONAR AND VISION SENSORS IN DYNAMIC ENVIRONMENTS

机译：在动态环境中使用声纳和视觉传感器同时定位和映射

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

An algorithm for Simultaneous Localization and Mapping (SLAM) using sonar and vision sensors in dynamic environments has been suggested and is considered in this paper. The algorithm is capable of differentiating static and dynamic objects in the environment. There are three parts in the algorithm: localization, mapping and dynamic objects detection. Two occupancy grids are maintained. One models the static parts of the environment, and the other models the dynamic parts. At the same time, we detect the dynamic objects by three-frame differencing and adaptive background subtraction technique. Two grid maps are updated by previous static map and the observations information from sonar and are emendated by the detected dynamic objects using monocular camera. The emendated static map is used to do SLAM based on the Rao-Blackwellized particle filter. The vision information can get rid of the mistakes that some dynamic objects are considered as static ones and help SLAM with the right landmarks. The results of experimental test of the algorithm are given.
机译：提出了一种在动态环境中使用声纳和视觉传感器的同时定位和制图（SLAM）算法。该算法能够区分环境中的静态和动态对象。该算法分为三个部分：定位，映射和动态对象检测。维护两个占用网格。一个模型为环境的静态部分建模，另一个模型为动态部分。同时，我们通过三帧差分和自适应背景减法技术检测动态物体。两个网格图由以前的静态图和来自声纳的观测信息更新，并由使用单眼相机检测到的动态对象进行修正。改进的静态图用于基于Rao-Blackwellized粒子滤波器进行SLAM。视觉信息可以消除一些动态对象被视为静态对象的错误，并帮助SLAM确定正确的地标。给出了该算法的实验测试结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号