...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Systems Science & Control Engineering >Improved grid mapping technology based on Rao-Blackwellized particle filters and the gradient descent algorithm
【24h】

Improved grid mapping technology based on Rao-Blackwellized particle filters and the gradient descent algorithm

机译：基于RAO-Blackwellized粒子滤波器和梯度下降算法的改进的网格映射技术

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

TRecently, the Rao-Blackwellized particle filter (RBPF) has been used to solve the problem of simultaneous localization and mapping (SLAM). Using the odometer information of robot to calculate the proposed distribution requires a number of sampled particles, which increases the calculation complexity in the RBPF operation. In this paper, we integrate the odometer measurement and sensor observation into the proposed distribution, effectively reducing the particle sample scale. To reduce the inconsistency in the map model caused by the cumulative error of the odometer information of robot, we applied a gradient descent algorithm to fuse the sensor data to obtain the real-time attitude angle. This combination method, based on the robot operation system (ROS), runs on a platform of self-built mobile robot equipped with a laser rangefinder. The experimental results show that this method can realize the online real-time high-precision grid map which provides a new approach for robot navigation and SLAM.
机译：超然，RAO-Blackwellized粒子过滤器（RBPF）已被用于解决同时定位和映射（SLAM）的问题。使用机器人的里程表信息来计算所提出的分布需要许多采样粒子，这增加了RBPF操作中的计算复杂度。在本文中，我们将里程表测量和传感器观察整合到所提出的分布中，有效地减少了粒子样本量表。为了减少由机器人的里程表信息的累积误差引起的地图模型中的不一致，我们应用了梯度下降算法来熔化传感器数据以获得实时姿态角度。基于机器人操作系统（ROS）的这种组合方法在配备有激光测距仪的自制移动机器人的平台上运行。实验结果表明，该方法可以实现在线实时高精度网格图，为机器人导航和猛击提供了一种新的方法。

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