首页> 外文会议>IEEE Chinese Guidance, Navigation and Control Conference >The use of optical flow for UAV motion estimation in indoor environment
【24h】

The use of optical flow for UAV motion estimation in indoor environment

机译:使用光学流量在室内环境中的UAV运动估计

获取原文

摘要

To solve the navigation problem of unmanned aerial vehicles (UAV) in GPS-degraded or GPS-denied environment, we realize optical-flow-based motion estimation for UAV by using the camera, inertial measurement devices, altimeter and other sensors. Detailed comparisons are made among several representative methods of computing optical flow field in this paper, including differential method, block matching method and feature-based method. After that, we use Lucas-Kanade method based on SURF features to determine optical flow from video sequences, and apply forward-backward error which is aimed to make tracking more reliable and improve the accuracy of features matching. Finally, depended on pin-hole image plane approach, we derive the motion field estimation. By comparing with the experimental results of PX4FLOW, we find that the method used in this paper is feasible and it can achieve velocity estimation with high accuracy in indoor environment.
机译:为了解决GPS降级或GPS拒绝环境中无人机(UAV)的导航问题,我们通过使用相机,惯性测量装置,高度计和其他传感器来实现无人机的基于光流的运动估计。 本文在计算光学流场的几种代表性方法中进行了详细的比较,包括差分方法,块匹配方法和基于特征的方法。 之后,我们使用基于冲浪功能的Lucas-Kanade方法来确定来自视频序列的光流,并应用前后误差,旨在使跟踪更可靠,提高具有匹配功能的准确性。 最后,依赖于引脚孔图像平面方法,我们推导出运动场估计。 通过与PX4Flow的实验结果进行比较,我们发现本文中使用的方法是可行的,并且可以在室内环境中实现高精度的速度估计。

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