In the autonomous flight of unmanned aerial vehicles (UAVs), real-time acquisition of its pose information is the basis for navigation and control. For the pose estimation of UAVs, this paper proposes a RANSAN algorithm based on prior information to implement the pose estimate of UAVs. This method firstly uses the SURF algorithm to process the sequence images acquired by the UA V in different angles of the same target area in the actual flight to achieve the extraction and matching of feature points between the images. With the assistance of the pose information provided by the GPS and IMU systems, the RANSAC algorithm combined with the five-point algorithm is used to obtain the corresponding pose information of the UA V at each moment. Experiments show that this method is more accurate than simply using visual information or GPS and IMU system to realize the pose estimation of UA V. It can meet the needs of the actual projects within the allowable range of error, and can enrich the pose estimation theory of UA V to some extent.
UAV Pose Estimation Based on Prior Information and RANSAC Algorithm
01.06.2018
967378 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Image Matching Algorithm Based on Improved FAST and RANSAC
British Library Conference Proceedings | 2021
|Preemptive RANSAC for Live Structure and Motion Estimation
British Library Conference Proceedings | 2003
|Lane detection based on guided RANSAC
Tema Archiv | 2010
|Preemptive RANSAC for live structure and motion estimation
British Library Online Contents | 2005
|