Pose Tracking And Path Planning For UAV Based On Multi-Sensor Fusion

In recent years,with the rapid development of computer vision,sensor technology,artificial intelligence and other fields,the technical of UAVs has made great progress,providing security and convenience for the country and society.In order to ensure the safety and reliability of drone flight,the real-time location and path planning technology of UAVs has become more and more important,which has gradually attracted the attention of researchers.Therefore,in the face of the actual needs and new difficult challenges,this dissertation uses multi-sensor fusion technology for pose tracking and designs two three-dimensional space path planning methods.Firstly,aiming at the shortcomings of precision of Simultaneous Localization and Mapping(SLAM)for single sensor,this dissertation uses the Global Navigation Satellite System(GNSS),Inertial Measurement Unit(IMU),and Stereo Visual jointly to construct a multi-sensor fusion pose tracking system.Using GNSS to provide global positioning information for calibration,using IMU to provide carrier motion information for auxiliary positioning,using Stereo Visual to obtain image information for visual pose tracking,giving full play to the complementary advantages between sensors,bringing strong robustness to the system.In addition,this dissertation designs two functional modules of GNSS-IMU and GNSS-IMU-Stereo Visual,and cooperates to complete the tracking of UAV poses.Compared with the traditional visual SLAM method,it has better performance in positioning accuracy.After obtaining the accurate position information of the drone,combined with the three-dimensional environment information acquired by the sensor,the initial environment modeling can be performed.Furthermore,this dissertation uses the combined filter denoising method and the de-redundant map stitching method to complete the construction of the three-dimensional space model,and uses the octree structure to obtain the three-dimensional spatial raster map.On this basis,two path planning methods for UAVs are proposed,which can quickly and robustly complete the path planning task while ensuring the flexibility of UAV movement,further ensuring the safety of UAVs during flight and the efficiency of performing tasks.