导航

ACTA AERONAUTICAET ASTRONAUTICA SINICA

    Next Articles

Heterogeneous collaborative SLAM based on fisheye and RGBD camera

  

  • Received:2022-06-14 Revised:2022-09-23 Online:2022-09-30 Published:2022-09-30
  • Supported by:
    National Natural Science Foundation of China

Abstract: SLAM technology has a broad application prospect in autonomous navigation in satellite navigation denied envi-ronments. To combine the advantages of monocular fisheye SLAM to obtain more texture information and RGBD-SLAM to directly obtain scale information, a heterogeneous collaborative SLAM system based on a monocular fisheye camera and a RGBD camera is designed. Firstly, a method for checking the 3D gray centroid direction con-sistency between feature points is designed to screen candidate matching points between heterogeneous images. Then, a step-by-step optical flow and projection matching method between heterogeneous images is designed to achieve high-performance feature point matching and relative pose estimation between fisheye and RGBD camera. Finally, based on ORB-SLAM2 framework, a heterogeneous collaborative SLAM system based on fisheye and RGBD camera is proposed. The experimental results show that, compared with the traditional feature point match-ing methods, the proposed feature point matching method shows higher performance in the image feature match-ing task for heterogeneous cameras. Compared with monocular fisheye SLAM and RGBD-SLAM system, the pro-posed heterogeneous collaborative SLAM system has better performance under the conditions of rapid camera movement, camera close to the scene, low frame rate, texture loss, pure rotation of the camera, outdoor large scenes, etc. its robustness, anti-drift ability and trajectory accuracy are greatly improved compared with mo-nocular fisheye SLAM and RGBD-SLAM system.

Key words: Heterogeneous collaborative SLAM, Heterogeneous image feature matching, Fisheye camera, RGBD camera, Candidate matching points filtering

CLC Number: