The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-3/W1
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W1, 219–224, 2017
https://doi.org/10.5194/isprs-archives-XLII-3-W1-219-2017
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W1, 219–224, 2017
https://doi.org/10.5194/isprs-archives-XLII-3-W1-219-2017

  25 Jul 2017

25 Jul 2017

A ROBUST METHOD FOR STEREO VISUAL ODOMETRY BASED ON MULTIPLE EUCLIDEAN DISTANCE CONSTRAINT AND RANSAC ALGORITHM

Q. Zhou1, X. Tong1, S. Liu1, X. Lu2, S. Liu1, P. Chen1, Y. Jin1, and H. Xie1 Q. Zhou et al.
  • 1College of Surveying and Geo-Informatics, Tongji University, Shanghai, China
  • 2China International Engineering Consulting Corporation, Beijing, China

Keywords: Visual Odometry, Stereo Vision, Robot Navigation, RANSAC Algorithm

Abstract. Visual Odometry (VO) is a critical component for planetary robot navigation and safety. It estimates the ego-motion using stereo images frame by frame. Feature points extraction and matching is one of the key steps for robotic motion estimation which largely influences the precision and robustness. In this work, we choose the Oriented FAST and Rotated BRIEF (ORB) features by considering both accuracy and speed issues. For more robustness in challenging environment e.g., rough terrain or planetary surface, this paper presents a robust outliers elimination method based on Euclidean Distance Constraint (EDC) and Random Sample Consensus (RANSAC) algorithm. In the matching process, a set of ORB feature points are extracted from the current left and right synchronous images and the Brute Force (BF) matcher is used to find the correspondences between the two images for the Space Intersection. Then the EDC and RANSAC algorithms are carried out to eliminate mismatches whose distances are beyond a predefined threshold. Similarly, when the left image of the next time matches the feature points with the current left images, the EDC and RANSAC are iteratively performed. After the above mentioned, there are exceptional remaining mismatched points in some cases, for which the third time RANSAC is applied to eliminate the effects of those outliers in the estimation of the ego-motion parameters (Interior Orientation and Exterior Orientation). The proposed approach has been tested on a real-world vehicle dataset and the result benefits from its high robustness.