Monocular SLAM for Visual Odometry
ABSTRACT The ego-motion online estimation process from a video input is often called visual odometry. Typically optical flow and structure from motion (SFM) techniques have been used for visual odometry. Monocular simultaneous localization and mapping (SLAM) techniques implicitly estimate camera ego-motion while incrementally build a map of the environment. However in monocular SLAM, when the number of features in the system state increases, the computational cost grows rapidly; consequently maintaining frame rate operation becomes impractical. In this paper monocular SLAM is proposed for map-based visual odometry. The number of features is bounded removing features dynamically from the system state, for maintaining a stable processing time. In the other hand if features are removed then previous visited sites can not be recognized, nevertheless in an odometry context this could not be a problem. A method for feature initialization and a simple method for recovery metric scale are proposed. The experimental results using real image sequences show that the scheme presented in this paper is promising.
- SourceAvailable from: Norrima Mokhtar[show abstract] [hide abstract]
ABSTRACT: In a visual odometry system, location of a mobile robot is automatically estimated (localized) from video. When the video is captured by an "upward" camera fixed to an indoor mobile robot, a panorama image of the ceiling (ceiling map) is generated by using a visual motion between two adjacent frames in the video. Similarly, location of another robot can be estimated on the ceiling map by using a visual motion between the current frame and the previously generated ceiling map. Under the assumption that the robot goes straight or rotates around a fixed point, there is no problem on the localization as far as the floor is flat. However, when there is debris on the floor, the estimated location contains error. In this paper, we reduce this error by utilizing visual motions in video from the "forward" camera fixed to the robot. This is a visual compensation of motions in the "upward" camera's video with those in the "forward" camera's video. It was experimentally confirmed that the maximum absolute value of the error was reduced to approximately 11%.Scientific research and essays 02/2011; 6:131-135. · 0.32 Impact Factor
Conference Proceeding: Camera-aided inertial navigation using epipolar points[show abstract] [hide abstract]
ABSTRACT: Due to the rapid error growth of navigation systems using low-cost inertial measurement units there is a need to fuse the information with complementary sensors. In this paper a monocular camera is used to aid the system. Unlike SLAM-like approaches the problem of estimating the location of each feature point viewed in a scene is avoided, instead estimated epipolar points on the image plane are used. By maintaining a buffer of past views, the method mimics a short-term visual memory which imposes multiple constraints on the estimation problem. The result is a Sigma-Point Kalman filter in square-root form with a linear and efficient time-update. A simulation study is presented indicating the filter's capacity to constrain the rate of error growth of an inertial navigation system. The filter may also find useful applications when fusing with additional sensors.Position Location and Navigation Symposium (PLANS), 2010 IEEE/ION; 06/2010
- [show abstract] [hide abstract]
ABSTRACT: In the SLAM (Simultaneous Localization and Mapping) technology, an environmental map is generated by a mobile robot. When it results in failure, it is necessary to inspect the scene. This localization and browsing require transmission of video signal to a remote place. In the system in this paper, an indoor mobile robot has two cameras. One is the "upward" which captures scenery of ceiling. The other is "forward" for scenery in front of the robot. Video signals from the cameras are encoded and transmitted from the robot to a remote server. It causes a problem that data size is too huge to be transmitted. To cope with this problem, the Functionally Layered Coding (FLC) was reported. In the existing FLC, visual motions are estimated by using the rotation invariant phase only correlation (RI-POC) technique. It can estimate two kinds of motions -translation and rotation. However, it requires doubled computational complexity and many components to be transmitted. In this paper, we analyze relation between kinetic movements of a robot and visual motions observed in videos, and propose to replace RI-POC by a simple POC. It was confirmed that the proposed method reduced data size for transmission to 61.6%.International Journal of the Physical Sciences. 01/2011; 5:2652-2657.