Vision-based global localization and mapping for mobile robots

Abstract

Abstract-We have previously developed a mobile robot system which uses scale-invariant visual landmarks to localize and simultaneously build three-dimensional (3-D) maps of unmodified environments. In this paper, we examine global localization, where the robot localizes itself globally, without any prior location estimate. This is achieved by matching distinctive visual landmarks in the current frame to a database map. A Hough transform approach and a RANSAC approach for global localization are compared, showing that RANSAC is much more efficient for matching specific features, but much worse for matching nonspecific features. Moreover, robust global localization can be achieved by matching a small submap of the local region built from multiple frames. This submap alignment algorithm for global localization can be applied to map building, which can be regarded as alignment of multiple 3-D submaps. A global minimization procedure is carried out using the loop closure constraint to avoid the effects of slippage and drift accumulation. Landmark uncertainty is taken into account in the submap alignment and the global minimization process. Experiments show that global localization can be achieved accurately using the scale-invariant landmarks. Our approach of pairwise submap alignment with backward correction in a consistent manner produces a better global 3-D map.

Sub-goal based Robot Visual Navigation through Sensorial Space Tesselation

“…Landmark-based localization methods rely on the assumption that landmarks can be detected and accurately interpreted from raw sensor readings [2], [5]. However interpretation from sensor readings to accurate geometric representation is complex and error prone.…”

RAFI - A Stereo Vision Based Autonomous Mobile Area Mapping Robot with 16 DOF

“…Stephen Se et al proposed vision based simultaneous localization and mapping by tracking SIFT (Scale Invariant Feature Transform) features [15]. Our approach is to make 3D submaps tracking SURF features and recover depth.…”

On-line SLAM using clustered landmarks with omnidirectional vision

“…The SIFT algorithm has become very popular in several robotics applications, as it can be seen in Se et al (2001), Se et al (2005), Ledwich and Williams (2004), and introduces several properties of invariance that are specially useful when extracting features directly from omnidirectional images, as it is the case in this work. Rotational invariance is important because detected objects can appear in any orientation depending on the angle between them and the robot, and so is scale invariance since resolution rapidly decreases in the outer ring of the image, changing the apparent size of observed objects.…”