Panoramic annular SLAM with loop closure and global optimization

TL;DR

This paper introduces PA-SLAM, a visual SLAM system using panoramic annular lenses that improves loop closure detection and reduces error through global optimization, achieving high accuracy and robustness.

Contribution

The paper presents a novel hybrid point selection strategy and a complete SLAM pipeline tailored for panoramic annular lenses, enhancing loop closure detection and global consistency.

Findings

Reliable loop closure detection with hybrid point strategy

Significant reduction in accumulated error and scale drift

Achieves state-of-the-art accuracy in panoramic SLAM

Abstract

In this paper, we propose panoramic annular simultaneous localization and mapping (PA-SLAM), a visual SLAM system based on panoramic annular lens. A hybrid point selection strategy is put forward in the tracking front-end, which ensures repeatability of keypoints and enables loop closure detection based on the bag-of-words approach. Every detected loop candidate is verified geometrically and the $Sim(3)$ relative pose constraint is estimated to perform pose graph optimization and global bundle adjustment in the back-end. A comprehensive set of experiments on real-world datasets demonstrates that the hybrid point selection strategy allows reliable loop closure detection, and the accumulated error and scale drift have been significantly reduced via global optimization, enabling PA-SLAM to reach state-of-the-art accuracy while maintaining high robustness and efficiency.