Non-Iterative SLAM: A Fast Dense Method for Inertial-Visual SLAM

The goal of this paper is to create a new framework of Visual SLAM that is light enough for Micro Unmanned Aerial Vehicle (MUAV). Feature-based and direct methods are two mainstreams in Visual SLAM. Both methods minimize photometric or reprojection error by iterative solutions, which are always computationally expensive. To overcome this problem, we propose a non-iterative approach to match point clouds directly. In particular, a new inertial-visual framework is proposed to reduce computational requirement in two steps. First, the attitude and heading reference system (AHRS) and axonometric projection are utilized to decouple the 6 Degree-of-Freedom (DoF) data, so that point clouds can be matched in independent spaces respectively. Second, the matching process is carried out in frequency domain by Fourier transformation, which reduces computational requirements dramatically and provides a closed-form non-iterative solution. In this manner, the time complexity can be reduced to $\mathcal{O}(n \log{n})$ where $n$ is the number of matched points in each frame. To the best of our knowledge, this method is the first non-iterative approach for data association in Visual SLAM. Compared with the state of the arts, it runs at faster speed and obtains 3D maps with higher resolution yet still with comparable accuracy.