We present a novel approach to robust pose graph optimization based on Graduated Non-Convexity (GNC). Unlike traditional GNC-based methods, the proposed approach employs an adaptive shape function using B-spline to optimize the shape of the robust kernel. This aims to reduce GNC iterations, boosting computational speed without compromising accuracy. When integrated with the open-source riSAM algorithm, the method demonstrates enhanced efficiency across diverse datasets. Accompanying open-source code aims to encourage further research in this area. https://github.com/SNU-DLLAB/AGNC-PGO
@article{arxiv.2308.11444,
title = {Adaptive Graduated Non-Convexity for Pose Graph Optimization},
author = {Seungwon Choi and Wonseok Kang and Jiseong Chung and Jaehyun Kim and Tae-wan Kim},
journal= {arXiv preprint arXiv:2308.11444},
year = {2023}
}
Comments
4 pages, 3 figures. Accepted for the workshop on Robotic Perception and Mapping(ROPEM): Frontier Vision & Learning Techniques, organized at the 2023 International Conference on Intelligent Robots and Systems (IROS)