Deep Weakly Supervised Positioning
Abstract
PoseNet can map a photo to the position where it is taken, which is appealing in robotics. However, training PoseNet requires full supervision, where ground truth positions are non-trivial to obtain. Can we train PoseNet without knowing the ground truth positions for each observation? We show that this is possible via constraint-based weak-supervision, leading to the proposed framework: DeepGPS. Particularly, using wheel-encoder-estimated distances traveled by a robot along random straight line segments as constraints between PoseNet outputs, DeepGPS can achieve a relative positioning error of less than 2%. Moreover, training DeepGPS can be done as auto-calibration with almost no human attendance, which is more attractive than its competing methods that typically require careful and expert-level manual calibration. We conduct various experiments on simulated and real datasets to demonstrate the general applicability, effectiveness, and accuracy of DeepGPS, and perform a comprehensive analysis of its robustness. Our code is available at https://ai4ce.github.io/DeepGPS/.
Cite
@article{arxiv.2104.04866,
title = {Deep Weakly Supervised Positioning},
author = {Ruoyu Wang and Xuchu Xu and Li Ding and Yang Huang and Chen Feng},
journal= {arXiv preprint arXiv:2104.04866},
year = {2021}
}
Comments
8 pages, 8 figures, submitted to IEEE Robotics and Automation Letters (RA-L) and 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021)