RINS-W: Robust Inertial Navigation System on Wheels
This repo contains a real-time approach for inertial navigation based only on an Inertial MeasurementUnit (IMU) for self-localizing wheeled robots. The approach builds upon two components: 1) a robust detector that uses deep neural networks to dynamically detects zero velocity; and 2) a state-of-the-art Kalman filter which incorporates this knowledge along with no lateral slip and vertical velocity as pseudo-measurements for localization.
Our implementation is done in Python a Pytorch for the adapter block of the system. The code was tested under Python 3.5.
Install pytorch. We perform all training and testing on its 1.5 version.
Install required packages, e.g. with the pip3 command
pip3 install requirements.txt
git clone https://github.com/mbrossar/RINS-W.git
Coming soon.
Coming soon.
This repo is mainly based on the paper "RINS-W: Robust Inertial Navigation System on Wheels", International Conference on Intelligent Robots and Systems (IROS), 2019 [IEEE paper, ArXiv paper]. The main differences with the paper are
You can also see also the paper "AI-IMU Dead-Reckoning," IEEE Transactions on Intelligent Vehicles, 2020 [IEEE paper, ArXiv paper].
If you use this code in your research, please cite:
@INPROCEEDINGS{brossard2019,
author={M. {Brossard} and A. {Barrau} and S. {Bonnabel}},
booktitle={2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
title={{RINS-W: Robust Inertial Navigation System on Wheels}},
year={2019},
volume={},
number={},
pages={2068-2075},
}
@ARTICLE{9035481,
author={M. {Brossard} and A. {Barrau} and S. {Bonnabel}},
journal={IEEE Transactions on Intelligent Vehicles},
title={{AI-IMU Dead-Reckoning}},
year={2020},
volume={},
number={},
pages={},
}
Martin Brossard*, Axel Barrau* and Silvère Bonnabel*
*MINES ParisTech, PSL Research University, Centre for Robotics, 60 Boulevard Saint-Michel, 75006 Paris, France