Skip to content

Latest commit

 

History

History
132 lines (91 loc) · 3.81 KB

README.md

File metadata and controls

132 lines (91 loc) · 3.81 KB
Raw

Pytorch UWB Localization

Official page of RONet, which is published @IROS'19

Since the original code is based on Tensorflow, now I have ported the original algorithm to PyTorch.

before

ToDo

  • Port RONet
  • Port Bi-LSTM
  • Run on test data
  • Set training pipeline
  • Visualize training procedure
  • Autosave the best model

Environments

（1）install anaconda3 or miniconda

conda create -n env_name python=3.6
conda activate env_name

（2）install pytorch

because ubuntu20.04's support version need CUDA >=11.0 and cudatoolkit >= 11.0, so the pytorch's mini is cudatoolkit=11.0 and pytorch==1.7.0

conda install pytorch==1.7.0 torchvision==0.8.0 torchaudio==0.7.0 cudatoolkit=11.0 -c pytorch

（3）install other libs

Please refer to requirements.txt to install other libs

conda install matplotlib==3.1.3

（4）For instance Training

python3 main.py --arch LSTM --gpu 0 -b 1000

Descriptions

What is the UWB sensor?

UWB is abbv. for Ultra-wideband, and the sensor outputs only 1D range data.

validation

More explanations are provided in this paper.

In summary, UWB data are likely to be vulnerable to noise, multipath problems, and so forth.

Thus, we leverage the nonlinearity of deep learning to tackle that issue.

Data

All data are contained in uwb_dataset and a total of eight sensors are deployed, whose positions are as follows:

idnames

And each csv consists N (the num. of sequences) x 10 whose columns denotes:

range @id0, range @id1, range @id2, range @id3, range @id4, range @id5, range @id6, range @id7, x of GT, y of GT

Note that our experiment was conducted on real-world data by using Pozyx UWB sensors and the motion capture system.

(Please kindly keep in mind that Pozyx systems do not give precise range data :( )

Training

The training scripts come with several options, which can be listed with the --help flag.

python3 main.py --help

The point is that it only takes a few minutes because the data of UWB are lightweight and simple! :)

Training results will be saved under the results folder. To resume a previous training, run

python3 main.py --resume [path_to_previous_model]

Validation

python3 main.py --evaluate [path_to_trained_model]

Benchmark

On validation data

Methods RMSE (cm)
RNN 4.050
GRU 3.918
LSTM 4.855
Bi-LSTM TBA
RONet TBA

Citation

  • If you use our code or method in your work, please consider citing the following::
@INPROCEEDINGS {lim2019ronet,
  author = {Lim, Hyungtae and Park, Changgue and Myung, Hyun},
  title = {Ronet: Real-time range-only indoor localization via stacked bidirectional lstm with residual attention},
  booktitle = {Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages={3241--3247},
  year = { 2019 },
  organization={IEEE}
}
@INPROCEEDINGS {lim2018stackbilstm,
  author = {Lim, Hyungtae and Myung, Hyun},
  title = {Effective Indoor Robot Localization by Stacked Bidirectional LSTM Using Beacon-Based Range Measurements},
  booktitle = {International Conference on Robot Intelligence Technology and Applications},
  pages={144--151},
  year = { 2018 }
  organization={Springer}
}

Contact

Contact: Hyungtae Lim ([email protected])

Please create a new issue for code-related questions. Pull requests are welcome.