This repository contains an Android implementation for the IMUNet paper.
It has three parts:
1- An application for collecting a new dataset that uses the AR-Core API for collecting the ground truth using the SLAM techniques and IMU measurements. The modification has been implemented on the code provided here and the Tango part for collecting the ground truth trajectory has been replaced with AR-Core API that makes every android user able to collect a new dataset.
2- The test part of the RONIN for the ResNet18 model using all the proposed models and some samples of the collected dataset has been implemented on Android. Samples can be downloaded here and must be put in the raw folder.
3- A comparison has been implemented to show the efficiency and accuracy of the proposed model. The result can be seen in the video below:
@article{zeinali2024imunet, title={IMUNet: Efficient Regression Architecture for Inertial IMU Navigation and Positioning}, author={Zeinali, Behnam and Zanddizari, Hadi and Chang, Morris J}, journal={IEEE Transactions on Instrumentation and Measurement}, year={2024}, publisher={IEEE} }