Dead Reckoning

pedestrian dead reckoning using phone sensors.

Indoor positioning method using Pedestrian Dead Reckoning (PDR) based on phone sensors data: Accelerometer, Magnetometer and Gyroscope.

I implemented and used this localization technique for my internship at Continental: Implementation and optimization of a localization system for connected vehicles.

_For more details about the PDR, please refer to this Article or this one.

_For more details about my internship, please email me at [email protected] [email protected]

Tools

• Phyphox
• Android Studio

Demos

In the following example, I made a complete turn in a hallway, i.e. a looped path of 116m and 168 steps.

• Tracking result using PDR:

• Length estimation per step:

Installing the requirements

This project was developed in python.

To install all the packages used in this project:

pip install -r ./requirements.txt

Uploading data sensors

Upload the csv files of your data sensors on ./data/ file:

• Accelerometer.csv
• Magnetometer.csv
• Gyroscope.csv
• Location.csv

Usage

To display sensors data on graphs:

python ./src/plot.py

To display compute steps methods on graphs:

python ./src/computeSteps.py

To display length estimation per step:

python ./src/computeLength.py

To display tracking:

python ./src/deadreckoning.py