This project contains the source-code of the prototype that was implemented for the master's thesis "Using Mobile Edge Computing Technologies for Real-Time Cornering Assistance" at the TU Wien. If you use the source code, pls. make the following citations:
- Hong-Linh Truong, Matthias Karan, "Analytics of Performance and Data Quality for Mobile Edge Cloud Applications", Preprint PDF, Mar 2018.
- Matthias Karan, "Using Mobile Edge Computing Technologies for Real-Time Cornering Assistance", Master Thesis, TU Wien, Feb 2018.
To contribute to safer driving in any type of car, in the thesis "Using Mobile Edge Computing Technologies for Real-Time Cornering Assistance" we introduce a novel system that assists drivers in real-time while cornering. The system is designed in a way that it can be deployed to both the cloud and/or the emerging edge-computing infrastructure.
Goals:
- warn drivers ahead of curves in real-time
- recommend safe speeds to enter a curve
- use novel edge- and cloud-computing architectures and algorithms
See the poster for more information about the thesis.
The following videos demonstrate the prototype application:
See Demo - README to run the demo yourself.
The software components of the thesis' prototype are split up into the following directories:
android: A native android prototype that shows upcoming curves and recommends a safe speed live on the road.
commons: A Java library that contains common code used in the services (recommendation and detection).
detection: Source code for the detection application written in Apache Apex
docker: Docker configurations for deploying the prototype, run experiments and a simple demo (described below).
recommendation: Source code for the recommendation service.
simulator: Java applications that simulate client/car functionalities.