Thesis in external company
keywords BEHAVIORAL MODELING, DATA ANALYSIS, MOBILITY, SENSOR FUSION
Reference persons GIUSEPPE CARLO CALAFIORE, ALESSANDRO RIZZO
External reference persons Dr. Pino Castrogiovanni, Dr. Claudio Borean, TIM SWARM Joint Open Lab
Research Groups SYSTEMS AND DATA SCIENCE - SDS
Thesis type EXPERIMENTAL/MODELLING
Description The thesis activities are in the framework of the experimentation of novel solutions aimed at changing the mobility patterns of people, leveraging ICT technologies and the Internet of Things (IoT), to encourage the adoption of sustainable behaviors by individuals and to improve the overall mobility experience and the environmental impact in cities.
The thesis goal is the design and prototyping of a system able to produce "fingerprints" of people movement patterns, automatically revealing their paths and the transportation means used. The system will constitute the basis for the development of a system of incentives of virtuous behaviors. To this aim, the system should guarantee an adequate security level, to prevent malicious alteration of the data by the users.
The thesis work will start from the study of sensory data coming from a smartphone (e.g. GPS and/or accelerometer), to evaluate their potential to be processed in order to provide such fingerprints, and to assess how the assets of a mobile communication provider (localization fingerprints from the mobile network, or Call Detail Records - CDR) can be used to improve the overall system reliability and security. On the basis of the results of this analysis, a first Android app will be developed and prototyped, in order to identify the paths and transportation means utilized by the users, using only the sensors on the smartphone and some cloud-based software components to process and certify the collected data.
The prototype will be validated on real users. Then, a second version of the app will be designed and prototyped. This version will make use of additional IoT devices (e.g. wearable devices, other devices installed on cars or bikes) to improve the reliability of the system or to extend the set of behaviors that can be certifiable (e.g. automatic identification of the number of people present in the same care in a car-pooling application).
Finally, the second prototype will be validated and, based on the obtained results, a further experimentation phase will be considered in the framework of the initiative "Torino Living Lab" (http://torinolivinglab.it)
Required skills - Great coding skills
- Good knowledge of Java and Android
- Flexibility/proactivity, ability in problem solving and availability to work in team
Deadline 28/11/2016 PROPONI LA TUA CANDIDATURA