Cities are experiencing unparalleled growth, facing new challenges as they endeavor to become healthier, more sustainable and safer places for citizens to live and work. One of such challenges arise in urban mobility and related fields like transport planning and traffic engineering, which are evolving to embrace the myriad of new opportunities that mobile personal devices, together with data science, may bring to citizens. Traffic engineering, for instance, has traditionally been based on stationary sensor data like vehicle counters, induction loops and more recently thermo-cameras to understand core traffic characteristic parameters including flow speed, street saturation and the distribution of cars, trucks, and pedestrian participants. With the ubiquity of mobile personal devices (e.g., wearables, smartphones) and the development of machine learning (ML), however, new opportunities have emerged to collect and analyze mobility data at the granularity of a driver or a pedestrian.
This kind of personal data, which is composed of the geolocated registers left by devices that can be considered to be attached to a single individual, can provide valuable information to understand mobility and traffic patterns and enable new kinds of analytics like origin-destination flow, detecting critical traffic junctions, heavily frequented transfer areas and road segments, estimate street maintenance and parking needs, and many other applications. With this information, for example, a city may seek to better understand the current mobility flow patterns and design an effective strategy for managing traffic and reduce the associated pollution.
The MOBILYTICS project focuses on such a scenario in which multiple private companies and public entities may be collecting, through their respective apps and services, mobility data of their users and may be interested, for example, in acquiring data from other companies/entities to complement theirs, in releasing the data for the public good or research studies, or they may wish to monetize their data through other means.
Anonymizing a microdata (i.e., a database containing records at the individual level) faces two opposing objectives that need to be harmonized. On the one hand, the privacy of the individuals appearing in the microdata must be guaranteed, and this is typically achieved by modifying the original data. On the other hand, the utility of the released data must be preserved, that is, analyses on the anonymized data should not yield results that are too different from those on original data. The field of statistical disclosure control (SDC), also known as anonymization, provides three main techniques to find a suitable trade-off between privacy and utility: (i) non-perturbative masking (methods that reduce the level of detail of the original data while preserving their truthfulness); (ii) perturbative masking (methods that modify the original data without preserving their truthfulness); and (iii) synthetic data (methods that build a new data set from scratch based on a model of the original data).
The development of a more sustainable mobility, the core goal of MOBILYTICS, has numerous immediate benefits to our society as a whole, such as reducing the number of polluting vehicles in big cities, increasing the use of green or alternative public transport services, which ultimately may help reduce environmental impact and atmospheric pollution, and improve people's daily lives.Read More... Read Less...
MobilitApp is a tool to help to analyse the urban mobility in our cities. MobilitApp periodically records 3 sensors (accelerometer, magnetometer and gyroscope) from the citizens' smartphones during their multimodal trips. The information obtained is processed using deep learning techniques to predict the transportation mode used by the user, i.e. bicycle, bus, car, e-bicycle, e-scooter, metro, motorbike, run, stationary, train, tram and walk. The user's information is treated completely anonymously.
Copyright © 2023 MOBILYTICS . Template by W3layouts