Cognitive Driving with SIMPLI-CITY Mobile App - overview
Cognitive Driving with SIMPLI-CITY Mobile App is an application which looks at driver anxiety, traffic, wearable and calendar data to suggest personalized routes.
Team: Freddy Lecue, Mark Purcell, Martin Stephenson, Michael Barry, Randy Cogill, Rodrigo Ordóñez (University College Dublin), Joe Naoum-Sawaya (Ivey Business School, University of Western Ontario)
Mobile application developed as part of EU SIMPLI-CITY Project:
Our cognitive driving application was developed as part of an EU funded project called “SIMPLI-CITY- The Road User Information System of the Future” which fosters the usage of full-fledged road user information systems – helping drivers to make their journey safer, more comfortable, and more environmentally friendly. This 3 year project concluded in November 2015 with a live testing of the prototype. IBM was responsible for data integration, data processing and reasoning (explaining and predicting traffic condition) and cognitive driving functionalities (personalised driver experience through the use and interpretation of open, government, real-time, sensor, user and body data / information). IBM also developed this Android mobile application for personalised driving experience as part of the project.
The scope of the prototype for this data processing and contextual driving experience:
-
Diagnosing road traffic conditions related to User-Centric and Open Data
-
Predicting road traffic condition using Open Data and User-Centric Data sources
-
Capturing user contexts and car sensor contexts for personalized navigation
Personalised driving experience by analysing user behaviour when driving:
The SIMPLI-CITY mobile app provides cognitive mobility enabling a new generation of vehicles to recommend (and justify) personalized routes based on analysis and interpretation of open data from real-time traffic and various IoT devices (e.g., weather station), (ii) social data from tweets feeds, (iii) driver-related data such as her/his body information (e.g., anxiety) from wearables and also (iv) calendar data. The application will then suggest personalized routes that fit drivers' ability while ensuring safer traffic for other vehicles.
Diagnosing road traffic conditions related to User-Centric and Open Data
The User-Centric and Open Data Management component of SIMPLI-CITY handles the Open Data sources, which are available to SIMPLI-CITY such as traffic and weather information as well as data, provided by the end-user such as calendar information from Google calendar, or profile information, which indicates preferences. Both the Open Data and User-Centric data feed into the SIMPLI-CITY Unified Data Model and are used in the Data Processing component where all available data (including sensor data in the final prototype) are transformed into the Unified Data Model.
This allows contextualization and reasoning over heterogeneous datasets. While the Open Data is fetched directly within the User-Centric and Open Data Management component, the User-Centric Data is accessed via the Sensor Abstraction and Interoperability Interfaces. The Sensor Abstraction and Interoperability Interfaces provide a Java interface on the Personal Mobility Assistant (PMA) side as well on the server side to directly access the user data. In addition, to provide external SIMPLI-CITY server components with access to the user data, a RESTful interface is provided on the server side. This component is then transformed into the SIMPLI-CITY Unified Data Model in the Data Processing component. This allows for contextualizing over the data, i.e., finding relationships between user locations, weather and traffic conditions for the relevant region and to perform reasoning, predictive and diagnosis tasks.
Within the application a GEO Services component binds information from Google Maps API with related anomalies found bySTAR-CITY, a semantic reasoning system developed within the SIMPLi-CITY project. STAR-CITY provides RESTful APIs for access to accurate and consistent detection, analysis, diagnosis, exploration and prediction of road traffic conditions and anomalies.
Prototype Sensor Technology included:
The following sensors integrated with the SIMPLI-CITY mobile app prototype are:
Car Sensor: An ELM 327 Interface
The ELM327 is a programmed microcontroller produced by ELM Electronics for translating the on-board diagnostics (OBD) interface found in most modern cars. The ELM327 command protocol is one of the most popular PC-to-OBD interface standards and is also implemented by other vendors.
Human Sensor:
An Empatica E4 wearable wristband was used to gather the data from the individual to monitor physiological signals in real-time.
The wristband has the following sensors: PPG Sensor: Photoplethysmography Sensor - Measures Blood Volume Pulse (BVP), 3-axis Accelerometer: Captures motion-based activity, Event Mark Button: Tags events and correlate them with physiological signals, EDA Sensor (GSR Sensor) Electrodermal Activity Sensor:Used to measure sympathetic nervous system arousal and to derive features related to stress, engagement, and excitement, Infrared Thermopile: Reads peripheral skin temperature and Internal Real-Time Clock: Temporal resolution up to 0.2 seconds in streaming mode
SIMPLI-CITY Mobile App Functionality:
A user of the SIMPLI-CITY enters their credentials which allows the system to access their calendar. They can then select an activity to attend and they are then presented with a suggested route. The SIMPLI-CITY app will inform the user of traffic anomalies along route, vehicle information (ex: If low on fuel then navigation to nearest service station), and a user can also receive SMS information within the app. In a future version of the app the SMS will be read aloud to the driver.