Understanding driving behaviors is essential for improving safety andmobility of our transportation systems. Data is usually collected viasimulator-based studies or naturalistic driving studies. Those techniques allowfor understanding relations between demographics, road conditions and safety.On the other hand, they are very costly and time consuming. Thanks to theubiquity of smartphones, we have an opportunity to substantially complementmore traditional data collection techniques with data extracted from phonesensors, such as GPS, accelerometer gyroscope and camera. We developedstatistical models that provided insight into driver behavior in the SanFrancisco metro area based on tens of thousands of driver logs. We used noveldata sources to support our work. We used cell phone sensor data drawn fromfive hundred drivers in San Francisco to understand the speed of traffic acrossthe city as well as the maneuvers of drivers in different areas. Specifically,we clustered drivers based on their driving behavior. We looked at driver normsby street and flagged driving behaviors that deviated from the norm.
展开▼