An NSF project named “study of driving volatility in connected and cooperative vehicle systems” aims at extracting driving volatility, characterized by hard acceleration/braking, jerky movements, sharp lane changes or turns, and abnormally high speeds in a connected vehicle environment. The objective of this project is to model computationally efficient algorithms for predicting driver actions and volatility using information about their prior behaviors combined with positions and motions obtained via wireless communications.
In previous years, the research team has developed and calibrated a base model implemented in MATSim and SUMO. This virtual testbed simulates an 8-million-person population and includes cars, trains, bus, bikeshare, taxi, and other for-hire vehicles calibrated to the year 2016. The team is building the architecture to host this virtual test bed and developing system design and user guide documentation.
These new sensors are focused on obtaining data about the bicyclist’s behavior, which will complement the current data and contribute to new findings. An iOS platform is also being developed to track the devices and visualize real-time data collected with them. More units are expected to be implemented into the device in the near future for crowdsourcing.