MRI: CloudCar: Development of a Diverse Distributed Instrument for Vehicles in the Cloud
This project, developing an instrument referred to as CloudCar, aims to build a cloud-based infrastructure to monitor and collect data about drivers, vehicles, and road conditions. The type of data to be collected includes driver's biometrics (eyeball tracking, heart rate, blood pressure, and EEG brain wave analysis), sensory data from the vehicle, traffic data, and sensed data about road condition. The collected data is analyzed in the cloud and used to enable a wide spectrum of applications, including assessing drivers' behavior and ability to drive, the "health" of the vehicle, dissemination of information about road conditions to prevent accidents and road congestion, and vehicular crash detection and notification.
The basic equipment requested includes cloud servers, software and tuning tools, the OBD-II and CAN software/hardware, a number of 14-channel EEG equipment for the alpha, beta, and gamma wave analysis during driving, EEG headbands, and mobile devices. AvaCars will be developed to allow researchers and application developers to access information about virtual cars. This work extends current infrastructure to include the cloud for real-time monitoring and feedback. A successful integration of the cloud into the system has great potential to enhance the accuracy and reliability of the system.
The proposal raises multiple research questions and establishes research directions that can be pursued to address these questions when the proposed infrastructure is made available. The following research activities will be pursued:
- Design of a portal with AvaCar of a vehicle and road conditions with all the timeline of events;
- Measurements of driver attention and vehicle condition using mobile phones, OBDII, and CAN bus;
- Measurements of the driver distraction using EEG-based headband;
- Design of an infrastructure for near real-time notifications and alerts to drivers in an area;
- Real-time guidance to the driver for optimal fuel consumption, energy-efficient localization, and speed tracking;
- Proactive prediction of drivers? behavior during accidents, hazards, and construction; and,
- Performance analysis, sensitivity analysis, and calibration for V2C2V.
The proposed instrument can greatly impact the capability to assess and/or improve the driving performance of people with disabilities, senior citizens, teen drivers, physiotherapists, and proper physical exercises. Overall, the project contributes to personal and road safety, optimized fuel consumption, and energy-efficient localization. The project incorporates the use of the infrastructure in the curricula of the different institutions involved. Course content, focused on cloud computing and vehicular communications, will be developed and tailored to computer science, electrical engineering, mechanical engineering, and cognitive science programs at participating departments. The research team will build on their existing outreach and educational programs to attract diverse undergraduate and graduate students and involve them in the MRI project.