Data Collection for Real-World Driving Studies: Challenges and Lessons from a Project on assessing EV Viability in Rural Michigan

Real-world driving data is an invaluable asset for several types of transportation research ranging from emissions estimation to vehicle control systems development. Traditional methods of real-world driving data collection utilize advanced and expensive GPS-based data logging equipment. Such equipment may possess unnecessarily excessive capabilities depending on the type of study, increasing study complexity and expenditure. This paper proposes the implementation of the Google Maps Application as a surrogate measurement device for real-world driving data logging, reducing study complexity and cost. Driving data can be collected from drivers who use Google Maps on their smartphones, making the process of implementation minimally invasive. The merits and disadvantages of this type of data logging will be analyzed through the case study of the ‘Rural Driving Study – Analysis of the Viability of Electric Vehicles in Rural Michigan’ research project that is currently being undertaken at the University of Michigan, Ann Arbor. This project solely relies on the usage of the Google Maps Application to log driving data, thereby creating an effective testbed to evaluate this technology for research. Furthermore, the paper will also present recommendations to optimize the effectiveness of the data collection process, and the type of studies for which this technology can be applied. Findings suggest that this technology is optimized for collecting non-granular data such as trip distances, durations, and start/end locations. Connectivity, data transfer, and privacy concerns are challenges with this technology. The technology is not recommended for studies requiring constant sampling rates and acceleration data.