Vehicle Energy Dataset (VED), A Large-Scale Dataset for Vehicle Energy Consumption Research

We present Vehicle Energy Dataset (VED), a large-scale dataset of fuel and energy data collected from 383 personal cars in Ann Arbor, Michigan, USA. This open captures GPS trajectories vehicles along with their time-series fuel, energy, speed, auxiliary power usage. A diverse fleet consisting 264 gasoline vehicles, 92 HEVs, 27 PHEV/EVs drove real-world Nov, 2017 to 2018, where the were through onboard OBD-II loggers. Driving scenarios range highways traffic-dense downtown area various driving conditions seasons. In total, VED accumulates approximately 374,000 miles. discuss participant privacy protection develop method de-identify personally identifiable information while preserving quality data. number case studies demonstrate how can be utilized for vehicle behavior studies. The investigate impacts factors known affect economy identify energy-saving opportunities that hybrid-electric eco-driving techniques provide. Potential research include data-driven consumption modeling, driver calibration traffic simulators, optimal route choice prediction human behaviors, decision making self-driving cars. believe an instrumental asset development future automotive technologies. accessed at https://github.com/gsoh/VED .