A novel large-scale database for fuel and energy use of diverse vehicles in real-world.
VED captures GPS trajectories of vehicles along with their timeseries data of fuel, energy, speed, and auxiliary power usage, and the data was collected through onboard OBD-II loggers from Nov, 2017 to Nov, 2018. The fleet consists of total 383 personal cars (264 gasoline vehicles, 92 HEVs, and 27 PHEV/EVs) in Ann Arbor, Michigan, USA. Driving scenarios range from highways to traffic-dense downtown area in various driving conditions and seasons. In total, VED accumulates approximately 374,000 miles.
A number of examples were presented in the paper to demonstrate how VED can be utilized for vehicle energy and behavior studies. Potential research opportunities include data-driven vehicle energy consumption modeling, driver behavior modeling, machine and deep learning, calibration of traffic simulators, optimal route choice modeling, prediction of human driver behaviors, and decision making of self-driving cars.
Link to the paper:
Vehicle Energy Dataset (VED), A Large-scale Dataset for Vehicle Energy Consumption Research
Geunseob (GS) Oh, David J. LeBlanc, Huei Peng
IEEE Transactions on Intelligent Transportation Systems (T-ITS), 2020.
The paper is also available on Arxiv.
Contact: gsoh@umich.edu.
GS Oh, Ph.D. Candidate, University of Michigan.
VED consists of Dynamic Data (time-stamped naturalistic driving records of 383 vehicles) and Static Data (Vehicle parameters for the 383 vehicles)
Dynamic Data: "VED_DynamicData.7z" contains a number of "VED_mmddyy_week.csv" files
- Includes a week worth dynamic data, for mmddyy ~ (mmddyy + 7 days)
- Columns represent: DayNum, VehId, Trip, Timestamp(ms), Latitude[deg], Longitude[deg], Vehicle Speed[km/h], MAF[g/sec], Engine RPM[RPM], Absolute Load[%], Outside Air Temperature[DegC], Fuel Rate[L/hr], Air Conditioning Power[kW], Air Conditioning Power[Watts], Heater Power[Watts], HV Battery Current[A], HV Battery SOC[%], HV Battery Voltage[V], Short Term Fuel Trim Bank 1[%], Short Term Fuel Trim Bank 2[%], Long Term Fuel Trim Bank 1[%], Long Term Fuel Trim Bank 2[%]
- Notes: Each combination of VehID, Trip is unique. DayNum represents elapsed days since a reference date. (DayNum 1 = Nov, 1st, 2017, 00:00:00, DayNum 1.5 = Nov, 1st, 2017, 12:00:00) For the details, refer to the VED paper
Static Data: "VED_Static_Data_ICE&HEV.xlsx", and "VED_Static_Data_PHEV&EV.xlsx"
- Includes parameters of all 383 vehicles (264 gasoline vehicles, 92 HEVs, and 27 PHEV/EVs)
- There are 3 pure EV vehicles in the dataset. All of them are 2013 Nissan Leaf with an advertised battery capacity of 24 kWh.
- Columns represent: VehId, EngineType, Vehicle Class, Engine Configuration & Displacement Transmission, Drive Wheels, Generalized_Weight[lb]
License under the Apache License 2.0