Accounting for the Variation of Driver Aggression in the Simulation of Conventional and Advanced Vehicles

Abstract

Hybrid electric vehicles, plug-in hybrid electric vehicles, and battery electric vehicles offer the potential to reduce both oil imports and greenhouse gases, as well as to offer a financial benefit to the driver. However, assessing these potential benefits is complicated by several factors, including the driving habits of the operator. We focus on driver aggression, i.e., the level of acceleration and velocity characteristic of travel, to (1) assess its variation within large, real-world drive datasets, (2) quantify its effect on both vehicle efficiency and economics for multiple vehicle types, (3) compare these results to those of standard drive cycles commonly used in the industry, and (4) create a representative drive cycle for future analyses where standard drive cycles are lacking. INTRODUCTIONHybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), and battery electric vehicles (BEVs) offer the potential to reduce both oil imports and greenhouse gases, as well as to offer a financial benefit to the driver. However, assessing these potential benefits is complicated by several factors, including the local climate, the cleanliness of the grid supplying electricity (for PHEVs and BEVs), and the driving habits of the operator, among many other things. Driving habits can be divided into two topics for consideration: (1) trip patterns, i.e., the distribution of trip times and distances, and (2) aggression, i.e., the level of acceleration and the velocity characteristics of travel. Investigation of trip patterns in [1] and [2] found that the variation in the distribution of daily miles traveled observed in real-world, multi-day drive data [3] produces significant variation in gasoline savings and total cost of ownership.Herein, we focus on driver aggression to (1) assess its variation within large, real-world drive datasets, (2) quantify its effect on both vehicle efficiency and economics for multiple vehicle types, (3) compare these results to those of standard drive cycles commonly used in the industry, and (4) create a representative drive cycle for future analyses where standard drive cycles are lacking. By doing so, we aim to supply an approach for vehicle performance simulation and testing that accurately captures the variation between different drivers, in particular for high-level techno-economic analyses performed using the National Renewable Energy Laboratory's (NREL's) Battery Ownership Model. This work is supported by the U.S. Department of Energy's Vehicle Technologies Program. ANALYSISIn this study, we apply a high-resolution vehicle simulator to real-world drive data to calculate a distribution of vehicle efficiencies for multiple vehicle types and operational modes. We then analyze these distributions and compare them to standard drive cycles commonly employed in the industry. From there, we synthesize and validate an artificial drive cycle that characterizes the average of the complete set of real-world drive data. We also describe in detail a set of associated scaling f...

A Survey of Autonomous Driving: Common Practices and Emerging Technologies

“…Typically, driving style is defined with respect to either aggressiveness [217]- [221] or fuel consumption [222]- [226]. For example, [227] introduced a rule-based model that classified driving styles with respect to jerk.…”

Updating United States Advanced Battery Consortium and Department of Energy battery technology targets for battery electric vehicles

“…The construction of this duty cycle is detailed in Refs. [11], where it has been shown to yield exceptionally accurate predictions of average fleet efficiency.…”

Aging of lithium-ion batteries for electric vehicles

“…The analysis of battery temperatures and ambient conditions helps to refine cooling requirements. Additionally, field data can be used to further develop lab test procedures, that is, to enhance load profiles (Neubauer and Wood, 2013;Peterson et al, 2010). Using field data, lifetime models can be adapted or trained, and the predictability can be improved (Liaw and Dubarry, 2007).…”