Quantifying Repeatability of Real-World On-Road Driving Using Dynamic Time Warping
2022-01-0310
03/29/2022
- Event
- Content
- There are numerous activities in the automotive industry in which a vehicle drives a pre-defined route multiple times, such as portable emissions measurement systems (PEMS) testing or real-world electric vehicle range testing, in which the speed profile is not the same for each drive cycle due to uncontrollable real-world variables such as traffic, stoplights, stalled vehicles, or weather conditions. It sometimes can be difficult, then, to compare each run accurately. To this end, this paper presents a method to compare and quantify the repeatability of real-world on-road vehicle driving schedules using dynamic time warping (DTW). DTW is a well-developed computational algorithm which compares two different signals describing the same underlying phenomenon but occurring at different time scales. In this paper, DTW is applied to real-world on-road drive cycles, and metrics are developed to quantify how similar these drive cycles are to a common reference cycle. It goes on to explore the development of a nominal drive cycle which is representative of driving a given route repeatedly. This methodology is vehicle-agnostic and can be applied to conventional light-duty vehicles, hybrid or fully electric vehicles, or heavy-duty on-road vehicles.
- Citation
- Lobato, P., Rayno, M., Daily, J., and Bradley, T., "Quantifying Repeatability of Real-World On-Road Driving Using Dynamic Time Warping," SAE Technical Paper 2022-01-0310, 2022, .