System Identification for a Wheeled Off-Road Vehicle Based on Dynamics
Mode Decomposition
Abstract
With the advent of connected vehicle technology and widespread
long-distance communication, leveraging data from real-world operational
cases to identify and characterize vehicle dynamics and control is now
possible. However, there exists a limited number of studies in the field
of off-road vehicle identification as compared to on-road vehicle
identification. This study explores the applications of Dynamic Mode
Decomposition with Control (DMDc) for linear vehicle dynamics, linear
human-driver-controller identification, and possible use for off-road
vehicle identification problems. An implementation is tested using
real-world data collected from the University of Illinois Modified
R-Gator(UIMR), an autonomy-capable, drive-by-wire, 6x4, Ackerman Steer,
wheeled off-road vehicle platform. The UIMR, partially developed by the
University of Illinois at Urbana-Champaign’s Autonomous and Unmanned
Vehicle Systems Lab, part of the Industrial and Systems Engineering
department, is presented for the first time.