Recovering Models of a Four-Wheel Vehicle Using Vehicular System Data

Abstract—This paper discusses efforts to parameterize theactuation models of a four-wheel automobile for the purposesof closed-loop control. As a novelty, the authors used theequipment already available or in use by the vehicle, ratherthan expensive equipment used solely for the purpose of systemidentification. After rudimentary measurements were taken ofwheelbase, axle width, etc., the vehicle was driven and datawere captured using a controller area network (CAN) interface.Based on this captured data, we were able to estimate thefeasibility of certain closed-loop controllers, and the modelsthey assumed (i.e., linear, or nonlinear) for control. Exampleswere acceleration and steering. This work served to inform theseparation of differences in simulation and vehicle behaviorduring vehicle testing. I. INTRODUCTIONA major complexity of vehicle control is an accuratemodel of the vehicle for controller design, and simula-tion. Poor vehicle models can result in unstable behaviorwhen applied to hardware, resulting in unsafe situations forthose involved, or frustrating demonstrations that divergesignificantly from simulation results. As a result, significanthardware-in-the-loop testing is necessary prior to controlsystem design, to minimize the risk.This is undesirable from many aspects. First, hardware-in-the-loop experiments are costly in terms of personnel timeand equipment, as well as any facilities which must be rented.Additionally, it predicates the implementation of algorithmsfor higher-level performance on already known or well-understood platforms upon which those algorithms will run.