Iterative control in automotive testing

Abstract

Road simulator test rigs investigate both the durability and the squeak and rattle characteristics of vehicle structures. The test specimens are the whole vehicle or a sub-structure taken from the vehicle. In laboratory simulation tests, random time series of strain, displacement, or acceleration, which represent service conditions, are reproduced at specified locations within the structure. These responses are formulated from measurements made on the road or at a specialized proving ground. The premise is that reproducing these responses recreates service strain throughout the structure. The test rigs are normally driven by hydraulics and form multiple channel non-linear dynamic systems. Time series drive files that reproduce the strain are generated to excite the test rig. These are determined using an iterative procedure that overcomes the deficiencies of the hydraulics, channel interaction, and non-linearity in the system. The current paper outlines a new iterative algorithm with better convergence than the commonly used iterative algorithm which does not always converge. This new iterative algorithm is demonstrated in numerical simulations that represent typical automotive engineering systems.

Reproduction of real-world road profiles on a four-poster rig for indoor vehicle chassis and suspension durability testing

“…[21][22][23] This kind of control has been chosen among others [24][25][26] because it features important advantages. For example, each cycle starts and ends in a steady-state condition and the target signals do not have to be targets for the closedloop controller variables as reported in Plummer.…”

Design of a Road Simulator for Motorcycle Applications

“…The four-poster itself features the possibility to reproduce any road surface accurately, starting from a real-world acquisition, and then using a complex iterative deconvolution process [18][19][20][21]. This controller is separate from the one employed to "play" standard waveforms and used to perform standard tests as reported in Section 3.1.…”

A method for improving test rig performance using passive components

“…The parameters θ considered in (11) are property parameters of the passive component and location parameters of its connection(s) with the test specimen.…”

“…[8,9] multi-output (MIMO) applications of the approach are concerned. The TWR algorithm belongs to a set of methods known as iterative learning control (ILC) -Smolders et al [10] and Hay and Roberts [11,12] have introduced nonlinear adaptations of ILC for rig tracking control.…”