A new structure for non-linear black-box system identification using the extended Kalman filter

Abstract

The method works iteratively in the time domain using an Extended Kalman Filter. The model retains a state space structure in modal canonical form, which ensures that a minimal number of parameters need to be identified and also produces additional information in terms of system eigenvalues and dominant modes. This structure is completely black-box, requiring no physical understanding of the process for successful identification, and it is possible to easily expand the order and complexity of nonlinearities, whilst ensuring good parameter conditioning. A simple nonlinear example illustrates the method, and identification of a highly nonlinear brake model is also presented. These examples show the method can be applied as a mechanism for model order reduction; it is equally well suited as a tool for nonlinear plant system identification. In both capacities this new method is valuable, particularly as the generation of simplified models for the whole vehicle and its subsystems is an increasingly important aspect of modern vehicle design.

Real-time characterisation of driver steering behaviour

“…This needs to incorporate both the vehicle and driver models as it must also estimate the driver-vehicle dynamic states. Kalman filters have widely been used in state estimation of nonlinear systems and have been applied for nonlinear system identification in [24,25]. The real-time adaptation required here may be well suited to the application of either an extended Kalman filter (EKF) or an unscented Kalman filter (UKF) [24].…”

“…Of course, no 'model' exists for the parameter states, sȯ T p = 0, K ug = 0 (25) and the filter adapts the parameters through the assumption of non-zero modelling error on these additional states alone; set…”

Comparison of Transfer Function Models to Represent the Correlation Between Vehicle Lateral Acceleration and Head Tilting Angle in Motion Sickness

“…The index of MS severity MSI [27] is typically predicted using mathematical fits through experimental data [28]- [30]; based on that, SI will predict the fits using the black-box method. SI is a MATLAB toolbox that allows the user to choose an adequately precise model structure to fit its unknown parameters to map the existing I/O data properly [31]. Additionally, it can estimate the behavior of the system's output towards an unknown input.…”