Deakin University
Browse

File(s) under permanent embargo

An optimal washout filter based on genetic algorithm compensators for improving simulator driver perception

conference contribution
posted on 2015-01-01, 00:00 authored by Houshyar AsadiHoushyar Asadi, Shady MohamedShady Mohamed, Kyle Nelson, Saeid Nahavandi, M Oladazimi
A simulator motion platform cannot exactly reproduce the longitudinal and rotational motions
experienced in a real vehicle as it is constrained by its physical limits. The aim of this research is to provide an
optimal Motion Cueing Algorithm (MCA) that can generate the most realistic motions and high fidelity vehicle
accelerations and angular velocities, within the simulator’s physical limitations. The optimal washout filter based
on Linear Quadratic Regulator (LQR), which takes the recent vestibular system mathematical model and
simulator motion in to account, has been proposed to constrain the human perception error between the simulator
and real driving, within the limits of the platform motion. This paper presents a new strategy based on optimal
control theory and the Genetic Algorithm (GA) to reproduce a signal that can closely follow the reference signal
and avoid false motion cues. An optimization method for adjusting the obtained optimal washout filter transfer
functions, based on genetic algorithms is used. Three additional compensatory linear blocks are integrated into
the LQR-based optimal washout filter, to be tuned based on GA in order to modify the performance of the filters
and minimize the fitness value if increasing the order is required. This is achieved by taking a series of factors
into account, including: the vestibular sensation error between real and simulated cases; the human threshold
limiter in tilt coordination; the human sensation error fluctuation; and cross correlation coefficient; where the effects
of these aspects have been previously ignored. The proposed optimized motion cueing algorithm based on
compensators using GA is implemented in the MATLAB/Simulink software packages. The results show the
superiority of the proposed method due to its better increased performance, enhanced motion fidelity, improved
human sensation, and reduced workspace usage compared to previous optimal washout filters.

History

Event

Driving Simulation. Conference & Exhibition (2015 : Tübingen, Germany)

Pagination

1 - 10

Publisher

Max Planck Institute for the Advancement of Science

Location

Tübingen, Germany

Place of publication

Munich, Germany

Start date

2015-09-16

End date

2015-09-18

Language

eng

Publication classification

E1 Full written paper - refereed

Copyright notice

2015, Max Planck Institute for the Advancement of Science

Title of proceedings

DSC 2015 : Proceedings of the Driving Simulation Conference & Exhibition

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC