### Abstract

Lingua originale | English |
---|---|

Stato di pubblicazione | Published - 2005 |

### Cita questo

**Stochastical Real Time Finite State Machine LPC for Planar Manipulator Control System Model estimation.** / Raimondi, Francesco Maria; Melluso, Maurizio.

Risultato della ricerca: Paper

}

TY - CONF

T1 - Stochastical Real Time Finite State Machine LPC for Planar Manipulator Control System Model estimation

AU - Raimondi, Francesco Maria

AU - Melluso, Maurizio

PY - 2005

Y1 - 2005

N2 - This paper presents a new stochastical real-time LPC (Last Principal Component) algorithm to estimate single-input-single-output (SISO) and multiple-input-multiple-output (MIMO) varying time models from input output data clusters of non stationary black boxes. Each of data clusters is on a time window. An application to estimate the control system model of a planar manipulator is developed. In fact many mathematical models of physical systems are non stationary such as industrial manipulator model. A real time estimation algorithm via stochastical LPC algorithm and an appraiser called "finite state machine" is then described For every data cluster the finite state machine updates the parameters of a Gaussian varying time model via an optimality design criterion that maximises the Likelihood function. The estimated steady-state parameters are constant values. By applying to two links planar manipulator, numerical tests of simulation in Matlab 6.5 demonstrate the effectiveness of this algorithm.

AB - This paper presents a new stochastical real-time LPC (Last Principal Component) algorithm to estimate single-input-single-output (SISO) and multiple-input-multiple-output (MIMO) varying time models from input output data clusters of non stationary black boxes. Each of data clusters is on a time window. An application to estimate the control system model of a planar manipulator is developed. In fact many mathematical models of physical systems are non stationary such as industrial manipulator model. A real time estimation algorithm via stochastical LPC algorithm and an appraiser called "finite state machine" is then described For every data cluster the finite state machine updates the parameters of a Gaussian varying time model via an optimality design criterion that maximises the Likelihood function. The estimated steady-state parameters are constant values. By applying to two links planar manipulator, numerical tests of simulation in Matlab 6.5 demonstrate the effectiveness of this algorithm.

KW - LPC; digital filters; finite state machine; maximum likelihood; manipulator control; estimation

UR - http://hdl.handle.net/10447/27794

M3 - Paper

ER -