Linear Time Invariant State Space System Identification Using Adam Optimization

Author(s):  
Mark Naeem ◽  
Shady A. Maged
2008 ◽  
Vol 100 (5) ◽  
pp. 2537-2548 ◽  
Author(s):  
Eric Zarahn ◽  
Gregory D. Weston ◽  
Johnny Liang ◽  
Pietro Mazzoni ◽  
John W. Krakauer

Adaptation of the motor system to sensorimotor perturbations is a type of learning relevant for tool use and coping with an ever-changing body. Memory for motor adaptation can take the form of savings: an increase in the apparent rate constant of readaptation compared with that of initial adaptation. The assessment of savings is simplified if the sensory errors a subject experiences at the beginning of initial adaptation and the beginning of readaptation are the same. This can be accomplished by introducing either 1) a sufficiently small number of counterperturbation trials (counterperturbation paradigm [ CP]) or 2) a sufficiently large number of zero-perturbation trials (washout paradigm [ WO]) between initial adaptation and readaptation. A two-rate, linear time-invariant state-space model (SSMLTI,2) was recently shown to theoretically produce savings for CP. However, we reasoned from superposition that this model would be unable to explain savings for WO. Using the same task (planar reaching) and type of perturbation (visuomotor rotation), we found comparable savings for both CP and WO paradigms. Although SSMLTI,2 explained some degree of savings for CP it failed completely for WO. We conclude that for visuomotor rotation, savings in general is not simply a consequence of LTI dynamics. Instead savings for visuomotor rotation involves metalearning, which we show can be modeled as changes in system parameters across the phases of an adaptation experiment.


Filomat ◽  
2020 ◽  
Vol 34 (11) ◽  
pp. 3529-3552
Author(s):  
Namita Behera

We introduce generalized Fiedler pencil with repetition(GFPR) for an n x n rational matrix function G(?) relative to a realization of G(?). We show that a GFPR is a linearization of G(?) when the realization of G(?) is minimal and describe recovery of eigenvectors of G(?) from those of the GFPRs. In fact, we show that a GFPR allows operation-free recovery of eigenvectors of G(?). We describe construction of a symmetric GFPR when G(?) is symmetric. We also construct GFPR for the Rosenbrock system matrix S(?) associated with an linear time-invariant (LTI) state-space system and show that the GFPR are Rosenbrock linearizations of S(?). We also describe recovery of eigenvectors of S(?) from those of the GFPR for S(?). Finally, We analyze operation-free Symmetric/self-adjoint structure Fiedler pencils of system matrix S(?) and rational matrix G(?). We show that structure pencils are linearizations of G(?).


1996 ◽  
Vol 3 (1-2) ◽  
pp. 99-109
Author(s):  
Andrzej Maćkiewicz ◽  
Francisco López Almansa ◽  
José A. Inaudi

2014 ◽  
Vol 511-512 ◽  
pp. 867-870
Author(s):  
Su Zhen Li ◽  
Xiang Jie Liu ◽  
Gang Yuan

T-S model is linearized at sampling points into the form of linear time-invariant state space , and using supervisory predictive control and muti-step predictive control strategy, which reduces amount of calculation and improves the control performance. Introduction


Sign in / Sign up

Export Citation Format

Share Document