Synthesis of automatic feedback control loops: a more quantitative view

2004 ◽  
pp. 19-28
Keyword(s):  
Feedback Control ◽  
Control Loops ◽  
Automatic Feedback

Automatic Feedback Control of Mechanical Gas Face Seals via Clearance Control

2004 ◽  
Vol 47 (4) ◽  
pp. 500-507 ◽  
Author(s):  
S. S. Yelma ◽  
B. A. Miller ◽  
R. G. Landers
Keyword(s):  
Feedback Control ◽  
Automatic Feedback ◽  
Face Seals ◽  
Clearance Control

Dynamics of Nonlinear Feedback Control

2007 ◽  
Vol 19 (5) ◽  
pp. 1179-1214 ◽  
Author(s):  
H. P. Snippe ◽  
J. H. van Hateren
Keyword(s):  
Feedback Control ◽  
Feedback Loop ◽  
Mathematical Form ◽  
Nonlinear Feedback Control ◽  
Nonlinear Feedback ◽  
Pass Filter ◽  
Low Pass Filter ◽  
Control Loops ◽  
Low Pass ◽  
Multiplicative Control

Feedback control in neural systems is ubiquitous. Here we study the mathematics of nonlinear feedback control. We compare models in which the input is multiplied by a dynamic gain (multiplicative control) with models in which the input is divided by a dynamic attenuation (divisive control). The gain signal (resp. the attenuation signal) is obtained through a concatenation of an instantaneous nonlinearity and a linear low-pass filter operating on the output of the feedback loop. For input steps, the dynamics of gain and attenuation can be very different, depending on the mathematical form of the nonlinearity and the ordering of the nonlinearity and the filtering in the feedback loop. Further, the dynamics of feedback control can be strongly asymmetrical for increment versus decrement steps of the input. Nevertheless, for each of the models studied, the nonlinearity in the feedback loop can be chosen such that immediately after an input step, the dynamics of feedback control is symmetric with respect to increments versus decrements. Finally, we study the dynamics of the output of the control loops and find conditions under which overshoots and undershoots of the output relative to the steady-state output occur when the models are stimulated with low-pass filtered steps. For small steps at the input, overshoots and undershoots of the output do not occur when the filtering in the control path is faster than the low-pass filtering at the input. For large steps at the input, however, results depend on the model, and for some of the models, multiple overshoots and undershoots can occur even with a fast control path.

Feedback control of the neuromusculoskeletal system in a forward dynamics simulation of stair locomotion

2009 ◽  
Vol 223 (6) ◽  
pp. 663-675 ◽  
Author(s):  
A Selk Ghafari ◽  
A Meghdari ◽  
G Vossoughi
Keyword(s):  
Feedback Control ◽  
Degrees Of Freedom ◽  
Inverse Dynamics ◽  
Sagittal Plane ◽  
Human Movement ◽  
Stair Climbing ◽  
Dynamics Simulation ◽  
Forward Dynamics ◽  
Control Loops ◽  
The Stability

The aim of this study is to employ feedback control loops to provide a stable forward dynamics simulation of human movement under repeated position constraint conditions in the environment, particularly during stair climbing. A ten-degrees-of-freedom skeletal model containing 18 Hill-type musculotendon actuators per leg was employed to simulate the model in the sagittal plane. The postural tracking and obstacle avoidance were provided by the proportional—integral—derivative controller according to the modulation of the time rate change of the joint kinematics. The stability of the model was maintained by controlling the velocity of the body's centre of mass according to the desired centre of pressure during locomotion. The parameters of the proposed controller were determined by employing the iterative feedback tuning approach to minimize tracking errors during forward dynamics simulation. Simultaneously, an inverse-dynamics-based optimization was employed to compute a set of desired musculotendon forces in the closed-loop simulation to resolve muscle redundancy. Quantitative comparisons of the simulation results with the experimental measurements and the reference muscles' activities illustrate the accuracy and efficiency of the proposed method during the stable ascending simulation.

Sign in / Sign up

Export Citation Format

Share Document