Technical Briefs

Position Control Using Acceleration-Based Identification and Feedback With Unknown Measurement Bias

[+] Author and Article Information
Jaganath Chandrasekar

Department of Aerospace Engineering, The University of Michigan, Ann Arbor, MI 48109-2140jchandra@umich.edu

Dennis S. Bernstein

Department of Aerospace Engineering, The University of Michigan, Ann Arbor, MI 48109-2140dsbaero@umich.edu

J. Dyn. Sys., Meas., Control 130(1), 014501 (Dec 05, 2007) (9 pages) doi:10.1115/1.2807180 History: Received January 06, 2005; Revised June 27, 2007; Published December 05, 2007

A position-command-following problem for asymptotically stable linear systems is considered. To account for modeling limitations, we assume that a model is not available. Instead, acceleration data are used to construct a compliance (position-output) model, which is subsequently used to design a position servo loop. Furthermore, we assume that the acceleration measurements obtained from inertial sensors are biased. A subspace identification algorithm is used to identify the inertance (acceleration-output) model, and the biased acceleration measurements are used by the position-command-following controller, which is constructed using linear quadratic Gaussian (LQG) techniques.

Copyright © 2008 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Standard problem for designing a position-tracking controller Gc that uses biased acceleration measurements. To facilitate controller synthesis using LQG, the backward-path controller Gbp that is used to reject the sensor bias is included in the Plant G.

Grahic Jump Location
Figure 2

Control architecture for discrete-time LQG position control using acceleration feedback and a backward-path controller Gbp

Grahic Jump Location
Figure 3

Construction of the compliance Ĝcomp by cascading a double integrator with the identified inertance

Grahic Jump Location
Figure 5

Error between the actual position of m1 and the output ypos,1 of the identified compliance model when u1 is an impulse and u2=0. For position-tracking controller synthesis, the identified compliance is used.

Grahic Jump Location
Figure 6

Error between the position of m2 and the output ypos,2 from the identified compliance model when u1=0 and u2 is an impulse

Grahic Jump Location
Figure 7

Magnitude of the diagonal entries of Wr(z)

Grahic Jump Location
Figure 8

Magnitude of the diagonal entries of Gbp(z)

Grahic Jump Location
Figure 9

Position-command following for the two-mass system using an LQG controller and a backward-path controller. The LQG controller Gc produces the control input u to track the position command r, while the backward-path controller with zero dc gain, that is, a zero at z=1, rejects the sensor bias.

Grahic Jump Location
Figure 10

Acceleration measurements of the two masses. The sensor biases in the accelerometers are shown as dashed lines.

Grahic Jump Location
Figure 11

Magnitudes of the diagonal entries of Gsens,r, the sensitivity transfer function between the reference position command r and the position-tracking error zpos. The magnitude of the sensitivity function is low in the required bandwidth between 0.1Hz and 1Hz.

Grahic Jump Location
Figure 12

Magnitudes of the diagonal entries of Gsens,v, the sensitivity function between the bias v and position-tracking error zpos. The inclusion of a backward-path controller with zero dc gain ensures that as k→∞ the position-tracking performance is not affected by the sensor bias v.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In