Software Enabled Control

Chapter 12.8 - Conclusion

12.8.   CONCLUSION

An adaptation scheme is proposed for the online adaptation of mode
transition controllers designed via the blending local mode controllers approach.
When the active plant model of the adaptation scheme is a good
approximation of the system to be controlled, then it is expected that the
adapted controller will track the desired trajectory very well in the presence
of parametric changes and disturbances. However, if the active plant model
does not capture the local model and the input/output behavior of the
systems to be controlled, poor tracking performance and unstable tracking
can result. Further investigation is needed to add robustness capability to the
adaptation scheme so that it can avoid controller faults due to poor approximation
of the nonlinear plant. One way of preventing controller faults could
be to decrease the aggressiveness of the scheme when approximation errors
are significant and increase the aggressiveness when the errors are small.

Adaptive neural-net-based limit prediction and avoidance algorithms are
integrated with a flight controller for avoiding performance limits during
aggressive maneuvers of an unmanned helicopter. The limit avoidance pro-
vides ‘‘artificial limits’’ for the control. A pseudo-control hedging technique is
used to eliminate the undesirable effects of control saturation in the context
of controller adaptation.

ACKNOWLEDGMENTS

This research has been supported by the DARPA Software Enabled Control program.
The support, assistance, and contributions of the DARPA program management and
AFRL personnel are gratefully acknowledged.

UNLIMITED FREE
ACCESS
TO THE WORLD'S BEST IDEAS

SUBMIT
Already a GlobalSpec user? Log in.

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.

Customize Your GlobalSpec Experience

Category: Machine Control Software
Finish!
Privacy Policy

This is embarrasing...

An error occurred while processing the form. Please try again in a few minutes.