Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.


Minimum correction control for obstacle avoidance using set invariance

Due to the constraints of current automation technology there are still many cases of humans operating machinery. Prime examples of this are mining excavators, tele-operated systems and the automobile. These examples, cannot be fully automated due to changing and uncertain environments or tasks. In some of these examples, there are defined obstacles or no-go regions that an operator must avoid. In this talk we propose a method for assisting the operator to avoid these regions by minimally correcting his command. This correction takes into account that the future operator's commands are completely unknown and guarantees obstacle avoidance in the future. We define this as minimum correction control. One of the properties of this control law is that the operator will only be corrected if a collision will result from his current command.

The minimum correction control law depends on the construction of an appropriate control invariant set and the associated minimum correction control law. In this talk, we detail two methods for construction of obstacle avoidance control invariant sets suitable for the minimum correction controller.. The minimum correction control law is determined by solving a fixed number of convex mathematical program that determines the input closest to the operator's current input such that the successor state is in the invariant set. We compare this to previous methods and discuss the implementation of the minimum correction controller on a mining excavator.
Type of Seminar:
Public Seminar
Michael Kearney
Jul 25, 2008   17:00

Contact Person:

Colin Jones
No downloadable files available.
Biographical Sketch: