Dynamic Sensor Policies

Kun Krebsbach and Maria Gini

We propose to generate a tree of possible sensing policies offline (using dynamic programming), cache the optimal sensor decisions at each level, and subsequently use actual world states as indexes into this structure to make a rational sensor choice online. Because no states are discarded in the dynamic programming process (only non-optimal paths), we are always guaranteed of having earlier cached an optimal decision for each state we anticipate possibly encountering at execution time. This combination of offiine and online computation allows the sensor selection process to be sensitive to actual events, while making use of assumptions and regularities inherent in the structure of the domain.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.