PhD Thesis Proposal Defence


"Probabilistic Plan Recognition from Low-Level Sensors"

By Miss Jie Yin


Abstract:

Plan recognition has traditionally been developed for logically
encoded application domains with a focus on logical reasoning. In
this proposal, we propose a probabilistic plan-recognition
framework that integrates low-level sensor modeling with
high-level goal inference, and use a wireless LAN application
domain to illustrate the framework. In our framework, a
generalized sensor model, at the low level, is developed to
transform the low-level sensory observations into intermediate
states, such as locations and motion patterns. On top of this,
subsequent high-level inferences about a user's actions and goals
is enabled by constructing a behavior model at the high level.
Based on this framework, we propose two categories of plan
recognition algorithms: the first is location-based plan
recognition algorithm and the second is motion pattern-based plan
recognition algorithm. We demonstrate the effectiveness of our
proposed algorithms using the data collected in a real wireless
environment.


Date:     		Wednesday, 22 June 2005

Time:                   10:30a.m.-12:30p.m.

Venue:                  Room 2302
			lifts 17-18

Committee Members:      Dr. Qiang Yang (Supervisor)
                        Dr. Nevin Zhang (Chairperson)
			Prof. Lionel Ni
			Dr. Dit Yan Yeung


**** ALL are Welcome ****