rlm@432: Here I explore the design and capabilities of my system (called
rlm@432: =CORTEX=) which enables experiments in /embodied artificial
rlm@432: intelligence/ -- that is, AI which uses a physical simulation of
rlm@432: reality accompanied by a simulated body to solve problems.
rlm@432: 
rlm@432: In the first half of the thesis I describe the construction of
rlm@432: =CORTEX= and the rationale behind my architecture choices. =CORTEX= is
rlm@432: a complete platform for embodied AI research. It provides multiple
rlm@432: senses for simulated creatures, including vision, touch,
rlm@432: proprioception, muscle tension, and hearing. Each of these senses
rlm@432: provides a wealth of parameters that are biologically
rlm@432: inspired. =CORTEX= is able to simulate any number of creatures and
rlm@432: senses, and provides facilities for easily modeling and creating new
rlm@432: creatures. As a research platform it is more complete than any other
rlm@432: system currently available.
rlm@432: 
rlm@432: In the second half of the thesis I develop a computational model of
rlm@432: empathy, using =CORTEX= as a base. Empathy in this context is the
rlm@432: ability to observe another creature and infer what sorts of sensations
rlm@432: that creature is feeling. My empathy algorithm involves multiple
rlm@432: phases. First is free-play, where the creature moves around and gains
rlm@432: sensory experience. From this experience I construct a representation
rlm@432: of the creature's sensory state space, which I call \phi-space. Using
rlm@432: \phi-space, I construct an efficient function for enriching the
rlm@432: limited data that comes from observing another creature with a full
rlm@432: compliment of imagined sensory data based on previous experience. I
rlm@432: can then use the imagined sensory data to recognize what the observed
rlm@432: creature is doing and feeling, using straightforward embodied action
rlm@432: predicates. This is all demonstrated with using a simple worm-like
rlm@432: creature, recognizing worm-actions in video.
rlm@432: