Main Page
From BerkeleyRobocupWiki
Leon has set up this wiki for the sake of helping people learn about and use his action representation work.
Contents |
Overview
Leon Barrett did his dissertation on an architecture for computers to represent and analyze the sorts of real-world actions we humans do constantly. Humans effortlessly walk, tie our shoes, and cook, but these are almost impossible for computers or robots to perform or even reason about. The first step to enabling them to manage this is to have a way of describing actions in a very simple, precise way that computers can handle. Leon Barrett's dissertation describes such a tool, and this software is his implementation of it.
Or, to put it another way, here is that dissertation's abstract:
- I present a computational architecture designed to capture certain properties essential to actions, including compositionality, concurrency, quick reactions, and resilience in the face of unexpected events. It uses a structured internal state model and complex inference about the environment to inform decision-making. The properties above are achieved by combining interacting procedural and probabilistic representations, so that the structure of actions is captured by Petri Nets, which are informed by, and affect, a model of the world represented as a Probabilistic Relational Model. I give both a theoretical analysis of the architecture and a demonstration of its use in a simulated robotic environment.
Software
- The software described here is explained at Software: Petri Net / Bayes Net Simulator
- Software: Robocup Simulator
Reading
Demonstrations
Demonstrations of this software controlling simulated RoboCup soccer players may be seen at Demonstrations.
