Teleoreactive neural networks

Abstract

The Teleo-Reactive (TR) formalism has been presented as a new programming paradigm to write reactive robot control programs. The formalism is based in a circuit semantics that intuitively can be ported in a direct way to a layered neural network architecture.

But to capture the essence of the TR programs a more sophisticated mechanism, of synthesis must be developed, that allows to express in a neural architecture 1) the reactive nature of the programs, 2) the incremental learning or new TR sequences and trees and 3) the continuous feedback from the world.

We present an analysis of TR programs and a method to synthesize those programs into an ontogenic neural network model that captures all the features of the program and can evolve with the agent as he explore the world.

This method can be easily integrated to the learning architecture of the reactive agent.