What Sort of Architecture is Required for a Human-Like Agent?

Abstract

This paper is about how to give human-like powers to complete agents. For this the most important design choice concerns the overall architecture. Questions regarding detailed mechanisms, forms of representations, inference capabilities, knowledge etc. are best addressed in the context of a global architecture in which different design decisions need to be linked. Such a design would assemble various kinds of functionality into a complete coherent working system, in which there are many concurrent, partly independent, partly mutually supportive, partly potentially incompatible processes, addressing a multitude of issues on different time scales, including asynchronous, concurrent, motive generators. Designing human like agents is part of the more general problem of understanding design space, niche space and their interrelations, for, in the abstract, there is no one optimal design, as biological diversity on earth shows.

A Cloud Based Architecture for Working on Big Data with Workflow Management

“…Sloman [26] has a strong evidence for thinking and designing models at the level of architecture, instead of building independent component units. Designing the real world data fusion requires the collaboration of the distributed system (network), organizational concepts (command), and environmental perception (textures).…”

AEROSTACK: An architecture and open-source software framework for aerial robotics

“…To increase the degree of autonomy of robots, Aerostack includes a reflective layer based on cognitive architectures [27], [6], [2], [26] to simulate certain self-awareness able to supervise the other layers. The reflective layer helps to see if the robot is actually making progress to its goal and to react in the presence of problems (unexpected obstacles, faults, etc.)…”

Designing a Safe Motivational System for Intelligent Machines

“…An excellent way to begin designing such a human-like motivational system is to start with an attentional architecture based upon Sloman's architecture for a human-like agent (Sloman 1999). Reflexes and emotions could easily be implemented in accordance with Baars Global Workspace Theory (Baars 1997) which postulates that most of human cognition is implemented by a multitude of relatively small, local, special purpose processes that are almost always unconscious.…”