Using Multi-agent Systems to Pursue Autonomy with Automated Components
详细信息   
摘要
Humans have used tools to transform raw resources into valued outputs ever since society harnessed fire. The type of tool, amount of effort and form of energy required depends on the output or object being created. As tools evolved into machines, they enhanced operator productivity. Hence, industry continues to invest heavily in machines to assist people to do more with less physical control and/or interaction. This involves automating functions previously completed manually. Taylorism and the Hawthorn experiments all contributed to optimising industrial outputs and value engineers continue to promote a mecha- nized workforce in order to minimise business variations in human performance and their behaviour. Researchers have also pursued this goal using Computational Intelligence (CI) techniques. This process of transforming cognitive functionality into machine actionable form has encompassed many careers. Machine Intelligence (MI) is becoming more aspirational, with Artificial Intelligence (AI) enabling the achievement of numerous goals. More recently, Multi-Agent Systems (MASs) have been employed to provide a flexible framework for research and development. These frameworks facilitate the development of component interoperability, with coordination and cooperation techniques needed to solve real-world problems. However problems typically manifest in complex, dynamic and often hostile environments. Based on the effort to seek or facilitate human-like decision making within machines, it is clear that further research is required. This paper discusses one possible avenue. It involves future research, aimed at achieving a cognitive sub-system for use on-board platforms. The framework is introduced by describing the human-machine relationship, followed by the theoretic background into cognitive architectures and a conceptual mechanism that could be used to implement a virtual mind. One which could be used to improve automation, achieve greater independence and enable more autonomous behaviour within control systems.