Abstract

The goal of this chapter is to propose constraint satisfaction as a central design concept for intelligent robots. Some proposals for codes of robot ethics apparently presuppose the existence of certain technical abilities on behalf of the robot designer that, simply put, do not yet exist. Namely, given current robot design techniques, it is usually not possible to specify local or global limitations on a robot's behavior; or, if indeed we can specify such limitations, it is usually not possible to verify that the robot will always (or almost always) satisfy them. Through a case study of a simple robot soccer player, prioritized constraint satisfaction and hierarchical control decomposition are proposed as design techniques that will provide a more viable foundation for codes of robot ethics.


Type

Book Section

Publisher

Cambridge University Press

ISBN
978-0-511-97803-6

Creators

Anderson, Michael (Editor)
Anderson, Susan Leigh (Editor)
Mackworth, Alan K. (Author)

Place

Cambridge

Webpage

→ external link


Professional Fields