Battle, Steve (2022) Ontological Design for Robotics. In: Proceedings of Relating Systems Thinking and Design, RSD11, 3-16 Oct 2022, Brighton, United Kingdom.

Preview	Text Battle_Ontological_paper_2022.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (6MB) | Preview
Preview	Slideshow Battle_Ontological_slides_2022.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview

Abstract

Ontological design is a way of defining, or constructing, a world of possibilities through design practice. Robotics is often taught in a very rationalistic and didactic fashion, focusing on the kinematics of robot motion. Ontological design is fundamentally enactive, allowing cognition to arise through a dynamic interaction between an individual and their environment. With the ontological design of robots, the focus is on learning robotic design through practice and serendipity. This paper reports on the practice of running hands-on robot workshops for children and adults, the philosophical background to this, and future direction using tangible programming tools. We learn about robots controlled, not through conventional programming, but using simple neural circuits. The inspiration for this is Braitenberg’s Vehicles: Experiments in Synthetic Psychology (1984), which describes a cybernetic approach to modelling brain function. To realise learning as a fully shared experience, the ‘coding’ of the robot should be experienced as a tangible activity. We explore a sequence of robot designs that can be built using tools for augmented reality and consider the possibilities for ontological learning at each stage.

Item Type:	Conference/Workshop Item (Paper)
Uncontrolled Keywords:	ontological design, robotics, AI, artificial intelligence
Divisions:	Faculty of Design
Date Deposited:	10 May 2024 18:38
Last Modified:	10 May 2024 18:45
URI:	https://openresearch.ocadu.ca/id/eprint/4505

Actions (login required)

Edit View