Situated Design Computing

The grand challenge of producing computational constructs that match our notions of designerly activity still remains. Can we build systems that are capable of answering such questions about designerly behavior as: How it is possible for a designer to commence a design without all the necessary information being available? How is it possible for a designer to continue designing when all the necessary information is not available? How is that designers are able to produce novel solutions to what appear to be minor perturbations of existing problems? How is it possible for a designer to produce a different design when later presented with the same requirements.

Drawing ideas from James, Dewey and Bartlett a new kind of computational model, called situated design computing, is being developed. The model is based on notions of "situatedness" and "constructive memory". Situations are constructions that provide a system with a world view that drives what it does. Constructive memory makes memory a process rather than an object in a location. These two concepts allow us to capture memories and turn them into experiences: experiences that are capable of being used in the current situation.

We have extended the notions of computational memory and the role that interactions play in the development of memory. With these we are able to show that designerly behavior is potentially computable.

Funding comes from DARPA, NASA and NSF.

Projects include:

• computational models of constructive memory

• computational models of situatedness

• constructive memory as learning

• constructive interpretations

• situated, constructive memory agents

• situated analogy

• situated design optimization

• situated teams

• situated creativity

• ontology of situated systems

Publications

The foundational papers are:

• Gero, JS (1998) Conceptual designing as a sequence of situated acts, in IF Smith (ed), Artificial Intelligence in Structural Engineering, Springer, Berlin, pp. 165-177. (pdf)

• Gero, JS (1999) Constructive memory in design thinking, in G. Goldschmidt and W. Porter (eds), Design Thinking Research Symposium: Design Representation, MIT, Cambridge, pp. I.29-35. (pdf)

• Gero, JS (2006) Understanding situated design computing: Newton, Mach, Einstein and quantum mechanics, in IFC Smith (ed), Intelligent Computing in Engineering and Architecture, Springer, Berlin, pp. 285-297. (pdf)

• Gero, J. S. and Fujii, H. (2000) A computational framework for concept formation in a situated design agent, Knowledge-Based Systems 13(6): 361-368. (pdf)

• Smith, G and Gero, JS (2005) What does an agent mean by being "situated"? Design Studies 26: 535-561. (pdf).

Papers that develop the research in more detail include:

• Gero, JS (2003) Situated computing: A new paradigm for design computing, in A Choutgrajank, E Charoenslip, K Keatruangkamala and W Nakapan (eds), CAADRIA03, Rangsit University, Bangkok, pp 579-587. (pdf)

• Gero, JS and Kannengiesser, U (2012) Representational affordances in design, with examples from analogy making and optimization, Research in Engineering Design 23: 235-249. DOI: 10.1007/s00163-012-0128-y

• Gero, JS and Reffat, R (2001) Multiple representations as a platform for situated learning systems in design, Knowledge-Based Systems 14(7): 337-351. (pdf)

• Kannengiesser, U and Gero, JS (2002) Situated agent communication for design, in JS Gero and F Brazier (eds), Agents in Design 2002, Key Centre of Design Computing and Cognition, University of Sydney, Australia, pp. 85-94. (pdf)

• Kannengiesser, U and Gero, JS (2019) Design thinking, fast and slow, Design Science, 5(e10), doi: 10.1017/dsj.2019.9

• Peng, W and Gero, JS (2008) Learning while optimising a design: A situated agent-based design interaction tool, AIEDAM (pdf)

• Perisic, M, Storga, M and Gero, JS (2019) Exploring the effect of experience on team behavior: A computational approach, in JS Gero (ed), Design Computing and Cognition'18, Springer, pp. 595-612.

• Perisic, M, Martinec, S, Storga, M and Gero, JS (2019) A computational study of the effect of experience on problem/solution space exploration in teams, in Proceedings of the 22nd International Conference on Engineering Design (ICED19), Delft, The Netherlands, 5-8 August 2019. doi:10.1017/dsi.2019.4Liew, P and Gero JS (2004) Constructive memory for situated agents, AIEDAM 18(2): 163-198. (pdf)

• Saunders, R and Gero, JS (2004) Situated design simulations using curious agents, AIEDAM 18(2): 153-161. (pdf)

• Smith, G and Gero, JS (2004) Describing situated design agents, in JS Gero (ed), Design Computing and Cogition’04, Kluwer, pp. 439-457. (pdf)

People

The people who have or are working on this include:

• Haryuki Fujii

• Kaz Grace

• Udo Kannegiesser

• Nick Kelly

• Jarek Kulinski

• PakSan Liew

• Owen Macindoe

• Gourab Nath

• Wei Peng

• Rabee Reffat

• Somwrita Sarkar

• Vishal Singh

• Greg Smith

• Ricardo Sosa