Emergent Behaviour in Co-Evolutionary Design

An important aspect of creative design is the concept of emergence. Though emergence is important, its mechanism is either not well understood or it is limited to the domain of shapes. This deficiency can be compensated by considering definitions of emergent behaviour from the Artificial Life (ALife) research community. With these new insights, it is proposed that a computational technique, called evolving representations of design genes, can be extended to emergent behaviour. We demonstrate emergent behaviour in a co-evolutionary model of design. This co-evolutionary approach to design allows a solution space (structure space) to evolve in response to a problem space (behaviour space). Since the behaviour space is now an active participant, behaviour may emerge with new structures at the end of the design process. This paper hypothesizes that emergent behaviour can be identified using the same technique. The floor plan example of Gero and Schnier (1995) is extended to demonstrate how behaviour can emerge in a co-evolutionary design process.

[1]  John S. Gero,et al.  Shape Emergence by Symbolic Reasoning , 1994 .

[2]  Woalfgang Jonas,et al.  Design as problem-solving? or: Here is the solution —what was the problem? , 1993 .

[3]  Luc Steels,et al.  Towards a theory of emergent functionality , 1991 .

[4]  Mary Lou Maher,et al.  Modeling design exploration as co-evolution , 1996 .

[5]  John S. Gero,et al.  Artificial Intelligence in Design ’94 , 1994, Springer Netherlands.

[6]  Peter Ross,et al.  Solving design problems by computational exploration , 1994 .

[7]  John S. Gero,et al.  Creativity, emergence and evolution in design , 1996, Knowl. Based Syst..

[8]  John S. Gero,et al.  An evolutionary process model of design , 1992 .

[9]  J. Poon,et al.  Co-evolution of the fitness function and design solution for design exploration , 1995, Proceedings of 1995 IEEE International Conference on Evolutionary Computation.

[10]  Yu-Tung Liu,et al.  Some phenomena of seeing shapes in design , 1995 .

[11]  Stephanie Forrest,et al.  Emergent computation: self-organizing, collective, and cooperative phenomena in natural and artificial computing networks , 1990 .

[12]  John S. Gero,et al.  Object Emergence in 3D Using a Data-Driven Approach , 1994 .

[13]  Steven M. Smith,et al.  Creative Cognition: Theory, Research, and Applications , 1996 .

[14]  Pattie Maes,et al.  Situated agents can have goals , 1990, Robotics Auton. Syst..

[15]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[16]  John S. Gero,et al.  Towards a model of exploration in computer-aided design , 1994, Formal Design Methods for CAD.

[17]  Maja J. Matarić,et al.  Designing emergent behaviors: from local interactions to collective intelligence , 1993 .

[18]  David B. Fogel,et al.  Evolutionary Computation: Towards a New Philosophy of Machine Intelligence , 1995 .