Click here to view a brief video by Ashok Goel on biologically inspired design.
DANE 2.0 Home
DANE 2.0 was developed at the Design Intelligence Lab at the Georgia Institute of Technology. One of the primary research threads at the Design Intelligence Lab is analogical design. DANE, or the Design by Analogy to Nature Engine, was conceived of as a tool to facilitate particular kinds of analogical design activity, as well as to facilitate research into the cognitive underpinnings of analogical design. DANE 2.0 has four primary functions:
Facilitate Biologically Inspired Design Activity
As the name implies, DANE is intended to be an engine for design. In particular, DANE is intended to facilitate biologically inspired design. Two major challenges of biologically inspired design are (1) finding biological systems that are relevant in a design context, and (2) understanding those systems so that one can extract and transfer the appropriate working principles. DANE 2.0 provides designers with two principle design tools: the ability to create and maintain structured descriptions of biological systems, and the ability to quickly access descriptions useful to a designer's project.
Cognitive Research Platform
In addition to the very pragmatic goal of providing a community of designers with a tool for storing and retrieving information relevant to their design context, DANE 2.0 is designed to help us, as cognitive scientists, better understand the role of structured representation in the context of analogical design. By altering the content and/or behavior of the design tool and observing designers under these different conditions, we hope to learn more about the influence of structured representations on the process of analogical design. The insight garnered from this research, will inform changes that enable greater facilitation of the activity listed in the first goal.
Augmented Intelligence Platform
Although DANE 2.0 is not (yet) capable of "intelligent behavior", such as analyzing designs or making recommendations, it is the goal of this system to provide a mixed-initiative design experience, with the computational tool acting as a (limited) collaborative partner in the design process. The SBF models used in DANE 2.0 were selected particularly because several systems have already been built using them to accomplish, for instance, design analysis and recommendation.
Structured Representation Development
Structured representations used by DANE are based directly on the Structure-Behavior-Function ontology, developed by the Design Intelligence Lab over the course of many years. These representations can be used to explicitly represent biological systems in DANE in a format readable by both humans and machines. As causal-explanation models, we have seen some evidence that these representations facilitate deep learning of biological system working principles. We believe they will also assist in the transfer of those principles into the design domain. Furthermore, the organizational framework implied by the SBF ontology, for instance hierarchical decomposition along functional lines, is used as an organizing principle for DANE 2.0.