CUR - The Council on Undergraduate Research

Practical Applications of Strange Attractors in Generative Design

Elliot Dickman, Emil Polyak, Department of Digital Media, Drexel University, 3141 Chestnut St, Philadelphia, PA 19104

This research explores potential applications of strange attractors in generative design algorithms. The chaotic, fractal nature of strange attractors makes them a compelling subject for digital artists. Many artistic visualizations of strange attractors exist, but there are few examples of their use as a practical element to drive procedural algorithms. This research explores methods of visualizing different attributes of strange attractors, and the practical potential of these methods in procedural design. The primary function of strange attractors explored here is their use as an alternative to traditional noise algorithms. While this technique is more computationally expensive than perlin or simplex noise, it is easier to obtain visually unique results. It also benefits from fairly predictable repetitive patterns, while still maintaining a degree of variance to generate unique output values for every seed value. Manipulating the initial conditions of each attractor and implementing different algorithms for mapping the attractor values to output geometry generates unique and organic results. The applications explored here are primarily heightmap generation, variance in procedural object generation, and attractor-driven organic motion, however there is practical potential for this methodology in a wide variety of use cases.

Additional Abstract Information