Publication | ACM Symposium on User Interface Software & Technology 2017
Trigger Action Circuits
Leveraging Generative Design to Enable Novices to Design and Build Circuitry
Abstract
Trigger Action Circuits: Leveraging Generative Design to Enable Novices to Design and Build Circuitry
Fraser Anderson, Tovi Grossman, George Fitzmaurice
ACM Symposium on User Interface Software & Technology 2017
The dramatic decrease in price and increase in availability of hobbyist electronics has led to a wide array of embedded and interactive devices. While electronics have become more widespread, developing and prototyping the required circuitry for these devices is still difficult, requiring knowledge of electronics, components, and programming. In this paper, we present Trigger-Action-Circuits (TAC), an interactive system that leverages generative design to produce circuitry, firmware, and assembly instructions, based on high-level, behavioural descriptions. TAC is able to generate multiple candidate circuits from a behavioural description, giving the user a number of alternative circuits that may be best suited to their use case (e.g., based on cost, component availability or ease of assembly). The generated circuitry uses off-the-shelf, commodity electronics, not specialized hardware components, enabling scalability and extensibility. TAC supports a range of common components and behaviors that are frequently required for prototyping electronic circuits. A user study demonstrated that TAC helps users avoid problems encountered during circuit design and assembly, with users completing their circuits significantly faster than with traditional methods.
Download publicationRelated Resources
2024
BrepGen: A B-rep Generative Diffusion Model with Structured Latent GeometryA generative AI model that can learn from and generate boundary…
2024
The Problem of Generative Parroting: Navigating Toward Responsible AI (Part 2)Part two of our series on data parroting in generative AI models…
2021
Cross-Domain Few-Shot Graph ClassificationWe study the problem of few-shot graph classification across domains…
2021
UVStyle-Net: Unsupervised Few-shot Learning of 3D Style Similarity Measure for B-RepsBoundary Representations (B-Reps) are the industry standard in 3D…
Get in touch
Something pique your interest? Get in touch if you’d like to learn more about Autodesk Research, our projects, people, and potential collaboration opportunities.
Contact us