Design Assistant

When human designers work together, they can generate designs they may never have considered on their own. Similarly, computer agents have been used extensively as tools to augment or exceed human design processes. Humans and machines tend to have very different strengths and weaknesses, and recent work has looked at ways that human designers can improve computational design or vice versa.

Our research asks what would happen if humans and machines were instead to work together, physically, side by side, as peer collaborators in design tasks. How might they infer and account for each others’ tendencies, preferences, and mental models in real time? And what are the effects of this palpable interaction on design processes, experiences, and outcomes?

Publications

. Hammers for Robots: Designing Tools for Reinforcement Learning Agents. DIS, 2021.

Project Source Document

. Design Intention Inference for Virtual Co-Design Agents. IVA, 2020.

PDF Project

. Side-by-side Human-Computer Design using a Tangible User Interface. DCC (Best Paper), 2018.

PDF Project