Planning: Toward OpenHMI, A Community-Designed Infrastructure for Human-Machine Interaction Research

点击翻译按钮获取中文摘要

Human-machine interaction is an area of research where scientist study how users work together with robots and other devices. It is a broad, cross-cutting area of computing with great potential for supporting wellness, education, and training of people from various backgrounds and with diverse needs. The technology in this field is expanding rapidly, and leverages major areas of development, including computer vision and computer intelligence. However, this area of research is stifled by the lack of tools and software platforms that are accessible, affordable, and appropriate for use with real-world users in replicable real-world research studies. This planning project aims to spend a year understanding this field of research community needs toward designing, what is called “OpenHMI:, an open-source, affordable and modular platform for enabling scalable and accessible research and outreach in human-machine interaction (HMI)”. OpenHMI aims to broaden participation in research by enabling affordable and inclusive real-world studies and data collections. By creating an accessible low-cost and a community accessible platform, this infrastructure will broaden participation of researchers from a range of institutions and levels of research support, in particular opening doors to under-resourced researchers and groups. By involving researchers in the community in the design process, this work establishes an ecosystem that aims to remove barriers to entry for researchers from traditionally under-represented, under-resourced, and/or minority-serving groups and institutions, thereby expanding computing research ideas and projects. The low cost of OpenHMI hardware can also make it an affordable platform for demonstrating introductory computing, computer coding, robotics, and design topics to K-12 and enabling safe hands-on learning.This one-year planning project uses surveys, interviews, and workshops to collect information about the computing community needs toward developing OpenHMI. Specifically, three nation-wide surveys assess detailed needs of the community, including hardware and software infrastructure needed and specific prioritized features of each. Using that input, mockups, early prototypes, and simulations of hardware and software are developed and used in two nation-wide virtual workshops designed to be broadly accessible and inclusive. The workshops serve to train the participants with hands-on experiential activities while collecting information about usability, priorities, and preferences. The large corpus of collected community data informs the design of OpenHMI so that it can be community-informed and designed to ensure its ability to significantly advance research in human machine interaction research by being open-source, accessible, modular and scalable and to facilitate inclusive, safe, privacy-centered and effective integration of this field of research into everyday lives of users to better support ethical data-driven research. The community design process also establishes an open and collaborative ecosystem to accelerate the transition from research studies in small-scale short-term laboratory settings to large-scale long-term deployments in real-world environments.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.