CAREER: Drones in Public: Foundational Interaction Research

点击翻译按钮获取中文摘要

Small Unmanned Aerial Vehicles (sUAVs), or drones, are one of the most adopted types of robot and have become ubiquitous in outdoor public spaces for applications in agriculture and public safety. But design limitations and lack of interaction principles have limited their adoption for indoor uses. The goal of this research is to discover broad interaction principles for communicative sUAV flight paths in various spaces and with diverse users, so that when design limitations such as indoor navigation and battery life are solved the drones will be capable of safe, efficient, and informative interactions with people. To this end, the first challenge will be to identify key communication failures, or barriers between humans and robots, which can create unsafe and inefficient interactions, and to define strategies for avoiding these problems. Guidelines will be developed for gestural communication from a vehicle to a user in order to generate safer and more natural interactions, while respecting personal space requirements. The two central hypotheses underlying this research are that gestural communication will allow better visibility of robot state or knowledge to bystanders while being applicable to existing platforms without modifications, and that interaction distances will need to adapt to the design of the vehicle, the user, and the environment to maintain comfort and efficiency in interactions. This work will have broad impacts both on education and on society through directed activities to improve understanding of human-robot interaction. Education will be supported through activities such as lesson plans for K-12 students, integration of the research into an undergraduate human-computer interaction class, and informing a semester-long course for both undergraduate and graduate students on human-robot interaction (HRI). Additional outreach efforts will include a month-long interaction with 6-8th grade students from local Title 1 schools teaching K-5 students about the scientific method, and an evening class for adults to inform them about current robotics and HRI research. The PI will continue her ongoing efforts to broaden participation in computer science research. This research will encompass three parallel objectives related to gestural communications by sUAVs and comfortable distancing. Objective 1 will focus on communicative motions of state information from the vehicle; gestures will be generated through participatory design and tested using methods from the social sciences. Objective 2 will also be informed by methods from the social sciences, but will focus on distancing with sUAVs and will take advantage of an existing room in the PI's lab with movable walls, video cameras, and a motion capture system. Each of these two research objectives could be carried out in the absence of progress from the other; however, their exploration together is natural since the gestures are conveyed by different vehicles, scaled to spaces, and the robots perform tasks which may require human involvement. This combined investigation will comprise Objective 3, which will focus on efficient signaling, incorporating the distancing work into the flight paths, and reception with users performing a variety of tasks. Project outcomes will transform close-proximity HRI by enabling sUAVs to generate communicative flight paths that are optimized based on likelihood of receipt and comfort in highly independent human-robot teams.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.