Independent mobility remains a challenge experienced by people who are blind or visually impaired - a person may reach a general area but may encounter difficulty or lack the information to get to a very specific destination. Currently available navigation techniques are not capable of robust navigation and spatial tasks in the 2-3 meter range. In this research project, where I worked with the Accessibility Lab @ Carnegie Mellon, my team was tasked with creating an interactive tactile tile that could be used to improve this problem through pressure sensing. The hope is that these tactile tiles can be sold in a kit and be used as a low-cost implementation in any environment. I learned a lot about physical prototyping, such as vacuum forming, 3D printing, and arduino, along with how to create survey designs for accessible populations.
Augmented communicators (ACs) who use augmentative and alternative communication (AAC) devices to speak often experience challenges in face-to-face conversations. In particular, they face problems with non-verbal communication, as current AAC devices only provide speech-based communication. My team and I working on a creating a digital "sidekick" that would accompany an AAC's user wheelchair and help them with nonverbals, such as gathering attention or filling in a silence gap during a conversation. More information coming soon - we are working on publishing another paper, which is exciting!
The DARPA Subterranean Challenge is a multi-year competition to rapidly map, navigate and search underground environments using several semi-autonomous ground and aerial robots. The underlying goal of the competition is to ultimately support emergency first responders and combat operators in complex underground environments. Our client for this project was Team Explorer, which consists of 20+ students and faculty in the CMU Robotics Institute and Oregon State University’s Robotic Decision Making Lab. The goal was to help them improve their performance during the competition and see how we could use design to simplify managing multiple aerial and ground robots at the same time. This project was an incredible opportunity - I was able to travel to Colorado for the first leg of the compeition. Team Explorer placed first in the first official round of the competition and my team was able to increase efficiency in detections by 5-10%. More information coming soon!