Recent studies have shown that novel continuous dropout methods can be viewed as a Bayesian interpretation of model parameters, though most such studies have shown results using normal distributions. As the posterior distributions over neural network nodes and parameters are intractable, given that they are a result of artificial construction...
This work describes the design and modification of a commercial children’s ride-on car to effectively collect usage data. The design is cheap and could be easily modified to gather more information. This will serve as a viable way to collect information about how motorized independence can help children with developmental...
Robotic global navigation is often based on the assumption that the world is static. It assumes that obstacles present in maps will be there later and new obstacles will be dealt with by a local planning algorithm. When robots interact in an environment inhabited by people, this assumption does not...
Despite an increase in the number of people who rely on manual wheelchairs, there are still substantial economic barriers to affordable and accessible localization systems. As a result, there is a pressing need to build a versatile yet low cost localization system for manual wheelchairs. Such systems allow users to...
Robots have the potential to protect health care workers, provide patient care, and ultimately save lives in infectious disease outbreaks. Nevertheless, infectious disease outbreak scenarios present unique technological and social challenges for robotics. This work explores what robots can and should do in the fight against infectious diseases. We present...
Numerous diseases and injuries can limit a person's ability to perform everyday tasks -- things like getting dressed, bathing, and eating. Anything that requires physical activity can be affected; even simple things like turning on the lights can become difficult or impossible. Until recently, the only way for a person...
This dissertation focuses on personal privacy in human-robot interaction, which we call "privacy-sensitive robotics." Our understanding of "privacy" is very broad, including not just information privacy but also physical, psychological, and social privacy. We begin by surveying the scholarly literature on privacy and talking about why it applies to interactions...
When treating highly infectious diseases, patient compliance with caregiver recommendations is crucial. Many patients who contract Ebola Virus Disease have symptoms including vomiting and severe diarrhea and die from the resulting dehydration if they do not drink enough fluids. Telepresence robots are useful in the context of Ebola treatment for...
Automating pick-and-place operations for large steel castings in a foundry is challenging, particularly from a perception point of view. Accurately estimating the identity and orientation of a given part is an open and complex problem, especially if that part sits in a bin of very similar parts. The problem is...
This thesis aims to determine the impact that Augmentative and Alternative Communication (AAC) devices have on social interactions, and then improves the AAC user experience through a user focus design process. AAC devices enable people who cannot speak to communicate with others. Unfortunately, they are tedious to use and make...