Mohan Sridharan, Ph.D. - US grants

The Texas Tech University Research Experiences for Undergraduates Site program provides opportunities for ten undergraduate students each summer to conduct research on two of the most crucial aspects of computing, software engineering and service composition. This REU Site project meets the needs for an undergraduate computing research environment in the Texas South Plains and contributes to producing graduate students in two essential sub-disciplines of computer science. The overall basis for this REU Site is for research efforts that improve the development of software and service-oriented systems in a wide variety of application domains. There are three major objectives for the proposed REU Site: increase the percentage of women and underrepresented minority students; improve the retention of computer science students; and increase those undergraduate students entering into graduate computing education degree programs.

These objectives are met through research that addresses a broad spectrum of means for improving software development. The projects address: software specification generation based on user input and output; construction of a self-managed embedded software system using software specifications; and dynamic techniques for service composition and recovery in a service-oriented environment. The undergraduate student research topics include language development, software tool design and implementation, application development in a variety of application domains to demonstrate applicability, and measurement of the improvements. The technical results of the research efforts with undergraduate students will be disseminated in a broad set of publications. The broadening participation aspects of the broader impacts will be in the women and underrepresented minority students that are recruited for the program, the achievement of success at conducting research, and then continuation on to graduate school.

This funding renews a highly-successful CISE Research Experiences for Undergraduates (REU) site at Texas Tech University. The site covers research topics in cybersecurity, robotics, and software engineering. The intellectual focus of the REU site is on addressing the underlying theoretical concepts in each research project and on developing a research proposal that can be reasonably conducted during the research period. Modeling uncertainty is a common theme within several of the research topics, with application areas that include the Smart Grid, surveillance, assistive care, reconnaissance, unmanned vehicles, climate forecasting, and irrigation management. Cross-discipline seminars will be held at the beginning of the summer to help the students get a smooth start on their research projects.

This site will introduce 30 undergraduate students (over three years) to computing research, with a goal of directing more U.S. undergraduates into graduate computing programs. The site will aid the professional development of the participating students by providing weekly educational activities, such as how to write a research paper and pursue research in ethically sound ways.

This site is co-funded by the Department of Defense in partnership with the NSF REU program.

One approach to expanding the availability of elder care and reducing elder care costs is to increase the number of elders who can live independently in their own homes which is referred to as "aging in place". This is an attractive option because elders prefer to live independently and report better health and quality of life when they do, and because aging in place can reduce the increasing costs of elder health care. However, aging in place still requires caregivers, and caregivers are in short supply. One approach to expanding the availability of caregivers for elders who are aging in place is to develop robots that can serve as caregivers in these homes. Research already suggests that many elders are open to this possibility, especially if it allows them to retain their independence. Furthermore, many elders report that for many tasks they would actually prefer to be helped by a robot rather than another human. As is the case for human caregivers, robot caregivers will be enmeshed in a larger complex social network and technical system beyond just the caregivers and the person who is being cared for. This larger context will need to be understood in order for robotic caregivers to successfully deliver the care that is needed, such as to know who to contact and with what information if the person that they are assisting is not following his or her usual patterns of behavior. This project will explore and describe how caregiving robots should function by analyzing the current form of caregiving in elders' homes, by creating a detailed account of current elder care practices, and by translating this account into design recommendations for caregiving robots.

The analysis will begin by describing the complex socio-technical context of current and future caregiving systems, specifically in terms of the work function and the work domain of the human or robot caregiver. Work function will be described independently of who or what will perform those functions. Work domain will be analyzed using a variety of techniques such as Abstraction-Decomposition Spaces (ADSs) and Contextual Activity Diagrams (ConADs). ADSs will map the relationships between a home's overall objectives and physical resources at several different levels of abstraction, such as by mapping the objective of providing a safe physical environment to different resources that are available to assist with transfers. ConADs will relate different work functions to different work situations, such as by relating the administration of medications to the appropriate time of day and location in the home. Based on these analyses of work function and work domain, the project will draft, refine, and validate formal descriptions of design requirements for caregiving robots.