Robotics and computer vision research projects

CHS: Small: Socially-Aware Navigation

  • National Science Foundation, award #IIS-1719027
  • PI: David Feil-Seifer, Co-PI: Monica Nicolescu
  • $500,000
  • Sept. 1, 2017 - Aug. 31, 2020

The promise of robots to serve in assistive capacities for care and cooperation is invaluable. Socially assistive robots have the potential to make a positive impact on people in need of physical therapy, children in need of educational intervention, children with developmental disorders, and patients post-stroke. However, socially inappropriate behavior can harm the long-term adoption of these assistive platforms. In particular, while typically understood only as getting from point A to point B, robot movement can constitute communication without any use of words, and even when not intended. This can have serious ramifications for the success of a social interaction between one or more people and a robot. A robot in a long-term interaction (say in a hospital, school, or home setting) may have dozens or even hundreds of navigation destinations every day; sometimes, the "optimal" navigation behavior may not match the socially appropriate course of action. Socially inappropriate navigation behavior has been attributed as a major impediment to the long-term adoption of assistive robots in hospital settings. To address that problem, this project will develop a socially-aware navigation planner for autonomous robots that mimics the social conventions of human-human interactions, and will thereby advance the applications of autonomous robot technology.

Model-based navigation planning for social appropriateness will involve observing the activities and intentions of people in social proximity to a robot and/or other people, and then selecting robot trajectories that are both spatially and socially optimal. This work will then build upon that foundation by expanding the library of spatial communication that a robot can understand and communicate in an autonomous fashion. The research will incorporate high-level features, such as intent recognition, into the model development along with low-level features, such as interpersonal distance, to create models of complex behavior considering both the spatial scene and the inferred intention of the actors in that scene. The Pareto Concavity Elimination Transformation (PaCcET) multi-objective optimization framework will be utilized to develop socially appropriate navigation behavior policies. The developed planner will be validated on several types of robot platforms.

Deep Reinforcement Learning Framework for Robotic Manipulation

  • NASA Space Grant Consortium Research Infrastructure
  • PI: Hung M. La
  • $45,000.
  • September 2018- April 2019

Designing Collaborator Robots for Highly-Dynamic Multi-Human, Multi-Robot Teams

  • Office of Naval Research
  • PI: Monica Nicolescu, Co-PI: Dave Feil-Seifer, Mircea Nicolescu
  • $656,511
  • April 2016 - March 2019

The goal of this proposal is to establish at the University of Nevada, Reno a robotic infrastructure that would enable groundbreaking research in a wide range of topics related to mobile articulated robots. The proposed system consists of the Baxter humanoid robot and the PR2 robotic platform, whose extensive sensory, motor and computational capabilities open up the possibility of addressing research questions that could not have even been formulated in the context of the current publicly available robotic systems.

Developing Earth & Space Science Curricular Units to Ameliorate Problems of Teaching Practice and Support Student Learning

  • Nevada NASA Space Grant Consortium (NVSGC)
  • PI: Candice Guy-Gaytan, co-PI: David Feil-Seifer, Elizabeth de los Santos, Adam Kirn
  • $25,000
  • 2018-2019

Highly Accurate Image Processing for Concrete Images

  • Japan Nine Sigma, under Penta-Ocean Constructions Co., LTD
  • PI: Hung M. La
  • $150,000
  • November 2018- October 2019

High Throughput Tissue Microarray (TMA) Construction and Interpretation to Facilitate Biomarker Discovery - phase 2

  • University of Nevada, Reno
  • Co-PI: Monica Nicolescu (PI: Sanford Barsky, Co-PI: Emil Geiger)
  • $20,000
  • May 2013 - present

The goal of this project is to develop technology utilizing mechanical engineering, image analysis and knowledge of tissue patterns of disease that will enable us to analyze human tissues. Specifically using collaborative expertise, the project will develop a method of analyzing thousands of individual samples in a high throughput manner. These samples will be robotically constructed in a tissue microarray (TMA) and then interpreted through virtual microscopy and the application of imaging algorithms, which will facilitate biomarker discovery.

Inspecting and Preserving Infrastructure through Robotic Exploration (INSPIRE)

  • U.S. Department of Transportation
  • PI: Hung M. La, Co-PI: Sushil Louis
  • $721,995 (University of Nevada, Reno portion)
  • November 2016 - September 2022

Intent Recognition for On-Water Dynamic Maritime Domains

  • Office of Naval Research
  • PI: Monica Nicolescu, co-PI: Mircea Nicolescu
  • $685,355

Leveraging human electrocardiogram signals for UAV security enhancement

  • Nevada NASA Seed Research Infrastructure Development
  • PI: Hung M. La, Co-PI: David Feil-Seifer
  • $50,000
  • September 2018- June 2019

NRI: Multi-Modal Characterization of DOE-EM Facilities (in collaboration with Carnegie Mellon University)

  • Kostas Alexis
  • $1.3 million ($349,000 for University of Nevada, Reno)
  • September 2016 - September 2019

Research focuses on developing aerial robots to explore post-disaster nuclear sites.

NVSCG: Robotics and Big Data Curriculum for Undergraduate and Graduate Students of UNR College of Engineering.

  • Nevada NASA Space Grant Consortium Research Infrastructure
  • PI: Hung M. La, Co-PIs: David Feil-Seifer and Tin Nguyen
  • $50,000
  • July 2018- April 2019

RET Site: Cross-disciplinary Research Experiences on Smart Cities for Nevada Teachers: Integrating Big Data into Robotics

  • National Science Foundation (NSF), award #EEC-1801727
  • PI: Kostas Alexis, co-PI Lei Yang, Senior Personnel David Feil-Seifer
  • $581,073
  • 2018-2021

This Research Experiences for Teachers (RET) Site at the University of Nevada Reno aims to deliver a unique holistic research experience to K-12 teachers of Nevada in relation to future smart cities and the cutting edge topics of robotics and big data, as well as their intersection. The overarching goal is to enable a critical mass of teachers in Nevada to be able to help their students develop a passion for these domains, strengthen their critical thinking, and broadly nurture an interest to excel in STEM fields. Research in the respective domains is currently of paramount importance in order to address the multitude of open challenges to achieve truly autonomous driving and the realization of smart cities. The relevance and significance of cross-disciplinary research in robotics and big data cannot be overstated in the context of STEM education and its potential is daily growing. This site will attract K-12 teacher participants to STEM education by providing tangible research experiences and appropriately designed knowledge modules, in turn enhancing the STEM education for their students and further designing and deploying transferable research and education tools for the school environment.

This project aims to nurture an interest and passion for robotics and big data research to K-12 teachers of all levels, especially from the perspective of their intersection for smart cities. The RET Site will provide a tangible experience through a) educational activities on the fundamental principles of autonomous driving, machine learning, and big data, b) a custom-designed, miniaturized, and transferable test-bed of smart cities called "Robocity", and c) hands-on activities on an instrumented electric bus and a drive-by-wire Lincoln MKZ autonomous car. Through the envisioned activities, it is anticipated that Nevada teachers will be actively engaged and gain knowledge and research experiences in robotic perception and machine learning, control, path planning, big data analytics, and big data system design. To enhance the quality of STEM education in local K-12 schools at all levels, a team with both senior and young faculty has been assembled to support the development of curricular modules and provide a set of mechanisms that allow research to be seamlessly transferred to the school environment. The project will engage a thorough plan for project evaluation and a participant recruitment process that focuses on underrepresented groups and respective student populations. To ensure that the developed skills acquired by teachers are transferred to the classroom, the PI team will engage in follow-up activities with the RET participants to support the installation of Robocity. The PI team will also offer demonstrations and tangible exercises at schools, as well as workshops at the University of Nevada, Reno.

REU Site: Collaborative Human-Robot Interaction

  • National Science Foundation, award #CNS:1757929
  • PI: David Feil-Seifer, Co-PI: Shamik Sengupta
  • $360,000
  • Feb. 1, 2018 - Jan. 31, 2021

Human-Robot Interaction (HRI) has the potential to affect several real world domains such as hospitals, homes, schools, offices, or infrastructure sites where robots are uniquely able to assist a human team member to achieve a common goal. However, there are many challenges for effective human-robot collaboration that impinge on this promise, such as communicating with human partners using natural language, providing information to an operator through a visual interface, or quickly providing information so that a person is able to offer assistance. We propose an HRI direction that provides several closely-related projects for undergraduate research experiences. The project team will mentor 10 undergraduates each summer, pursuing research directly impacting assistive robotics, in the following technical areas: development of autonomous robot capabilities targeted for care environments, such as networked robotics devices cooperating with care professionals to perform complex tasks in support of aging in place; human command-and-control infrastructure for operations that utilize robots for technology support; and connectivity and security for a network of robots, computers, and embedded devices collectively used to mission-critical goals while utilizing unused parts of the wireless spectrum. The goal of this Research Experiences for Undergraduates (REU) site is to develop and evaluate robotic systems whose function is to help bridge the human-robot collaboration gap and achieve the above objectives efficiently and effectively. The summer activities for undergraduates will provide hands-on science and engineering activities related to current research projects, and professional development with training sessions on writing a graduate school application and how to apply for fellowships to support graduate education.

The site will develop new autonomous robot capabilities and supporting network and data science technology to address real-world challenges of operating autonomous systems in hospital, clinic, home, and infrastructure environments. This site will develop solutions for semi-autonomous robot behavior with humans in the loop, wireless network connections in rapidly changing frequency domains, and processing high volumes of real-world data. The proposed site presents five projects related to these computing domains: Empirical Study of Socially-Appropriate Interaction; Language-Based Human-Robot Collaboration; Autonomous Robotic Exploration of Underground Mine Environments; Multi-robot Collaboration and Human in-the-loop for Safe, Accurate and Reliable Inspection; and Network Management of Heterogeneous Robotics Devices in a Wireless Environment. This REU site links these projects together through the common objective of developing assistive technology. Students will develop domain knowledge, mathematical skills, and interdisciplinary competency. The possible intellectual properties resulting from this project will include study on long-term human-robot interaction, wireless networking for challenging signal environments, autonomous robot capabilities for human-robot teaming, and data science tools for processing high volumes of data from real-world systems.

Robotics and Big Data for Undergraduate and Graduate Students of University of Nevada, Reno College of Engineering 

  •  Nevada NASA Space Grant Consortium (NVSGC)
  • PI: Hung La, co-PI David Feil-Seifer
  • $25,000
  • 2018-2019

University/Museum Partnership For Informal Education in Robotics

  • Nevada Space Grant Consortium (NVSGC), award #NNX15AI02H
  • PI: David Feil-Seifer, Co-PI: Adam Kirn
  • $25,000
  • July 1, 2018 - April 9, 2019

Computing education is becoming a critical component for success in the modern job market. Computing is continually highlighted as a national priority of K-12 education. The experience and training required are a barrier to the introduction of courses or other educational models with computing content. In this proposal, we extend prior work that developed robotics modules for informal K-12 instruction in robotics and computing and utilize these materials for outreach to Nevada students through a collaboration with the Nevada Discovery Museum and the University's Mobile Engineering Education Lab (ME2L).

The current training paradigm for robotics instruction relies on primarily graduate study. Some advanced undergraduate courses may be available, but students typically have access to at most one or two of these courses. However, at the K-12 level, few robot enrichment activities exist. The result of this configuration is that students may not even learn that robotics is a field that they would like to pursue until late undergraduate education at the earliest. This could severely depress interest in robotics.

A wealth of literature suggests, however, that children learn best when presented with playful, real world, “learn by doing” challenges (see, for example, Dewey, 1938; Vygotsky, 1978). Unfortunately, much education concerning computer programming may be currently framed as equivalent to asking students to “eat their vegetables” and memorize content rather than participating and learning from activities that are inherently rewarding. Computing can be difficult to teach, especially at the pre-college level. One reason for this difficulty is that computing relies on several other skills, such as algebra, sequential logic, and planning. Subsequently, there is a dearth of computing education in Nevada.

Rather than focus on programming in a specific computer language, the proposed modules will focus on the fundamentals of computing: instructions, decisions, and loops; as well as the fundamental problems in robotics: sensing, planning, and actuation. Computing skills are often used in day-to-day activities. A recipe, for example, is a planned sequence of steps that when executed result in a dish. Directions are a similar example of programmatic sequences.

This proposal will engage with the Terry Lee Wells Discovery Museum and the UNR Mobile Engineering Education Lab (ME2L) to develop and execute this type of skill-based real-world inspired computing instruction and to disseminate them widely to students from the Reno area. These activities have been shown to encourage interest and identity in STEM careers and education. We are excited to expand on this content and bring this to a wider audience.