This page contains information about the work that I am currently doing at Carnegie Mellon as well as the research I did as an undergraduate at the University of Minnesota. Please visit my publications page to learn more about my findings.

PhD Thesis

My thesis, "Robot-Proxy Grounding," focuses on improving human-robot interaction for remote exploration robotics.

Downloads

Dissertation (v3.1 (final), 8/11/08): PDF
Diff (dissertation v2 to v3): PDF
Presentation slides (v2, 7/29/08): PDF
Presentation, handout version: PDF

Abstract

Exploration robotics has traditionally utilized an encoder-decoder model of communication between users and a robot. This means that users construct a plan (sequence of actions) to be sent to the robot; the robot executes the plan and returns data to the users, who then construct another plan. The problem with this interaction paradigm is twofold: (1) users must develop a complex mental model of the robotic system in order to create intricate plans, yet the data returned to them is not necessarily sufficient to help them develop such a model, and (2) the robot does not have the users' specialized domain knowledge, so the robot does not have any way to ensure that how it handles unexpected events in the field is consistent with the users' goals (what the users were trying to accomplish through the plan).

In order to address these problems, this thesis introduces Robot-Proxy Grounding, a novel interaction model for exploration robotics. Robot-Proxy Grounding is derived from common ground theory, a model of human-human communication introduced and experimentally validated by Herbert Clark and his colleagues. Robot-Proxy Grounding is also based on detailed observations and analysis of the Life in the Atacama exploration robotics project, which indicated that a majority of the errors and miscommunications which occurred during the project resulted from a lack of common ground between participants even as the robot became more autonomous.

Because the cost of communication with the remote robot is extremely high, this work introduces the concept of a ``robot proxy,'' a software system which models both the robot's capabilities and the user's goals. Robot-Proxy Grounding occurs as the proxy interacts with the user in real-time in place of the robot so as to promote common ground between the two.

A proof of concept study was conducted which compared the effects of an encoder-decoder planning system and a prototype robot proxy; the study suggested that the use of a robot proxy was effective in improving task efficiency and fostering feelings of collaboration. A full implementation of a robot proxy-based planning system was constructed and evaluated. A user study demonstrated that participants who used the robot proxy were more efficient at the task, collected higher-quality data, and possessed more accurate information about the robot's internal state and its context than participants without a robot proxy. The results suggest that the implementation was successful at promoting common ground with the user, resulting in improved task performance.

Life in the Atacama

This study is part of the larger Life in the Atacama project. The goal of the project is to conduct robotic astrobiology to search for life in the Atacama Desert of Chile. As part of the project, astrobiologists in America spent several weeks controlling an autonomous robot located in the desert. I have been observing the scientists in Pittsburgh while Pamela Hinds, a professor at Stanford University, has been observing the robot and the engineers in the desert. In conducting this work, I have also been collaborating with Carnegie Mellon faculty members David Wettergreen of the Robotics Institute and Sara Kiesler of the Human Computer Interaction Institute.

The Personal Exploration Rover

My advisor Illah Nourbakhsh and his students have designed a miniature robotic rover called the Personal Exploration Rover (PER). The PER was modeled after the two NASA Mars rovers, Spirit and Opportunity, that landed on Mars in January 2004. The PER team designed a museum exhibit in which visitors will have the opportunity to use the PER rovers in a simulated Mars environment to search for life. This exhibit was installed in four science centers across the United States. Because museum employees worked with the rovers in the exhibit for a period of months, the exhibit provided an excellent opportunity to explore long-term human-robot interaction. For this project, Illah and I collaborated with Kevin Crowley and Debra Bernstein of the University of Pittsburgh Center for Learning in Out of School Environments (UPCLOSE).

Distributed Robotics

Most of my research as an undergraduate focused on the improvement of a software architecture for a distributed robotic system. The system was designed to control a group of miniature, two-wheeled robots known as the Scouts. The Scouts were designed by Center for Distributed Robotics. My contribution to the Scout software architecture was the creation of software components to handle load balancing.

The Center for Distributed Robotics went on to create the MegaScout, a larger and more capable robot than the original Scout. I helped design and implement the MegaScout application programming interface, working both at a low level to ensure proper communication between the MegaScout's electronic devices, and at a higher level programming C++ to make these devices easily accessible by programmers wishing to design behaviors for the MegaScout.

If you would like to see demonstrations of these robots, visit this page.

Intelligent Transportation Systems

As an undergraduate, I also spent some time working with intelligent transportation systems, helping to improve a computer vision system designed to detect potential collisions at intersections. I mostly worked on the problem of multiple-camera calibration with the goal of integrating data from multiple cameras into the collision detection system.

User Interface Design

Another project that I worked on as an undergraduate was the design and implementation of a new game-style interface for helping beginning Japanese students learn Japanese characters (hiragana and katakana). The two programs that were the result of my efforts were Kana Card, a flashcard-type program, and Kana Warrior, which encouraged students to read Japanese characters quickly. I was able to conduct some user testing which suggested that Kana Warrior did indeed encourage students to read characters faster than Kana Card. Kana Card and Kana Warrior are open-source software projects available for download at Sourceforge, at The Kana Project homepage. The results of the user study are available from my publications page.