The University of Arizona

Home     About     Colloquia     People     Research     Undergraduate Program     Courses     Get Involved     Support     Contact Us

SISTA Projects

SISTA is home to many interdisciplinary projects, involving faculty at the University of Arizona and other universities. Please browse the following descriptions of projects and laboratory pages, and visit our Recent News page for articles and media about SISTA projects and community.

ARRG: Arizona Robotics Research Group

picture of a small robot

The goal of the Arizona Robotics Research Group (ARRG) is to gain a greater understanding of learning and development in humans by creating artificial agents- both softbots and robots- that have to solve many of the same problems that the developing human brain has to solve. Many robots in the ARRG are essentially baby robots, which are created with minimal skill sets and uninterpreted sensors and go on to develop higher-level motor and perceptual skills from prolonged interaction with the world. Other robots are designed primarily to learn about human-robot interaction, particularly human-robot teaching. Central to most of this work is focused on the idea of \"curiosity based learning\"- that is, robots who\'s primary drive is an intrinsic desire to better explain and control the world rather than the desire to maximize some external (human defined) reward.
>Learn more

B2E2: Student Scientists

Students looking at a projector screen

In collaboration with scientists at the UA Biosphere 2 and School of Natural Resources, we are developing information technology and methods to bring ongoing, cutting-edge science experiments into K-12 classrooms. The goal is to provide a direct conduit between scientists and the classroom to simultaneously enhance science education and enable experimental studies on a large scale. We have successfully run several variations of an evapotranspiration study in local schools and at the Biosphere 2; background information about the study was incorporated into the classroom science curriculum, and results from the experiments have directly informed the evapotranspiration study.
>Learn more

Bootstrap Learning: The Bootstrapped Learning Project

computer generated entities on a virtual plane with cubes

The Bootstrapped Learning Project seeks to develop an electronic student capable of being taught new concepts through natural human instruction. There are many facets to the challenge of building an instructable student, and we are participating in a multi-team DARPA program focussed on this problem. We are developing algorithms to notice information and patterns that are not the explicit subject of the lesson being taught, both in teacher/student dialog and world events.
>Learn more

MathGap: Helping English-Language Learners Do Better in Math

photo of Beal

This project studies the problem-solving strategies of students who are learning English, using word problems from the AnimalWatch online math tutoring system. The results will drive adaptations of the system to help ELLs improve their math skills.
>Learn more

Vision Research: Computer Vision

scene understanding dining room

The SISTA computer vision group studies automatically inferring semantics from image data in a wide range of contexts including fundamental research into scene understanding, inferring human activity from video data, extracting 3D structure from biological images, and improving access to image and video data. We take a statistical modeling approach where we build and/or learn models that explain image data based upon high level representations of objects, scenes, and activities. We then use Bayesian inference for interpreting visual data as evidence for what is in the world with respect to those representations. The vision group has roughly a dozen active projects. Undergraduate researchers contribute to many of these.
>Learn more