Alla Safonova, Ph.D. - US grants

"This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)."

The objective of this project is to build a state-of-the-art Human Motion Acquisition Facility capable of capturing human motion data using an array of synchronized multiple sensors. It includes hardware and software necessary to capture simultaneously multi-modal aspects of human performance including human movement, ground reaction forces, video data, muscle activation patterns, heart rate, temperature, eye movement, etc. Importantly, all the data is synchronized with each other.

The facility enables researchers to explore new frontiers in human motion synthesis. In particular, the captured data is used to study novel compact representations of human motions and to develop new algorithms to efficiently synthesize detailed realistic motions of a single character and crowds. The data is also used to build, animate and populate virtual worlds. The facility also greatly benefits existing computer graphics degree programs at UPenn (Ph.D, Masters, Undergraduate, and summer high school students). The use of the facility is an integral part of the curriculum in these programs.

Currently, there are no extensive, publicly available, high quality databases of human motion data synchronized with data from other sensors. All the data we obtain in this project will be made publicly available. The facility will therefore enable new research ranging from human motion synthesis in computer animation to medical studies of people with disabilities to researching novel simulators. As a result, it will foster inter-disciplinary research. The facility will also make computer graphics programs at UPenn even more exciting and will help in attracting more women to computer science.

This project will: (1) Build an experimental setup for conducting evacuation experiments inside 3D virtual online communities. (2) Analyze the realism of the experimental setup and research the ways to improve it. (3) Investigate the use of intelligent signs for controlling peoples' behavior in emergency situations. One of the major aspects in the design of buildings, stadiums and city blocks is their suitability for evacuation. Peoples' lives depend on how quickly these constructions can be evacuated in the emergency situations such as fires, earthquakes, terrorist attacks and collapsing structures. As a result of the 9/11 terrorist attacks however, it became apparent that current approaches to building design in regard to emergency evacuation for various building typologies such as high-rises, airports and stadiums need to be re-examined. In particular, the designs must be carefully evaluated against evacuation procedures. This raised a major concern about the lack of tools that would allow robust predictions of realistic human movements and interaction in the designed environments. It is clearly impractical to establish live experiments with thousands of people evacuating every possible building design for every possible emergency condition.

The research will not only evaluate virtual-world information technology as a tool for building design to facilitate evacuation, but it will also develop new information technology that could be incorporated in real-world buildings. For example, intelligent exit signs could automatically direct people during emergencies based on the known information about the construction, the type and location of emergency, and the current distribution of people. Building an experimental setup for conducting evacuation experiments within 3D virtual online communities requires answering such unorthodox questions as how to attract research subjects to the experiments, how to motivate the participants to evacuate buildings when emergency occurs, and what setup should be made to evaluate the efficiency of evacuation as truthfully as possible. The analysis and improvement of the realism of the setup requires research in realistic simulation of natural effects such as fire, real-time realistic human motion synthesis for hundreds of characters and developing effective means of measuring the immersiveness of participants. Finally, the design of control strategies for intelligent signs requires the research of real-time decision-theoretic planning under uncertainty algorithms suitable for the control of massive multi-agent systems.

The safety of buildings is of critical importance to any society. Natural hazards, terrorist attacks and fire accidents cause losses of thousands of lives every year. By designing buildings that are easier to evacuate, engineers can save a significant fraction of these lives. Yet, there is a lack of tools that would allow engineers to conduct high-fidelity evaluations of building designs in terms of evacuation efficiency. This research is directed towards providing engineers with access to conducting such large-scale high-fidelity experiments at drastically lower expense than previously possible. Progress in this direction will also have positive educational impacts. The project will play a significant part in the classes offered in an undergraduate Digital Media Design program, which has about 50% female students.