Robotics work targets stampedes

HAIBO HE, an associate professor of electrical, computer and biomedical engineering at the University of Rhode Island, has been awarded a three-year, $282,000 grant from the National Science Foundation in order to develop co-robots to lead people during an emergency evacuation. As a part of the grant, He will also engage students enrolled in the robotics program at the Community College of Rhode Island, as well as at local high schools. Over the past several years, He has received numerous grants from the Army Research Office, the Defense Advanced Research Projects Agency and the National Science Foundation in order to solve critical engineering application problems, including cybersecurity, human-robot interaction and smart grids.

What inspired you to develop this technology?

Crowd stampede is one of the most serious and disastrous forms of collective human behavior, especially under emergency situations. Even under normal conditions, such as directing people exiting large venues like sports arenas and concert halls, optimizing pedestrian flow is still a major challenge. This inspires my collaborators and myself to develop innovative robotics research to tackle such a challenge.

Could you describe how co-robots will be used in an evacuation situation?

- Advertisement -

We will develop three major components to accomplish this goal. First, we will study and develop models for human behavior under emergency, such as panic models. Second, we will develop human-robot interaction at the individual level and at the group level (i.e., collective human dynamics). Finally, we will develop adaptive control approaches for the robots to interact with local crowds in order to optimize pedestrian flow under an evacuation situation.

How will you develop an adaptable algorithm?

The key is to develop an adaptive learning and control approach with the integration of human-behavior modeling and robotics control. Specially, we will investigate different types of feedback control approaches to accomplish the goal. During this project, we will use camera systems to observe the human flow pattern, and feed this information to the robotics. •

No posts to display