Researchers in the Department of Computer Science and Engineering at the University of Minnesota have discovered a way to explain why people moving through crowds can do so without running into one another.
They call it “time-to-collision,” which is a mathematical law the researchers have used to explore why people move the way they do when many converge on a single location.
Even though crowd modeling is nothing new, the discovery has gained wide attention from local, national and international news outlets because past research has overlooked one major difference between how people move and how molecules move: Humans think.
At the center of this research are co-authors CS&E Assistant Professor Stephen Guy and Lead Researcher Ioannis Karamouzas.
In an interview for Australian Broadcasting Corporation’s Radio National’s (ABC RN) piece, “Avoiding the crush,” Guy described the concept behind “time-to-collision.”
“The original idea was, ‘Let’s just take the way people simulate particles or simulate planets and let’s try that with a crowd. Anybody who is a human being knows that there are really important difference between people and particles or people and planets. One of the most important things is that we can anticipate the future.”
Guy goes on to describe the mechanism that produces these mental calculations in the Boston Globe’s article, “In Crowds, human ‘particles’ follow laws of movement.”
“If you’re close to somebody and they’re walking away from you, you automatically account for [that] fact,” said Guy. “On the other side, if somebody is far away from you, but walking at you quickly, you’re likely to make a large adjustment to your velocity.”
Not only have the researchers found a better way to study crowds, but the implications of their crowd simulation discoveries may have a very real impact on the future, as the Minnesota Daily article, “Designing for a crowd eased by new discovery,” points out.
Karamouzas says in the article that "time-to-collision" could be used to create more accurate computerized simulations of crowds, while Guy explains that the law could help architects study movement beyond looking at patterns, allowing them to design better buildings that avoid “bottleneck situations.”
Guy, who is an expert in interactive computer graphics, including real-time crowd simulation, path planning, and intelligent virtual characters, went on to add that this new research is already being adopted into computer graphics and animation courses at the university.
To dive further into the “time-to-collision” law, read any of the articles mentioned above or listen to ABC RN’s segment, “Crowds and motion.”