A mobile robot has been built that can detect the distances between humans and roll over to remind them they're too close. It's not exactly Robocop, but perhaps automated rule enforcement has to start somewhere.
In PLOS ONE the University of Maryland, College Park PhD student Adarsh Jagan Sathyamoorthy and colleagues describe a robot they have built with depth perception allowing it to determine the distances between objects, or people.
Since the robot is also mobile, it can use the same perception to navigate its way through a crowded space to people who are too close together and encourage them to move to a safe distance. The paper claims this is done “discretely, using a mounted display,” although if a robot elbows its way through the crowd we wouldn't count on nobody noticing where it is going.
The authors have set the distance at the CDC-recommended two meters (7 feet), but with increasing evidence virus particles frequently travel further than that in still air, presumably the robot can be reprogrammed if policies change.
Detecting the distance between stationary members of a crowd is probably easy enough that publication in a prestigious journal might look like overkill, but Sathyamoorthy's creation can go much further. Not only can it identify ongoing breaches of physical distancing between people who are moving together, but it can also prioritize those whose proximity is most likely to facilitate transmission, for example larger groups.
The robot has proved its effectiveness on volunteers imitating crowd scenes using its own cameras and LiDAR system, and can also hook into CCTV cameras to really make you feel like big brother is watching.
To top it off, the as-yet-unnamed robot is equipped with a thermal camera to spot people with potential fevers. Not only will this allow it to send those people straight to the top of its priority list for those who must be kept apart, but it can also use this information for contact tracing. The authors claim this will be combined with de-identification to protect privacy but there's no guarantee that won't ever change.
The paper acknowledges the fact the robot can't “distinguish between strangers and people from the same household,” which might limit its usefulness where families are common.
Irrespective of how effective such robots prove in real-world situations, just knowing they are out there might be enough to keep some people from leaving their houses – potentially an effective method for disease control on its own.
Much as this may sound like something from a dystopian science fiction novel, not helped by the authors referring to their creation as a COVID surveillance robot but the authors make a fair point about the alternative. "A lot of healthcare workers and security personnel had to put their health at risk to serve the public during the COVID-19 pandemic,” they said in a statement. “Our work's core objective is to provide them with tools to safely and efficiently serve their communities."