A new ‘robocop’ which can identify social distancing in crowds and warn people to move apart has been developed by scientists.
The surveillance robot could help reduce the spread of the virus and also aid contact tracing, researchers from the University of Maryland said.
They explained how the mobile bot can detect people in crowds who are not observing social distancing rules using its own camera and sensor.
The robot can even tap into an existing CCTV system, if available, to detect breaches and can navigate to people to “encourage” them to move apart via a text that appears on a mounted display.
The study lead author, Adarsh Jagan Sathyamoorthy, said: “Previous research has shown that staying at least two metres apart from others can reduce the spread of Covid-19.
“Technology-based methods – such as strategies using WiFi and Bluetooth – hold promise to help detect and discourage lapses in social distancing.
“However, many such approaches require participation from individuals or existing infrastructure, so robots have emerged as a potential tool for addressing social distancing in crowds.”
The PhD student, who specialises in robotics, and his colleagues have now developed a new way to use an autonomous mobile robot for that purpose.
Mr Sathyamoorthy added: “The robot uses a novel system to sort people who have breached social distancing rules into different groups, prioritise them according to whether they are standing still or moving, and then navigate to them.”
He said the system employs a machine-learning method known as Deep Reinforcement Learning and Frozone, an algorithm previously developed by several of the same researchers to help robots navigate crowds.
The researchers tested their method by having volunteers act out social distancing breach scenarios while standing still, walking, or moving erratically.
Their robot was able to detect and address most of the breaches that occurred, and CCTV enhanced its performance.
Mr Sathyamoorthy said: “The robot also uses a thermal camera that can detect people with potential fevers, aiding contact-tracing efforts, while also incorporating measures to ensure privacy protection and de-identification.”
He said further research is needed to validate and refine the system, such as looking at how the presence of robots impacts people's behaviour in crowds.
Mr Sathyamoorthy added that the group's “core objective” was to provide healthcare workers and safety personnel with “tools to safely and efficiently serve their communities.”
The findings were published in the journal PLoS One.Internet Explorer Channel Network