This Robot Disinfects Classrooms

When COVID lockdowns began in 2020, the world had to find new solutions to new problems. How do you disinfect both the air and surfaces of large indoor spaces – and do so repeatedly, at scale, during a labor shortage? An interdisciplinary group of Purdue students collaborated on a solution: an autonomous robot that disinfects classrooms.



The robot is the brainchild of Richard Voyles, the Daniel C. Lewis Professor of Robotics in the Purdue Polytechnic Institute. Voyles is well known for creating unique robots, including those used in search-and-rescue operations. When the pandemic arrived, Voyles and his students shifted their attention to how those robots could positively contribute during the lockdown.

“We did some literature research, and found that there had been many projects aimed at building disinfecting robots that use ultraviolet light,” said Haoguang Yang, a PhD student on Voyles’ team. “There were also robots had been experimenting with using disinfecting sprays. But the problem with both of these approaches is that they are irritating to humans, and thus could only be used in rooms without occupants. What could we do in rooms with people?”

Voyles enlisted the help of Luciano Castillo, the Kenninger Professor of Renewable Energy and Power Systems in Mechanical Engineering, who brought in his expertise in fluid mechanics. “When somebody coughs, they release a plume of turbulent flow,” said Jhon Quiñones, a Ph.D. student under Castillo. “With computational fluid dynamics, we can predict the spread of these particles. We had the computer models, and they had the robots; so we saw the opportunity to team up and integrate them together.”  

Yang began adapting a standardized industrial robot, which uses four drivable wheels to navigate narrow paths. They installed an off-the-shelf HEPA air filter, and began conducting tests in a mock classroom with two rows of three seats each. On each seat was a water fogger, simulating the droplets that would come from human occupants. Using a combination of sensors and computational fluid dynamics models, the robot safely traversed through the mock classroom with its air filter, and spending longer in the areas where there were more droplets. Using particle counters, the team determined that this approach captured more particles than if the air filter sat stationary in the room.

But there was still room for improvement, particularly in the air filter. “Our goal is to minimize the amount of exposure that occupants would have with infected droplets in the air,” said Antonio Esquivel, PhD student in agricultural and biological engineering. “With our models, we know where the particles are, but we needed an airflow design that converged on those particles to capture them.”

Adapting work they had previously done with wind turbines, they created an entirely new air filtering system, which they call the Bernoulli Air Filtration Module. They added a pair of airfoils and a perforated cylindrical filtration pipe, creating a low-pressure zone between the airfoils which helps to drive the airborne particles directly into the intake of the HEPA filter, without generating a turbulent wake. This design, coupled with the robot’s calculated motion paths, showed a huge improvement, reducing exposure to particles by up to 26% versus typical air filtering setups.

Their research has been published in Robotics and Autonomous Systems.

“We don’t know of any other commercial robots that use air filtration for disinfection,” said Voyles. “If you’re trying to disinfect a room where humans are present, we’ve shown that this system is more effective than any other option.”

And they didn’t stop there. In addition to air disinfection, the team is building surface disinfection into the robot. “We are using microjets that spray a disinfecting solution with nanoparticles,” said Tanya Purwar, a Ph.D. student in mechanical engineering. “There are microjets on the bottom, which disinfect the floor. We also have a dynamic robot arm on the top with three different nozzle sizes, because different surfaces respond to the disinfecting solution differently. Our goal is for the computer to recognize this autonomously, see that a surface is wood or steel or fabric, and adapt its robot arm to spray in the most effective configuration.”

They are also working to develop their robot into a modular platform, with interchangeable mechanisms. So if a large unoccupied space requires ultraviolet disinfection, the Bernoulli module could be removed and an ultraviolet payload easily swapped in, with the software and sensors adapting to each custom environment. In this way, the robot will be ready for schools, cruise ships, office buildings, factories, or any other need that may happen in the future.

“My favorite part is that we are using computational fluid dynamics in a real application,” said Quiñones. “That’s why I’m really proud of our collaboration in this project. When I see this robot, and see that our work can really help to save lives, I really get excited about it.”

This technology is protected by patents under the Purdue Research Foundation’s Office of Technology Commercialization, technology track code 2021-CAST-69215.


Writer: Jared Pike,, 765-496-0374

Source: Richard Voyles,, 765-494-3733

Luciano Castillo,, 765-494-8607


Occupant-centric robotic air filtration and planning for classrooms for safer school reopening amid respiratory pandemics
Haoguang Yang, Mythra V. Balakuntala, Jhon J. Quiñones, Upinder Kaur, Abigayle E. Moser, Ali Doosttalab, Antonio Esquivel-Puentes, Tanya Purwar, Luciano Castillo, Xin Ma, Lucy T. Zhang, Richard M. Voyles

Abstract: Coexisting with the current COVID-19 pandemic is a global reality that comes with unique challenges impacting daily interactions, business, and facility maintenance. A monumental challenge accompanied is continuous and effective disinfection of shared spaces, such as office/school buildings, elevators, classrooms, and cafeterias. Although ultraviolet light and chemical sprays are routines for indoor disinfection, they irritate humans, hence can only be used when the facility is unoccupied. Stationary air filtration systems, while being irritation-free and commonly available, fail to protect all occupants due to limitations in air circulation and diffusion. Hence, we present a novel collaborative robot (cobot) disinfection system equipped with a Bernoulli Air Filtration Module, with a design that minimizes disturbance to the surrounding airflow and maneuverability among occupants for maximum coverage. The influence of robotic air filtration on dosage at neighbors of a coughing source is analyzed with derivations from a Computational Fluid Dynamics (CFD) simulation. Based on the analysis, the novel occupant-centric online rerouting algorithm decides the path of the robot. The rerouting ensures effective air filtration that minimizes the risk of occupants under their detected layout. The proposed system was tested on a 2 × 3 seating grid (empty seats allowed) in a classroom, and the worst-case dosage for all occupants was chosen as the metric. The system reduced the worst-case dosage among all occupants by 26% and 19% compared to a stationary air filtration system with the same flow rate, and a robotic air filtration system that traverses all the seats but without occupant-centric planning of its path, respectively. Hence, we validated the effectiveness of the proposed robotic air filtration system.