Students modify robot for remote disinfection

When students at the Center for Advanced Manufacturing, an interdisciplinary center at the Viterbi School of Engineering, realized that healthcare workers are more likely to become infected with the coronavirus, they began modifying their robot, ADAMMS, into an ultraviolet light disinfection robot: ADAMMS-UV. In just six months, the robot will be able to aid overwhelmed hospitals by allowing healthcare workers to disinfect rooms safely from a remote location. 

The original robot is used in manufacturing, and  extracts 3D printed parts late at night when factory employees are not present to collect parts for printers to begin producing the next batch. The new robot builds on those functions and integrates them with UVC light sources — ultraviolet light with wavelengths between 200 and 280 nanometers — to break apart the DNA of viruses and bacteria, sterilizing the area the light reaches. 

“We realized that a lot of the functions that we have in [ADAMMS] can be easily transferred for its use in disinfection … In the previous application we had to open doors, open drawers and pick up objects,” said Shantanu Thakar, a doctorate student studying mechanical engineering and the robot development manager. “In this case, we have to do the same things except we needed to add a disinfection motion where we move the [robot] arm to disinfect.”

The robot consists of a UV pillar, a large, vertical ultraviolet light on top of the body of the robot that shines in all directions and disinfects the surrounding air. There’s also a UV wand that the robot arm holds to disinfect hard to reach areas. Once the robot approaches an item or area to disinfect, the user can select boundary points and the robot arm will calculate the motion to perform following sterilization standards. 

The ADAMMS-UV arm and hand can move in myriad directions and interact with objects such as drawers and doors, an aspect the CAM team realized made its robot special among current disinfection robots.

“Having our robot have an arm can make it more versatile and do things which existing technologies cannot — that’s some motivation that all of us had to form a team immediately and show everyone that this concept can be really useful in times like these,” Thakar said.  

While most of the team is working from home building different parts of the software, Hyojeong Kim, Pradeep Rajendran and Sarah Al-Hussaini are the only three students authorized to enter the CAM space to work physically with the robot. They integrate new software with the robot, test it, then provide feedback to the rest of the team. For Kim, who visited the lab nearly every day for two weeks in April, working in the lab by herself proved challenging. 

“It was challenging to do because I’d actually move the robot, and usually some person should watch the movement of the robot and stop it if something happens, but I was the only one who could work in the lab space,” said Kim, a graduate student studying computer science with a specialization in intelligent robotics.

The ADAMMS-UV software can currently only be used on desktop computers, but the team is working on making it available on smartphones, tablets and laptops to make it more accessible for healthcare workers. 

“It definitely works well in our lab, but that’s not the main target,” said Rishi Malhan, a doctorate student studying mechanical engineering who works on the ADAMMS-UV research team. “It needs more testing to make sure it’s more robust, and it needs a user interface to make it more intuitive and user-friendly for whoever is operating it.”

With limited funding and under present circumstances with the coronavirus, the team estimates it will take at least six months for the current version of the robot to be used in a real setting. With more funding, however, the team is considering including more features and physical capabilities.  

Rajendran and Al-Hussaini are interested in adding another arm since a two-armed robot would likely be faster and more intuitive for a human to control through telepresence. The robot may also be customized to use a liquid disinfectant spray and a scrubber to help hospital personnel decrease their exposure to the virus. 

“For me, as a robotics researcher, I always feel like the best use case of robots is in the application areas where it’s risky for humans,” said Al-Hussaini, a doctorate student studying computer science. “When [the coronavirus] came up, that’s the first thing I thought about … How can we make the workspace safer while not endangering other people?” 

The CAM team continues to test its robot with the hope that it can be a tool to alleviate other coronavirus-induced problems aside from those healthcare workers are facing. 

“Not only does it save and protect humans, but it also gives certain people jobs back. That’s the vision for this,” Thakar said.