Researchers from the University of Southern California (USC) Viterbi School of Engineering have created a four-legged robot that features the locomotive agility of an animal but that is also equipped with chemical disinfection to fight COVID-19 and other viruses.
Four-legged robots have been applicable for a variety of uses such as working on an oil rig, conducting search and rescue missions, pulling a truck and more. But USC said this is the first time this type of robot has been combined with a disinfection task.
Called Legged Agile Smart Efficient Robot for Disinfection (LASER-D), a team of seven USC Viterbi master’s students worked on the robot that was tested on campus.
“This is the first time we’ve combined a legged robot with the disinfection task,” said Quan Nguyen, an assistant professor of aerospace and mechanical engineering at USC. “This can be challenging because we need to maintain mobility while positioning for disinfection. LASER-D conserves energy by walking and positioning its body simultaneously. Thus, we can just use the robot orientation to control the spraying of disinfectant, instead of attaching an extra arm to perform this task.”
While the robot is not yet autonomous, the team is working on improving the autonomy and increased distance from the human operator in future versions.
While LASER-D was built with disinfection as the key element, the ability to move while spraying could be useful in other segments such as agricultural in pesticide dispensing or precision irrigation. Another potential application is general cleaning with less focus on disinfection and more on aesthetics such as at a shopping mall or office space.
Machine learning
The robot’s vision is based on machine learning but on the cheap because the team only had limited data. The robot was trained on a pair of images with one being the surface prior to disinfection and the second image of the surface post-disinfection.
The vision system helps human operators review what the robot is seeing but also allows the operator to weigh in during the process. The dual image recognition helps deal with diverse images such as with multiple types of surfaces.
“When lighting conditions aren’t good, human eyes might not be able to detect areas that were not coated by the disinfectant, however, the vision system can,” Nguyen said.
LASER-D navigates through waypoints on a map. A human operator presents the map of the overall space and then designates waypoints to help the robot route accurately and efficiently through the space.
USC hopes to enhance the vision system further to improve the robot’s collision avoidance system by adjusting its path autonomously. This could allow the robot to traverse across the entire USC campus or an office.
“Locomotion of legged robots in rough terrain is a challenging problem in and of itself,” Nguyen said. “But the problem for us right now is combining locomotion with disinfection.” For example, LASER-D would need to maintain its ability to move and spray surfaces while crouching under a table.
