Anthony Whitehead, an associate professor and director of Carleton’s School of Information Technology, is in the process of taking the robotic world by storm.
Along with graduate students Colin Killby and Ashleigh Fratesi, Whitehead is building a human-like robot that can read emotions.
The humanoid robot is created with open-source parts that were 3D-printed and has the ability to pursue different aspects of human-robotic interaction, Whitehead said.
The goal is for the robot to recognize different emotions based on its computer vision and machine-learning algorithms.
Whitehead said the robot has two high-definition webcams in the eye sockets, which allow the robot to look at images and isolate the faces it can find. It then uses its software to recognize lines and patterns in a person’s face.
Afterwards, it examines the faces for general characteristics, such as a smile, and classifies the emotion based on the attributes it can identify.
The robot is a mechanism that works through a series of parts that are moved by servos, a powered mechanism producing motion or forces at a higher level of energy than the input level, Whitehead explained.
The servos move the body parts and multiple servos engage together to create motion of the robot as a whole. However, as a work in progress, there are several things yet to be discovered, such as the mechanism and use of legs.
According to Killby, one of his greatest challenges has been to be able to take a step back and understand how people who are unfamiliar with this type of technology will react and interact with it.
“I think the field of human-robotic interaction is changing at an incredibly fast rate, especially now,” Killby said. “It is definitely more in the forefront of our culture now than it has been in the past.”
The goal is to conclude the examination of basic human/robot interaction, but Whitehead said he suspects he can spend the rest of his career on the project.
“There is so much to do. The end date will be more a function of the number of students working on it,” he said. “As we are able to make the robot more sophisticated, we will undoubtedly uncover many more issues and problems that need to be solved.”
