Human Robot Interaction

January 23, 2018 - 2 minute read -
robotics ai psychology

This semester, I’m taking a course on Human-Robot Interaction (HRI) with Dr. Andrew Williams. HRI is a relatively young multidisciplinary field that draws from various aspects of cognitive science, robotics, psychology, and more. The ultimate goal, regardless, is to understand how robots can interact with humans in natural, intuitive, and meaningful ways.

What does that even mean?

…this isn’t ideal but it’s pretty close.

The earliest human-robot interaction I remember vividly was with the iDog, a robot dog that reacted to music from my iPod. It would flash different colored combinations of LED lights on its face and would dance to the beat of the music by wiggling its ears and shaking its head. I thought it was THE coolest thing.

That was early 2005-2006.

Now, just 13 years later, we have robots like Sophia. She is pretty close to being human-like. Sophia is what the robotics community call a “humanoid”. She can follow faces, maintain eye contact, recognize people, process speech, and have decent conversations. She can even dance!

It won’t be too long until we have fully functional social robots - embodied agents that are able to recognize each other and engage in meaningful social interaction - living among us. Can you imagine a Bernard always by your side? (If you seen haven’t Westworld, I highly recommend it).


Still, what does being socially functional even mean?

As we go through the course, we will learn more about what constitutes a meaningful human-robot interaction. This includes artificial intelligence, the psychology of social interactions, and a number of different computational models for social intelligence. By the end of the course, we should be able to design and program our own socially intelligent robot.

To get us even more excited about the class, Dr. Williams showed us a clip from the movie I, Robot starring Will Smith.

Sonny is quite a robot - a sentient humanoid with advanced social intelligence. He was able to learn and respond to his surroundings. He was able to pick up on cues like the photographs shoved in front of him which meant “Hey, look at these”. He was able to express intense emotions like anger, guilt, and fear. Sonny demonstrated sarcasm and even love, a concept that still heavily studied by humans.

Sonny is an embodiment of human growth and evolution. Sonny went from a simple machine, comparable to a human infant who knew nothing of the world its in, to a complex being who experienced self-actualization. By the end of the movie, Sonny revealed an understanding of judgment, morality, and ambition. Characteristics humans do not fully understand ourselves.

How, then, do we teach a robot this? How can we translate human emotion and understanding into mathematical models?