SFU Researcher Explores Robot Empathy Through Humour and Signals

A researcher at Simon Fraser University is investigating how robots can interpret human emotions, specifically through humour. Angelica Lim, an associate professor of computing science, leads a project aimed at teaching robots to recognize and respond to non-verbal cues such as facial expressions. This research could have significant implications for the development of robots that interact more naturally with humans.

During a demonstration, Lim introduced a robot named Pepper, which is equipped with algorithms that enable it to respond when prompted with, “Tell me a joke.” When it began its routine, Pepper’s mechanical voice delivered a light-hearted quip about bananas. However, it also assessed Lim’s neutral expression for feedback on its performance, showcasing its ability to gauge human reactions.

Lim’s work at the Robots with Social Intelligence and Empathy (ROSIE) lab is centered on enhancing robots’ social capabilities. The goal is to develop machines that can engage with people in meaningful ways, moving beyond the limitations of traditional AI assistants like Siri or Alexa.

Understanding Emotional Cues

This research delves into a critical question: can robots ever truly understand emotions? Lim’s initial fascination with robotics was sparked by her childhood experiences at Disneyland, where animatronic characters brought stories to life. Today, her focus has shifted to the ethical implications of creating machines that could mimic emotional responses. She argues that while robots may be designed to replicate empathy, doing so carries significant risks, particularly concerning mental health.

Lim draws attention to the potential dangers of overly empathetic AI, which can lead to unintended psychological effects on users. Instead, her research emphasizes teaching robots to interpret human body language and emotional signals appropriately. This approach could improve practical applications, such as robots assisting in hospitals or retail environments.

For instance, a robot delivering supplies in a hospital would need to recognize social dynamics, such as not interrupting conversations. Lim states, “Humans automatically understand these social cues, and we want robots to have that capability as well.” Developing unbiased algorithms that respect privacy while accurately interpreting social signals is a key challenge for her team.

Real-World Applications

The significance of this research extends beyond academic interest. AJung Moon, a computer scientist at McGill University, leads a lab focused on the ethical implications of autonomous systems. Her team is currently working on a robotic shopping assistant designed to assist customers with special needs.

In a hypothetical scenario, a visually impaired shopper with food allergies might rely on the robot to navigate a store. Moon envisions a robot that can pick up on subtle cues indicating the shopper’s urgency, enabling it to provide efficient assistance without necessitating constant verbal communication.

Both Lim and Moon agree that while robots can be designed to interpret human emotions, ethical boundaries must be established. The extent to which robots should simulate human emotions should correlate directly to their intended use.

At Simon Fraser University, Lim oversees a fleet of ten Pepper robots, originally donated by a now-defunct French company. These robots have been employed in various educational initiatives, including teaching coding to high school students and facilitating language learning for Indigenous studies students. Their versatile applications highlight the potential for robots to serve as valuable educational tools.

Lim remains committed to her vision of creating robots that can evoke joy and wonder. Inspired by recent advancements in robotics, such as Disney’s lifelike animatronics, she continues to explore how technology can enhance human experiences.

As research progresses, the ability for robots to understand and respond to human emotions could redefine interactions between humans and machines, ushering in a new era of empathetic robotics.