Learning to love robots: Making machines that people can welcome

May 6th 2018 | Bilge Mutlu, Special to the State Journal
Faculty, Natural & Physical Sciences
Back to News
Bilge Mutlu Fueling Discovery Slideshow1 645X415 A social robot, Wakamaru, provides its user with spoken instructions to assemble a pipe structure, following user progress by keeping track of the parts that are being assembled. It offers corrections and clarifications if there are errors in the user’s assembly. The research question here was to create an autonomous robot system that provides cognitive assistance by instructing humans and “repairing” misunderstandings. (Bilge Mutlu)
Bilge Mutlu Fueling Discovery Slideshow2 645X415 A robotic arm teams up with its user in performing a task and adapting its performance to the availability and task progress of its user. The robot tracks its user and plans when to hand the plates over according to when it estimates its user to be ready to receive the plates. The research question is whether the robot can improve user experience when it plans itself according to user availability rather than doing its task as fast as possible and waiting for its user. (Chien-Ming Huang)

We hear a lot about robots in the news, see them in movies and read about them in books. They are coming, we’re told, but when? What will they do? Will they walk among us? Will they take away our jobs?

Our relationship with robots is defined by ambivalence and uncertainty.

We are fascinated by the possibilities but are skeptical of these robots. They give us hope, but we fear their side effects. Although technology — the telephone, the internet — is often disruptive, and produces similar hopes and fears, there seems to be something special about robots.

My research aims to build human-centered methods and principles for the design of robotic technologies and for their successful integration into human environments.

I teach human-computer interaction and co-direct the Mad UX User Experience Design program. My students and I work to guide the design and development of robots with an understanding of the needs, expectations, and concerns of their future users.

We use what we know — and often learn what we don’t know — about human psychology to design the algorithms that control robots so that they interact with their users in natural, intuitive and non-threatening ways. We carefully study human environments and think about how robots can best be integrated into these environments so that people will welcome them.

There is no denying that bringing radically new technology into our environments is disruptive.

In the last nine years, we have developed educational robots that can understand whether students are distracted and regain their attention, robotic assistants that know when their users are busy and offer help exactly at the right time, flying drones that can communicate their intent to bystanders and companion robots for children that turn reading into a sustained social activity.

In all of our work, we repeatedly see that careful design starts with an understanding of human needs, expectations and concerns. This care creates robots that work better, that people welcome, and that get closer to what we hope robots will be.

There is no denying that bringing radically new technology into our environments is disruptive.

People working with robots every day in hospitals and in factories — and our own observations — tell us that people want robots that respect them, that work in familiar ways and that blend in.

These expectations are all the more reason to study humans’ environments, to understand how people work, and to carefully design new technologies and plan robots’ introduction — so that the robots will blend in and people will welcome them, not reject them.


Bilge Mutlu is an associate professor in the Department of Computer Sciences and the director of the Wisconsin Human-Computer Interaction Laboratory. His research program develops human-centered principles for designing robotic technologies that help people communicate, work and pursue personal goals. His research group combines design thinking with knowledge and methods from computer, social and cognitive sciences.

Fueling Discovery

"Fueling Discovery" is a joint effort of the UW-Madison College of Letters & Science and the Wisconsin State Journal featuring faculty members wiring about their work in their own words. The effort was financed through sponsorships and gifts from alumni and friends.