The Importance of Teaching Robots to Hug in IEEE Spectrum

June 5th, 2018

By Evan Ackerman

Hugs make us feel warm and safe and comforted and loved. They’re pretty great, if you’re into that sort of thing. If we need a hug and another human isn’t available, we can sometimes get a little bit of satisfaction from hugging inanimate objects like stuffed animals, but it seems like robots (that can hypothetically hug us back) might be able to be somewhat more fulfilling. While we’ve seen robots that are actively huggable before, and even a few that can hug you back, it’s not clear exactly how a robot hug compares to a human hug, and whether hugging a robot can confer any of the benefits that we get from hugging people.

At the ACM/IEEE International Conference on Human Robot Interaction (HRI) earlier this year, Alexis E. Block and Katherine J. Kuchenbecker from the Haptic Intelligence Department at the Max Planck Institute for Intelligent Systems in Stuttgart, Germany, presented a paper on “Emotionally Supporting Humans Through Robot Hugs.” Their work explores how robots can be more effectively designed and taught to give the kinds of hugs that humans will love. If you hug robots every time you see them (like I do) and sometimes wish those robots could be just a little bit warmer and softer, this research is definitely for you.

The importance of hugs goes way beyond “oh that feels nice thank you.” There’s plenty of research showing that hugs offer tangible health benefits, including lowering blood pressure and  increasing oxytocin levels, as well as significant improvements to mental health, like social support reinforcement and stress relief. And if that doesn’t convince you that physical affection is a really important thing, these horrific experiments with giving baby monkeys surrogate mothers made of either wire or cloth definitely will.

For many (if not most) people, hugs are a natural way of expressing affection, and giving a hug is something that comes almost instinctively. For the rest of us (myself included), hugs can be a little bit awkward at times, because they’re a deceptively simple interaction—a good hug involves proper positioning of the arms and hands and torso, along with awareness of the appropriate amount of force to use and knowing when the right moment is to let go. As humans, we get a lot of body language cues from the huggee, and as the hugger, we give our own cues too. For robots, all of this stuff needs to be made explicit, studied, and quantified, at least to the point where a robot knows how to safely hug someone without accidentally crushifying them.

Of course, simply refraining from squishing a human into a pulp is not all that a good hug requires. But for a robot, it’s not yet clear what the requirements are for a hug that will have a net positive effect on a human in need of support. The research presented at HRI is an attempt to figure some of this out, by conducting a study where participants hugged a specially modified PR2 named HuggieBot. HuggieBot could be configured with extra-soft layers of foam, cotton, and polyester, and those layers could also be heated to make them pleasingly warm. It was also programmed with a range of hug pressures (from loose to tight), and interactive hug durations (controlled by a pressure sensor behind its neck) ranging from very brief to five seconds after the participant attempts to disengage, which was long enough to sometimes result in low-key panic on the part of the human.

Read the Full Article, See the Video and Interview in IEEE Spectrum