Prosthetics with a human touch

For those who have lost arms, prosthetics that mimic the warmth and softness of a human hand can improve the quality of personal relationships.

JJCRecent years have seen great advances in the functionality of prosthetics, but the loss of an arm or hand affect more than just motion and dexterity. What happens when your child can’t feel the reassuring warmth of your hand on his back? Or your partner the softness of your touch? John-John Cabibihan, Associate Professor of Mechanical Engineering at Qatar University, is working to address this issue, and has demonstrated that it’s possible to create prosthetic hands that feel real to the touch. 

ResearchGate: Can you tell me what role touch plays in human relationships?

John-John Cabibihan: You can get a good idea of this if you think about the scenes that unfold around you at an airport. What do you see when people say goodbye or greet each other? There’s almost always some kind of intimate contact, depending on the closeness of the relationship. A hug, a handshake, a pat on the back. Touch is hardwired into human relationships. Hospitals recommend that a newborn child immediately be given to the mother to hold. Touch is important for creating that bond. From birth until death, we see touch of some kind at play in all human relationships.

RG: And what does that mean for people who have lost limbs and are therefore limited in their ability to touch others?

Cabibihan: It’s difficult for people who had touch hardwired into them, and because of war or an accident, suddenly lose the ability to touch as they did before. You can imagine the difficulty of not being able to reach out and connect with another person. Of course, losing limbs poses great practical challenges as well. You can’t open the door by yourself, it’s difficult to eat without dexterous hands and fingers. But the effect on human relationships can be even more devastating. We become isolated if we cannot touch one another, if our children can’t feel warmth when we hug them or our spouse the softness of our caress. The result can be feelings of anguish, distress and depression. Ultimately, the person who lost an arm or a finger may be overwhelmed by a sense of loss and even experience suicidal thoughts.

RG: How have prosthetic arms evolved over time to address the challenges people who have lost arms or hands face?

Cabibihan: The first priority in the development of prosthetics over the years was to bring back functional elements, and we’ve seen remarkable progress in the past 20 years in the user’s ability to control prosthetic arms and fingers. With some technologies, a prosthetic user can contract his muscles, and the fingers will grasp or close. Researchers have invested a lot of time in functionality, and we can already predict that in 5-10 years, mechanisms and controls will be fully developed. So we asked ourselves, what’s next? And this is what we arrived at: bringing back touch to those who’ve lost it. It’s our vision to fully return human contact and interaction, the ability to tickle a child or hold a loved one and have that touch feel authentic to the recipient.

RG: And how does that technology work? What are the methods that you’re using to replicate the feel of human touch?

Cabibihan: First we had to mimic the structure of a hand. I went to the hospital and did a CT scan of my own hands. Other robotic hands primarily use cylindrical finger design, which can easily be manufactured, but the human hand has lots of contours and lines. There’s also the skeletal structure, and the CT allowed us to accurately replicate that and all the many layers of the human hand.

Once we had the right structure, the next step was determining the right softness and warmth. We reviewed various materials to replicate the softness of the human hand, testing them with a robotic arm. To achieve the ideal softness, we asked subjects to feel material samples of different temperatures and identify the one that felt most lifelike. Their answers revealed an ideal skin temperature for a robotic hand.

RG: And how realistic was the resulting hand?

Cabibihan: Once we combined the soft material, the structure, and the warm skin, we did a final test. It was a modified Turing test, where the participants were touched with hands they couldn’t see: a human hand, a cold artificial hand, and a warm artificial hand heated to the ideal temperature with a heating system. We asked them which was the human hand and which was the robotic hand. Their answers showed that the warm, soft artificial hand is indistinguishable from a human hand. This was a really good result! We were able to show that if you combine the structure of a human hand with soft materials and warmth, it’s possible to mimic the effect of human touch from the point of view of the recipient.

RG: Is there anything that you think is particularly unique about human touch that can’t be mimicked?

Cabibihan: Well, there’s sweat. Our sweat rates vary depending on our moods and stress levels. That’s something we haven’t incorporated into prosthetics yet, but it would be interesting to investigate the possibility of introducing small pores. We’d have to see if that aspect could be replicated.

RG: In addition to prosthetics, you also work in social robotics. What are social robotics, and what are their applications?

Cabibihan: Simply stated, social robotics is the study of the effect interactive robots have on humans. I’ve done some work mapping out the trends in this field. There are a variety of applications. Social robots are helpful for the elderly and also children, especially children with autism spectrum disorders.  The latter is actually a project my team is working on parallel to our prosthesis work.

RG: Are there applications for the work that you’ve been doing simulating human touch that could also play into social robotics? 

Cabibihan: Yes, particularly for humanoid robots.  We have social rules and norms in our day-to-day interactions. We shake hands at the office, high-five our teammates when we play sports, pat each other on the back in celebration. As robots move into human spaces, these touch-related norms will have to translate. Humans expect it. Robot-human handshakes are already common, and other touch-centric interactions will be needed in the future as the field progresses. We’ve done studies on the biomechanics of human handshakes and how to replicate them in robots. It’s really not easy! In a human to human handshake, there are normally 2-3 pumps. After that you subconsciously signal your partner (and your partner signals to you) that that’s enough using the little finger and thumb. Just a quick, subtle signal that it’s time to release and stop shaking hands. It’s this kind of small, nuanced behavior surrounding touch we need to pass on to robots. Otherwise, the robot will be extremely awkward!

RG: And what are you working on now?

Cabibihan: Right now we’re very busy with the project I mentioned earlier: using robots to help children with autism. Children on the autism spectrum have difficulties with communication. Social robots can, for example, help teach them to say hello in a manner appropriate for the social context. They can break down a wave into parts: raise your hand, move the wrist left to right, etc. Or they can teach children to hug when it’s appropriate, to understand that a hug is coming and isn’t an attack. These are important and subtle human gestures that children with autism can learn using robots as a tool.

RG: And why use robots? Why not teach children these behaviors with human instructors?  

Cabibihan: In previous studies we found that children on the autism spectrum interact better with robots, computers, tablets, technology in general than with people. This is because humans have complicated facial movements. If you agree with somebody, you nod your head. If you have doubts, your eyes twitch and you knit your brow. These types of signals are very complex for children with ASD, and social robots don’t exhibit them. Their movements are simpler and more straight forward, so they are easier for children with autism, who are very sensitive to expressions and gestures, to interpret.

Featured image courtesy of iwishmynamewasmarsha.