Meet the researcher building robots to help disabled people travel safer

25 Apr 2018

Prof Aaron Steinfeld of Carnegie Mellon University’s Robotics Institute. Image: CMU

Prof Aaron Steinfeld of Carnegie Mellon University’s Robotics Institute is working on building robots that could one day help disabled people navigate major transport hubs.

Significant strides have been made to ensure our society is more inclusive of people with disabilities in terms of building accessibility and design.

But, increasingly, we are seeing technology being developed to bridge this gap even further, which is what Prof Aaron Steinfeld of the Robotics Institute in Carnegie Mellon University (CMU) is trying to achieve.

Having received his BSE, MSE and PhD degrees in industrial and operations engineering from the University of Michigan, he moved to California PATH at the University of California, Berkeley.

Here, he continued to work on advanced vehicle systems, including a snowplow that could be driven through white-outs, collision warning systems and autonomous vehicles.

In 2001, he joined CMU and extended his transportation research to include human-robot interaction and human interaction with artificial intelligence (AI).

What inspired you to become a researcher?

I have always loved technology. I would take apart my toys, build things, and try and figure out how machines worked. Also, my free time as a child was spent on Lego, electronics kits and model aircraft.

My grandfathers were an electrician and a metallurgist, which definitely made an impression.

We had a computer in the house well before I reached high school, so we were ahead of the curve for my generation.

I also grew up around research. My father, Edward Steinfeld, has a long and distinguished research career in making environments and products more inclusive to all members of society. My earliest memory of going to work with him was riding around in a wheelchair while he examined a building. While I don’t have a disability, I grew up with the idea that buildings, technology and transport should be accessible and useful to everyone.

Can you tell us about the research you’re currently working on?

My research focuses on human-robot interaction and advanced transportation.

One example of this is helping people with disabilities efficiently use transportation hubs and other complex buildings. Our research group is trying to develop guide and information desk robots that live in the building and provide tailored assistance based on each user’s needs. The goal is not to replace station agents, but to instead provide assistance where human employees are unable to help due to demand or location.

Another project in our lab is trying to make transit vehicle information on smartphones more useful for people with disabilities. It is not a big leap to imagine a future where a bus rider with a disability uses an app to reach a train station and is then met at the door by a robot that guides them to the correct platform.

We are also looking at issues related to autonomous vehicles. Many people with disabilities and older adults are unable to drive a personal vehicle. Fully autonomous vehicles could dramatically increase the quality of life for this community since large parts of the country are not well served by public transit.

In your opinion, why is your research important?

There are two large societal problems I hope our research helps address.

First, people with disabilities need more accessible transportation. Lack of efficient and usable transportation severely limits access to work, healthcare and independence in society.

Second, everyone knows that robotics and AI are likely to have a significant impact on society. However, we are still learning how these systems should interact with humans. Driving straight through is rude, and waiting for a large gap may make the robot late. Instead, it needs to drive in a socially appropriate way that is efficient yet inoffensive. This is a hard robotics challenge that humans do without thinking.

What commercial applications do you foresee for your research?

We’ve heard from several companies that socially appropriate navigation by robots is an important capability for future commercial robots.

There is also significant demand for robots and AI that help older adults and people with disabilities be more independent. Many developed countries are seeing their population age, with correspondingly large gaps in available caregiver support. Since human help is expected to be sparse, many hope AI and robots can fill this gap.

What are some of the biggest challenges you face as a researcher in your field?

My colleagues and I recently started a new project on how people might interact with AIs that migrate between devices in support of their user.

People already interact with social media this way in many companies, but what happens when you add a work AI to this mix? Do you want your work AI taking over your home robot? Are you comfortable with data used by your personal AI being stored on work devices? Are you OK with your friend’s AI taking over your television when they come to visit?

These questions are examples of how technological advances interact with human concepts of ethics and appropriateness. I think society is still struggling with these kinds of challenges, and it will require work from a wide range of disciplines to find solutions.

Are there any common misconceptions about this area of research?

I get a lot of questions about the impact of robots and AI on human jobs and personal safety.

For the first, there is a lot of concern that automation will lead to significant job loss and unemployment. I think the day where automation can do everything is far off but there will definitely be a transformative change in some industries. I think we’ll see humans shifting from doing tasks manually to supervisory and instructional roles since flexible, competent automation is really hard. Replacing humans was possible in large-scale manufacturing since the tasks were very repetitive and predictable.

Secondly, I hear a lot of ‘when robots attack’ jokes. I also get asked a lot whether autonomous vehicles will run over people and animals. Some robots are very dangerous and there are real safety concerns for most kinds of autonomy, but I also think society puts too much weight on the negative views of the future seen in entertainment.

Are you a PhD researcher? Can you explain your work in three minutes of engaging chat? Then you could be our next Researchfest champion. Find out how to apply here.