When you first meet Michael Goldfarb, his soft-spoken demeanor and infectious enthusiasm immediately impress you. When you ask what it is that motivates him, his answer is as unambiguous as it is unassuming. “My research and engineering passion is developing assistive technologies for disabled persons,” Goldfarb says.
“I’ve done other things in the past, but it’s that type of work where my passion truly lies. I’m very happy that during the past five years I’ve been able to fully pursue that. The work I’m doing right now is, for me, really rewarding.”
For this Vanderbilt professor of mechanical engineering, that work is focused on the development of robotic artificial legs and arms for amputees, and robotic lower-limb exoskeletons that will allow paraplegics to walk again. As director of the Center for Intelligent Mechatronics at Vanderbilt, Goldfarb heads a dedicated team of researchers, graduate students and engineers that is producing a new generation of artificial limbs that are remarkably close to the real thing in both form and function.
This near-human capability is owed in no small part to recent advancements in the related robotics technologies that make up the limbs’ electric motors, sensors, intelligence systems, power electronics and batteries.
“It’s exciting that all these things are coming together,” Goldfarb says. “If any one piece [of these technologies] were missing, we couldn’t do this.” To Goldfarb, it represents more than a coming together of technologies; it’s the meeting of a lifelong passion, persistence and opportunity.
Goldfarb did most of his growing up in Atlanta, and as far back as he can remember, he was always interested in engineering. “I started building things and taking things apart at a really young age,” he says. “I don’t know if it’s true, but there’s a family story that when I was 2, I took apart the vacuum cleaner. Whether I got it back together or not, I’m not sure, but there’s no question that I liked taking things apart and putting them back together again.”
Building and designing things have been common threads throughout his life. The genesis of his interest in prosthetics is more of a mystery. When he was 7 or 8 years old, Goldfarb remembers seeing some magazine articles and TV programs about disabled persons, and somewhere along the way the two notions combined—trying to use engineering to improve their quality of life. “I guess to some extent, I’ve always gotten the message from my parents that trying to make society a better place is an important thing,” he says.
His family always supported his interests, sometimes even covertly. “My grandmother was a full-time volunteer at the Atlanta Veterans Administration Hospital for more than 40 years,” Goldfarb says. “She ended up getting me a summer job when I was in high school, working in the orthotics and prosthetics lab at the VA.” Seeing the benefit patients received from artificial arms and legs was a tremendous experience that left an indelible impression. “I didn’t find out until 20 years later that it wasn’t a real job; my grandmother had arranged the whole thing. She even gave the money to my boss for my paycheck,” Goldfarb says.
Although his parents had divorced, Goldfarb’s father, a heart surgeon living in Phoenix, had followed his academic progress and arranged for him to skip his senior year of high school and enroll at the University of Arizona on a full scholarship. Goldfarb majored in mechanical engineering, eventually working on a functional artificial arm for his senior design project. He went on to do his graduate work at MIT, designing a robotic leg for above-the-knee amputees. His doctoral work was a computer-controlled exoskeleton that allowed paraplegics to walk in the lab.
“I was fortunate that from a young age, I knew what I wanted to do, and I’ve been fortunate enough to find opportunities to do it,” Goldfarb says.
Since joining the faculty at Vanderbilt in 1994, Goldfarb not only has thrived on the research aspects of his career, but on the teaching aspects as well, particularly in the unique environment of his lab. “Working with students in the research setting enables a deeper interaction than one can typically experience in the classroom, and importantly it often entails two-way interaction,” he says. “They learn from me, I learn from them, and we often learn together.”
Goldfarb initially came to Vanderbilt to partner with then-Vanderbilt professor Ephrahim Garcia. “We immediately connected technically and personally,” Garcia says, “and together we established the Center for Intelligent Mechatronics at Vanderbilt.” Although robotics wasn’t Goldfarb’s first love, the government was, at the time, providing substantial funding for robotics research. The pair proposed and won a couple of large grants from DARPA, the research and development agency for the U.S. Department of Defense, to design, construct and analyze robotic insects capable of covering long distances using only onboard power.
“There was a connection between the way we engineered those devices and our observation of natural systems that captures one’s imagination,” says Garcia, now an associate professor of mechanical and aerospace engineering at Cornell University’s College of Engineering. “I think Michael and I have continued to look to natural systems for design inspiration.
“Michael is brilliant and loves the details,” Garcia adds. “He amalgamates these details to arrive at a design that he may not have spoken much about, but was clear to him from the beginning.”
One such design is a robotic artificial leg that is not only vastly superior to the current state of the art in lower-limb prosthetics, but is now in the early stages of transitioning from the lab to the commercial marketplace. “I’m thrilled that some of this technology I’ve been working on for the last 20 to 25 years is beginning to make it out of the lab where someone will actually be able to use it to improve their life,” Goldfarb says.
“In effect, the leg learns the person, rather than the person learning to use the leg, which is quite revolutionary.”
At present, the most advanced artificial legs that lower-limb amputees can buy basically have dampers for knees (similar to the device that makes a screen door close slowly), and a carbon fiber leaf spring.
“It’s not really good at adapting from one activity or surface to another,” Goldfarb says. “The real problems come when amputees try to do things other than walk on level ground, like walking uphill or downhill, upstairs or downstairs. Those legs are just not adaptable enough or capable enough to provide normal kinds of motions. They’re very unstable, and amputees end up falling a lot.”
Goldfarb’s lab has developed a fully robotic battery-powered leg that can adapt to whatever the person is doing. If the user is walking downhill or downstairs, it recognizes that and provides the right kind of biomechanics. “Our leg basically recognizes what it’s doing and makes the proper adjustments. We’ve been really pleased with the performance so far.”
The leg utilizes many of the same technologies that have made hand-held devices such as smartphones so capable and so successful. “All the technologies that have enabled smartphones are the same technologies that are enabling us to do the things we can do with these artificial limbs and exoskeletons,” Goldfarb says. The same microsensors and accelerometers that allow a smartphone to know when the screen is tilted and adjust the display are utilized in the prosthesis, giving the robotic leg the ability to know not only its orientation in space, but the speed, acceleration and elevation of every part of the leg as well.
Even more astounding is how the device is controlled. There are no surface electrodes or implants transmitting the nerve impulses of the user to activate the power electronics of the leg. Instead, all the information used to control the leg is inferred by the motion of the user’s hip joint and the manner in which it swings the leg around.
“It has to make these decisions very quickly and very reliably,” Goldfarb says. “It has a lot of sensors and a lot of software, and it uses these, plus some very modern pattern-recognition software, to look at patterns from the sensor information and decide what the person wants to do. It’s similar to pattern-recognition software for voice-recognition programs. So, in effect, the leg learns the person, rather than the person learning to use the leg, which is quite revolutionary.”
This technology was recently licensed to the largest U.S. manufacturer of lower-limb prosthetics, and is currently being developed to enter the commercial market. This real-world application is also why Brian Lawson, a second-year doctoral student, was initially attracted to Goldfarb’s lab. “I saw the potential and the motivation in his research to reach beyond academia and contribute directly to the well-being of the amputee,” Lawson says. “What I have found subsequently is a remarkable juxtaposition between the impact of his work and his unassuming demeanor. I believe Dr. Goldfarb’s attitude should be a reminder to researchers that when you do good work, it will speak for itself.”
That’s a sentiment echoed by longtime colleague Ephrahim Garcia. “Michael sees a world where he challenges the status quo and accepted wisdom,” Garcia says. “He sweats the details, and what emerges is seminal work that can change the world.”
“My greatest hope is that people can buy [the robotic leg] on the commercial market and for it to make a real difference in their lives,” says Goldfarb. “That would be immensely rewarding to me—and as much as I could hope for.”