Hartford Business Journal

C T I N N O V A T O R S , 2 0 2 3 1 1 decide you want to stop, how to stop — that's the realm of math and physics. at's executed by computers way better than any human ever can." Since 2010, his autonomy work has been carried out as part of the Sikorsky Innovations group. e idea behind the group's formation was to narrow down research and development efforts to a manageable number of highly relevant projects and invest heavily in them. Cherepinsky says the central question they try to answer is, "what will transform vertical flight?" e first research pillars chosen for Innovations were autonomy, speed and intelligence. Back then, his boss at Innovations was Chris Van Buiten, the founding director of the new unit. Van Buiten says he knew from the start of their work together that Cherepinsky was something special. "I feel blessed that during my life, I've worked with a lot of brilliant people, but he is by far the smartest," Van Buiten said. Van Buiten says what's different about Cherepinsky is his ability to both be an expert in the highly technical details, and also see the bigger picture. "Just talking to him, you quickly understood that he was much more than a flight-controls expert," he said. "His technical expertise was very broad, whether it was aerodynamics or structures or electronics, electrical systems — and then he also had a really savvy business side to him." Building the team at Innovations was an interesting exercise. At any one time, the unit has about 100 dedicated team members, with other Sikorsky staff being assigned temporarily. at can mean the unit fields as many as 400 experts. Van Buiten says the key skill is being self-starting. "Self-propelled people," he said. "You give them a very general direction and help them with the resources, and they'll just go and attack and make it their own, and have the courage to try and the courage to fail." Cherepinsky, he says, "has that in spades. His work ethic is just enormous." Van Buiten says Cherepinsky also intuitively understands how to get the best out of the engineers he's working with. "One of the parts of the job I always enjoyed," says Van Buiten, "super-brilliant people may not always be the easiest to work with, and the normal systems and measurements don't value them appropriately. ey prefer working at 2 in the morning (rather) than showing up at 8 a.m. You know — they're just different." Cherepinsky, he says, "knows how to leverage that." In May 2020, Cherepinsky became director of Sikorsky Innovations. a development and demonstration technology for the company's innovations in autonomous flight. It was this aircra in 2016 that flew the 30 miles from Sikorsky's Stratford headquarters to an airstrip in Plainville controlled solely by a tablet. e pilot was able to click and drag to replan what the helicopter was doing — never touching the cockpit controls. at work was the first phase of a project Cherepinsky and his team undertook in collaboration with the federal government's Defense Advanced Research Projects Agency (DARPA). e ALIAS — Aircrew Labor In-cockpit Automation System — was supported by Sikorsky's MATRIX autonomy technology. In November 2022, the DARPA project successfully demonstrated how a Black Hawk helicopter flying autonomously — with no pilots on board — could complete a cargo resupply mission and rescue operation. In the staged "mission" carried out and filmed at the U.S. Army's Yuma Proving Ground in Arizona, the unmanned aircra flew more than 80 miles carrying 400 units of blood and flying, at times, as low as 200 feet. It then resupplied troops with an external load. Mid-mission, a ground operator — again armed only with a tablet — took over control of the aircra, and re-routed it to evacuate a casualty. Cherepinsky says one of the big challenges in developing that capability was building a way to communicate with the machine. "We actually built a language and a user interface where almost anyone can convey what the mission really is in almost a human- like exchange," he said. "Based on that, the machine figures out a plan and presents a plan to you and says, 'here's what I'm going to go do.'" It's then up to the pilot to approve the plan, edit it, or tell the machine to try again. Human and machine While society is slowly getting used to the idea of self-driving cars, autonomous flight might seem a step too far to many of us. "We humans are terrible at giving up control," said Cherepinsky. "We always think we can do a better job than we actually can in a lot of cases." Instead, he is a firm believer in human and machine working together. "It's a healthy tension between, what does a human do well, versus what does a machine do well," he says. To explain it in terms most of us can understand, he takes the example of anti-lock brakes, something we all now take for granted. "at's your perfect example of where the human is the mission commander," according to Cherepinsky. "So we say, 'I want to stop.' A machine can't figure that out — that's hard. But once you CHER EPINSK Y "Anytime you try to apply process to innovation you end up stifling creativity. I hate using words like controlled chaos, but there is a little bit of that." – Igor Cherepinsky Igor Cherepinsky Director Sikorsky Innovations Education: Brooklyn Polytechnic Institute (now NYU Tandon School of Engineering) Age: 47 Continued on next page

• Cover