Born: *Unknown
Primarily active in: United States of America

Robert Ernst, Technical UAS Chief Engineer, Naval Air Systems Command

Within the Naval Air Systems Command (NAVAIR) Multi-Mission Tactical Unmanned Aerial Systems (UAS) program office (PMA-266), Robert Ernst manages a growing technology portfolio that includes the MQ-8 Fire Scout helicopters, fixed-wing MQ-9 Reaper and still-to-be-defined MUX – the Marine Air-Ground Task Force (MAGTF) Unmanned Aircraft System (UAS) Expeditionary — system. His team mixes engineering disciplines.

“We have a lot of software people, mostly on the contractor side. It’s a lot of systems engineers. I have cyber security. I have reliability and maintainability engineers. I have manufacturing engineers. Then I have all the technical area experts required to do the flight clearance to certify airworthiness.” In addition to major NAVAIR acquisitions, PMA-266 still manages the CQ-24 unmanned K-MAX cargo helicopter under a cooperative research and development agreement with Kaman, and it orchestrates counter-UAS efforts. Ernst acknowledges, “I have enough really good deputies who are empowered. My job is seeing where the data is and using lots of years of experience to help them out.”

Years of Fire Scout experience has advanced naval UAS development processes. Ernst explained, “We’re separating flight-critical and mission-critical functions in our architecture. If we’re going to change a payload, I don’t want to have to go back and do all the flight control regression testing.” He continued, “We are using model-based engineering. The systems are so complex now with so many perturbations, we can’t do it like I did 20 years ago, let alone like my dad did 50 years ago. We’re trying to work with industry to see how we can model the system in code first, simulate it in labs, and fly it on surrogates to make sure it’s done right.”

Test facilities and labs like the one being built for MUX now come early in UAS development, and NAVAIR engineers plan live virtual constructive testing and verification at the very start of systems development. “I’m challenging my engineers, so we understand how we’re going to test it from Day One,” explained Ernst. “We can’t design a feature in six months and take 18 months to test it.”

With MUX to be defined, Ernst also urges his engineers to work with “agile” requirements: “Work them; generate them; try them; fix them; model them; correct them,” he said. The objective is early risk reduction. “We visited 14 different companies that potentially have an interest in MUX and we asked them where their risk areas were, and we’ve put in budget requests to burn those down.” NAVAIR is also considering far-reaching aspects of UAS cyber security. “We’re really leading some of the efforts on cyber as the latest emerging issue we’re trying to attack,” said Ernst. “To deal with an unknown, that’s kind of fun for me.”

Engineer’s Eyes
Robert Ernst grew up in Beltsville, Maryland, just four miles from the University of Maryland with its celebrated aerospace engineering department. His father, George Ernst, was NAVAIR’s chief engineer for the AWG-9 radar and AIM-54 Phoenix missile system on the F-14 Tomcat fighter. “I grew up with the development of the F-14,” recalled Bob Ernst. “I wanted [to go into] aerospace engineering from the time I was in seventh grade. My dad said, ‘You ought to go into electrical engineering,’ but I was fascinated by aircraft and their performance. When I heard a loud jet engine, I’d get goosebumps. Most people covered their ears.”

Passion and practicality led the would-be engineer to UMD. “I chose Maryland because they had a good program, and because it was cheap, and it was close,” said Ernst. “[Professor] John Anderson was there teaching aero, and we had a small department and a really hard-working group of people who went on and did well in a lot of things. Anderson was the main faculty member, and they just drove us. The other one was [Dr.] Dick Hallion who was big into the history of aerospace. I love history and how aircraft have evolved.”

On graduation, the new aerospace engineer went to work for the Navy. “I really wanted to go out into industry and do something practical,” recalled Ernst. “We were just beginning the Reagan buildup in 1982, and manufacturers were hiring a lot of engineers, but they wanted three to five years’ experience.” A college friend who interned at NAVAIR told Ernst of broad opportunities. “NAVAIR put me in the aircraft performance branch where I was looking at flight test data, deriving aircraft performance characteristics, and doing a lot of hands-on engineering, so I liked that.”

A follow-on assignment had Ernst analyze alternative fighter engines. “I literally sat there late at night running the programs, wrote the results on the back of an envelope, went and took my graduate course final, and presented the results to the F-14 community on the different engines for the F-14D.” NAVAIR’s focus on aging aircraft subsequently made Ernst chief engineer on the S-3 Viking antisubmarine warfare jet but kept him away from helicopters. “I was supposed to look at performance including rotorcraft performance,” he noted. “I always got shunted to the fixed-wing side. I didn’t start working rotorcraft until after aging aircraft started winding down.”

Unmanned Ups and Downs
NAVAIR leaders sought experienced engineers to develop unmanned systems and tapped Bob Ernst to work on the MQ-8B vertical takeoff and landing UAS (VTUAS). “They said, ‘We want you to be the chief engineer for Fire Scout.’ My wife said, ‘You told me helicopters weren’t real air vehicles,’ and I said, ‘Well, I guess I’m going to learn about them.’”

The VTUAS effort in 2008 faced problems landing the Schweizer 330 unmanned helicopter on Navy ships. According to Ernst, “The MQ-8B was in test. It was supposed to go to the ship, but the software wasn’t mature enough. They were having a lot of ‘unexpected maneuvers and behaviors.’ I had to get it certified for the ship and get the landing dynamics straightened out. The problem with unmanned systems is every time you wave-off or change state, you’re changing state in a machine . . . you have to automate the thought processes of a pilot.”

Ernst recalled, “We went out to the ship about six times to get a dynamic interface envelope. Eventually, in about 2009, we deployed with the first workup on the guided missile frigate USS McInerney. We were not necessarily building on a community with lots of experience. We were getting H-60 Seahawk squadrons and giving them secondary duty. You had to make the UAS a lot more robust and a lot easier to learn. That’s kind of the antithesis of helicopters.”

NAVAIR ultimately implemented 42 engineering change proposals from 2010 to 2015 to enhance MQ-8B robustness and reduce false alarms from sensors and systems. Ernst offered, “The challenge of an unmanned system is you don’t have the typical cues and situational awareness from a pilot. Helicopter pilots detect icing by feeling vibration in their feet and their butts; then they look out the window and see ice forming on the windscreen. We don’t get any of that, so we try to give air vehicle operators [AVOs] information without overloading them. The original Fire Scout would get 50 alerts an hour. All that’s doing is taking away useful information. I’ve got to figure out what to do to get that situational awareness so the AVO has enough time to operate. That’s a big issue with unmanned systems and one that’s not getting enough play.”

The Fire Scout endurance upgrade introduced the MQ-8C air vehicle. “The really nice thing about the MQ-8C is it’s based on the Bell 407 airframe which is in production and is a very, very robust airframe,” said Ernst. “So that made life a lot easier, but it still had its challenges. We made the software common between the MQ-8B and C. We went to full agile software development and had to modify the commercial software process itself. We improved some of the automatic testing of scripts. I knew we had to turn unknown changes faster.”

MQ-8C integration also improved NAVAIR coordination with Northrop Grumman and Bell. “The other thing I did was set up a lot of what I consider real-time communications with my contractor engineering counterpart. We processed about 120 changes over the course of those 18 months.” Bringing the MQ-8C to flight test close to schedule and on cost was a major achievement. “We turned around a flight clearance in 4.8 days — including first flight. We were fast. We were not slowing up the contractor. That to me is just an astounding effort.”

Improvements in software qualification have paid off. “We’re turning software around with probably about 10 times fewer defects than we had 10 years ago,” explained Ernst. “The way we did it was to incentivize the contractor and create the culture of doing more. If they find something to fix in test, I tell them to ‘go do it.’ Or more importantly, ‘do it and tell me about it later.’” The UAS chief engineer added,
“There’s been about a 300 to 400 percent improvement in throughput and probably a six-fold improvement in defects. What that’s allowed us to do is to put a payload on the aircraft and get it to flight test in nine to 12 months. I want to get it down to three months.”

MQ-8B weapons integration was another accomplishment. “We’ve really had to define safety requirements and use cases for how we’re going to certify weapons coming aboard a ship,” said Ernst. “You can’t tell the ship guys you’re bringing forward-firing ordnance aboard and there’s no pilot to jettison it, but ‘trust me; we’ve got this.’” The Navy has no requirement to arm the production MQ-8C but NAVAIR has money for non-recurring engineering associated with armament.

Autonomy and Applications
Shipboard UAS operations still pose unique challenges. Ernst observed, “The pilot will do everything possible not to fly through that big, gray thing called a ship. The robot will just follow mechanics. The robot never makes mistakes — it’ll only do what you program it to do. But you’ve got to consider all your frames of reference. Fire Scout navigation, for example, transitions through absolute navigation, position relative to the ship and landing under UCARS [the UAS Common Automatic Recovery System]. As you change those modes and something fails, you’ve got to make sure all those variables translate without a hiccup. That’s why we’ve gone to a lot of automatic testing, because you just can’t sit down and go through all of those perturbations.”

NAVAIR is also working on greater UAS autonomy, teaming with the US Federal Aviation Administration (FAA) on sense-and-avoid and ice detection technologies. “Our landing systems are very automated,” observed Ernst. “We’ve automated some of our responses to ice detection and how we respond to cascading effects. I think that the eVTOL [electric vertical takeoff and landing] world has got to really understand, not only how they automate systems, but how they certify them and under what bounds. That’s why they need to get curmudgeons like me involved in the process and say, ‘These are the things that can go wrong.’” With a distinguished history in fixed- and rotary-wing engineering, Bob Ernst recently joined the Vertical Flight Society. “The reason is the Forum last time was one of the best exchanges of data I’ve seen. We really have to leverage what the commercial world is doing.”


Leadership Profile: Vertiflite May/June 2019