The Consumer Electronics Hall of Fame: DJI Phantom Drone

Dozens of companies vied to build the first commercially successful recreational drone; the competition was won by a Chinese engineer barely out of adolescence

Advertisement

DJI’s original Phantom drone was a remarkable piece of engineering, but if the little flier’s engineering and marketing hadn’t reinforced each other, DJI never would have soared to the top of the commercial drone market.

Around 2005, there were hundreds of companies capable of entering the race to commercialize drones; a few dozen did. And yet somehow it was a guy in China, barely out of adolescence, who managed to completely dominate a market absolutely everyone saw coming. The emergence of the market was not much of a surprise, because in the years and months leading up to it anyone could go on YouTube to watch videos filmed with one-off DIY drones that various research labs had been posting. The first ones featured quadcopters (exotic at the time) wobbling in the air, soon followed by footage of prototypes performing aerial maneuvers with the agility of dragonflies and hummingbirds.

Frank Wang was still a student at Hong Kong University of Science and Technology when he and three friends founded DJI in 2006. DJI was going to commercialize a drone helicopter that Wang had built for a school project. The four moved to Shenzhen, for the same reason that U.S. entrepreneurs flock to Silicon Valley—it’s the place to go in China to find money, manufacturing resources, and engineering talent.

If there was any precedent for drone makers, it might be radio-controlled (R/C) cars. The fat part of the market, at least in terms of unit sales, is for R/C cars priced at US $100 or so. At that price point, they are basically somewhat expensive toys. Some R/C cars are larger, more durable, and more powerful, with prices that range into the thousands of dollars. But for all the extra horsepower and ruggedness, they don’t do much more than the $100 models. They’re still just toys.

DJI got a good start building helicopter drones, but Wang wanted to expand into quadcopters. The company began developing its own technology, including motors, a flight-control system, and gimbals, which are the mechanisms that hold a camera steady even as the drone carrying it pitches and wobbles. The idea was to produce the most agile, stable, and reliable drones possible.

DJI introduced its first quadcopters in 2011 (by then, Wang was the last of the founders remaining at the company). They were only partially assembled, sold for hundreds of dollars, and appealed mainly to professionals and what DJI called “extreme hobbyists.”

DJI was not interested in selling toys, and besides, it couldn’t manufacture its drones at a toylike price point without seriously compromising quality. The company needed to identify markets, and in doing so it took a brilliant tack: It began to catalog the uses not for drones but for flight in general. The list included cinematography, agriculture, energy-sector inspection, infrastructure and construction, and emergency response. Then DJI developed a marketing plan for each of those applications.

Cinematography was the obvious first target market, as many of DJI’s extreme hobbyists were already using its drones for aerial photography. The costs of hiring an airplane or helicopter to take a camera crew aloft reached into the thousands of dollars for even a very short and simple flight. A stable, agile drone that cost hundreds of dollars would make aerial cinematography cheap and easy.

When it came time to design the Phantom, the company had already crafted components that would keep the drone stable enough for photography, including not only its own motor and gimbals but also a GPS-based navigational unit. GPS would enable the drone to remain stationary when aloft, and GPS was also the basis for a feature that would let the drone return to its original launch site should it lose a signal from the remote control. More than a few drones were lost that way before DJI made the use of GPS standard on hobbyist drones.

DJI engineers nailed not only the big things but the small details, too. The company was experienced with software that made remote control simple and easy. And the kicker for the amateur-videography market was a camera mount custom crafted by DJI to hold a GoPro video camera.

Priced at $629, the Phantom was DJI’s first fully assembled model and among the first drones designed explicitly for consumers. To understand its impact when it hit the market in 2013, you need to first understand that essentially all the other quadcopters on the market then were almost impossible for newbies to fly. The drones’ lack of GPS and maddeningly twitchy controls meant they were badly buffeted by breezes and would sometimes go zooming off crazily (and perhaps irrecoverably) in response to tiny joystick movements. But the Phantom was flyable right out of the box for most people. It was affordable enough for serious hobbyists, and cheap enough for anyone who wanted to use it for professional purposes.

The intended audience immediately embraced it. The footage the average person could get could was impressive enough to be compared with the efforts of film professionals. With incremental improvements in the technology, subsequent Phantom models were even steadier in the air.

The company has never discussed the innards of its Phantom drones. There have been multiple references to the company graduating from an 8-bit microprocessor unit to a 32-bit MPU, suggesting that it was an 8-bit MPU in the original Phantom. A 2015 report quotes Chinese-language sources claiming that DJI has evaluated MPUs from MediaTek; the timing suggests that would have been for the current Phantom 4 model. DJI is widely reported to be using Ambarella digital signal processors from WT Electronics Co., almost certainly new to the Phantom 4, introduced in 2016.

All along, the company has never wavered from its integrated engineering-marketing approach, and it has kept pursuing target markets from its original list. For example, in 2015 it introduced a $15,000 8-rotor model, the Agras MG-1, specifically for spraying crops.

The commercial drone market is still small compared to the military one. Estimates of its size vary wildly, but the research firm Gartner predicted in 2017 that revenues from the sale of personal-use nonmilitary drones that year would total $2.36 billion. And with a market share of roughly 75 percent, by some accounts, DJI has no real competitor in the segment.

Advertisement