28-year-old self-made billionaire started his self-driving car company in his parents' garage—now he's worth $1.6 billion

When Austin Russell was a 17-year-old high school student, he founded Luminar Technologies, which makes laser sensors that can help self-driving cars detect nearby objects.

"I started off at a very early age, always wanting to know the hows and whys of how to build and create things," Russell tells CNBC Make It. "When I was in my early teens, I converted my parents' garage into an optics and electronics lab ... I saw opportunities to be able to build new kinds of optical systems from the ground up."

Now 28, Russell is the world's youngest self-made billionaire, according to Forbes, as Luminar competes with major companies like Tesla and Alphabet's Waymo.

Russell has an estimated net worth of $1.6 billion, and his Orlando, Florida-based tech startup — which went public in 2020 — currently has a market cap of $2.59 billion. Luminar reported $40.7 million in 2022 revenue, and has partnerships with companies like Mercedes-Benz and Volvo.

The company didn't really get off the ground until after Russell dropped out of Stanford University in 2012, at age 18. He applied for and won a Thiel Fellowship, which pays students $100,000 to focus on their startups full-time.

Even without the fellowship, Russell says he would have still left school to build Luminar: "It was going to happen anyway."

But self-driving cars aren't exactly commonplace on American roads, meaning Russell and Luminar have a lot of work ahead of them, despite their lofty valuations.

Here's how Luminar's technology works, how it differs from Tesla's self-driving tech and what improvements still need to be made before fully autonomous vehicles can dominate public roads.

Luminar uses laser-powered lidar technology, which stands for "light detection and ranging," to generate a 3-D map of a car's surroundings in real time. Once a car can detect nearby objects, it can theoretically avoid hitting them.

"It sends out laser pulses and we're able to do this millions of times to be able to successfully measure exactly how far away everything is, down to centimeter-level precision," Russell says.

The technology is meant to help cars automatically brake in front of unexpected pedestrians, or safely avoid other vehicles drifting into their lanes — not to replace drivers altogether, says Russell. Car crashes may be impossible to permanently eradicate, but if applied appropriately, the tech can still "save lives," he adds.

Elon Musk's Tesla also seeks to popularize self-driving cars, and Musk's approach to the technology is entirely different. Among all the companies striving to make autonomous driving safe, Tesla alone uses "vision-only" technology, says author and artificial intelligence expert Mario Herger.

Specifically, that means "using only cameras," Herger says. "Namely, eight cameras." The cameras map a car's environment "at up to 250 meters of range," helping it autonomously switch lanes or find you in a parking lot, according to Tesla's website.

Tesla's "autopilot" features have contributed to 736 U.S. crashes, at least 17 fatal incidents and five serious injuries since 2019, the Washington Post reported last month. (Tesla didn't immediately respond to CNBC Make It's request for comment.)

Lidar could potentially change that, Russell says: Cameras can be helpful "for certain use cases," but lidar is "a huge step forward."

"It's a true 3D perception of the world, as opposed to a 2D perception [where] you're trying to be able to guess where things are," he says, adding: "When it comes to safety and autonomy, it's all about that last 1% of all the different things that can happen ... Getting something 99% accurate is not nearly good enough for cars."

The lack of self-driving cars on most roads could indicate that nobody's quite achieved 100% accuracy yet. The tech needs to improve in two areas, says Herger: collaboration and affordability.

Autonomous vehicles will be safer once cameras, lasers and sensors can all be combined together, he says: Using multiple technologies simultaneously and "creating algorithms" that process information faster could reduce car crashes, whether by human error or vehicle-caused.

"There may be situations where you need another sensor because [the cameras] can't see through a light, or they can't look through fog," Herger says. "If the lidar says [there's an object] right there when you can't see it, could it be wrong? Is it not functioning?"

But safety may not matter if nobody can afford to buy the car. Tesla's full self-driving beta costs upwards of $15,000. Lidar technology can cost "tens or hundreds of thousands of dollars" to develop and manufacture, Russell says.

Luminar's sensors currently cost between $500 and $1,000 apiece, Russell adds. With mass production, he hopes to reduce those prices — to as low as $100 per sensor, reportedly.

"When you commit to doing something, you have no option but to be able to get it to work out," he says. "Failure is not an option. And I know it certainly won't ever be for for Luminar."

DON'T MISS: Want to be smarter and more successful with your money, work & life? Sign up for our new newsletter!

Take your business to the next level: Register for CNBC's free Small Business Playbook virtual event on August 2 at 1 p.m. ET to learn from premier experts and entrepreneurs how you can beat inflation, hire top talent and get access to capital.