For Self-Driving Cars, the Hot New Technology Is… Radar

One of the hottest new technologies for next-generation auto-safety systems and planned “autonomous” vehicles has roots going back a century.

Yes, we’re talking about radar, the same technology that began as a curiosity in the early 20th century, helped Britain to repel the Luftwaffe in World War II, and has long enabled weather forecasting and allowed air-traffic controllers to keep our skies safe.

Today, radar is no longer just for aeroplanes and military installations. A number of new companies on both the hardware and software side are making radar an integral part of safety systems to detect cyclists and pedestrians.

The need for such technology is urgent: Pedestrian deaths in the U.S. have surged in the past few years, even as Americans logged fewer miles on the road. And increasing levels of autonomy in new vehicles that enable features like collision warning, automatic braking, and blind-spot detection—not to mention the driverless cars of the future—are entirely dependent on advanced sensory systems. Such systems are also essential for auto makers to meet their promises to incorporate automatic braking systems into all vehicles by 2022.

To reduce and prevent carnage on our roads, companies like Mobileye, a subsidiary of Intel, are working on chips bristling with tiny radar antennas. General Motors recently invested in Oculii, an all math-and-software startup that uses machine learning to shape the kinds of signals automotive radar systems use. The software company MathWorks is developing algorithms that can allow auto makers to integrate data from radar and other sensors into a trustworthy picture of the world around a vehicle.

For engineers who work on vehicle sensors, now is a time of rapid change, says Erez Dagan, executive vice president of products and strategy at Mobileye. Cameras used in automobiles continue to become higher resolution, and are able to sense a wider range of natural light than they used to. Lidar, which bounces lasers off surrounding objects to “see” the world in 3-D, is becoming less expensive than it used to be. (Lidar is common in robot taxi prototypes, such as those from Waymo, the Google sister company, GM’s Cruise and Amazon’s Zoox.)

Radar, which bounces radio waves off objects—the term was born as an acronym for “radio detection and ranging”—has been used on some first-generation safety systems in vehicles since the 1990s. Automotive radar systems have a number of advantages. They’re tough enough to survive years of jostling and temperature swings when mounted on cars. They’re much, much less expensive than lidar, good at instantaneously measuring the velocity of objects, and able to peer through the kinds of inclement weather, like fog and rain, that can foil both cameras and lidar systems. But they have until recently had one major drawback: They have only a fraction of the resolution of those other systems, which means in essence that the images they produce are much blurrier.

Oculii’s technology works by changing the shape—also known as the waveform—of the radar signal sent out by radar on cars. The physics are complicated, but by changing the nature of the radar signal depending on what sort of objects it’s bouncing off of, it can resolve objects whose shape would be impossible to “see” otherwise. The result, says Chief Executive Steven Hong, is that existing automotive radar sensors, which cost around $50 apiece, can generate three-dimensional images of a car’s surroundings with a much higher resolution. The company’s software is set to make its debut in radar-based safety systems in the forthcoming Lotus Lambda SUV, to be released in 2023.

Leveraging the chip manufacturing abilities of parent company Intel, Mobileye is working on individual microchips covered with nearly 100 tiny antennas. By using artificial intelligence software to process the noisy signals they receive, Mobileye says its systems can do things like identify pedestrians, at least in the lab. That’s something that previously could only be achieved with cameras and lidar.

There is no unanimity among automobile technologists about what configuration of cameras, lidar and radar will become the standard way to achieve various safety systems or autonomous driving, but nearly all agree that the best solution will be some combination of them.

The resolution that even the best automotive radar can achieve is only as good as the worst lidar systems available, says Matthew Weed, an engineer and senior director of product management at Luminar, which makes lidar systems for automobiles. Luminar’s system, which Mr. Weed says is superior to radar for most applications, costs $1,000, however.

Mr. Weed says that Luminar’s lidar-based systems could justify their cost by being so good that they could lower drivers’ insurance costs by preventing accidents and pedestrian deaths. Even with such a system on a car, radar would be a good backup for when it fails or can’t handle severe weather, he adds.

Mobileye uses lidar, cameras and radar in its most advanced systems. CEO Amnon Shashua has said that while lidar systems have come down in price, they are still 10 times the cost of radar, and likely to remain so for the foreseeable future, on account of the complexity of the hardware involved.

Elon Musk’s Tesla has gone all-in on its bet that the company can achieve true autonomous driving in its vehicles using only cameras.

Cameras have the advantage of extremely high resolution, and they’re affordable and compact thanks to years of advances in smartphone cameras. But for a system that can achieve the highest safety standards, and even eventually full autonomy, cameras need backup sensors that fail under different sets of conditions than they do, Mr. Dagan adds.

Take fog, which looks like an obstacle to both camera-based and lidar-based systems, potentially causing vehicles to stop when they shouldn’t. In research published in 2020, radar-based automotive sensors had no trouble penetrating fog and correctly identifying stopped vehicles hidden within it, says Dinesh Bharadia, an assistant professor of engineering at the University of California San Diego who contributed to the work.

Dr. Bharadia says his team found that one key is using multiple radars, spaced at least five feet apart on a vehicle. It’s the same principle at work in the ever-expanding number of cameras on the backs of our smartphones, he adds. It’s possible to build up an “image” of a car’s surroundings using multiple low-cost radar sensors, just as our phone can use multiple small and inexpensive cameras, and then recombining the images they gather into something much sharper.

Bringing all the sensors on a car together into a single, coherent view of the reality outside a vehicle requires fusing all that data together, says Rick Gentile, an engineer who used to work on radar systems for defence applications and is now a product manager at MathWorks, a software company that builds tools to help process data. For example, while radar might be able to detect that there’s a sign up ahead, it can’t see its colour, which is critical to quickly identifying what kind of sign it is.

For so-called robot taxis, the way to make up for the gaps in the capabilities of each kind of sensor is to use all of them. The goal is “full redundancy,” says Mr. Dagan, so that even if one sensor has an error, others perceive the world correctly. This, he argues, is the fastest way to give vehicles senses that are at least as good as a human’s. (Whether those vehicles will have judgment sufficient to actually drive themselves around safely is a separate matter.)

Until we get real autonomous vehicles—something that could be years, if not decades away—auto makers will have to choose among radar, lidar and cameras, or some combination of the three, to create safety systems that can meet their promises to make automatic braking systems standard by 2022, and to continue to improve those systems. All three sensor types continue to get better, but the difference in cost among them has led automakers to favour one technology or another, depending on how well they think they can make up for its deficiencies with software and AI.

This has led to healthy competition among makers of safety systems, sensors and supporting software—whomever you talk to, they argue their systems are the best.

As all of these companies jostle for a place on your car, the goal of all of these technologists is to profit by significantly reducing road fatalities of every kind, when a human is behind the wheel. It’s a goal that they all agree is much closer at hand than fully autonomous vehicles.

In our 20s, my new husband and I took a year off from our fledgling careers to travel in Southeast Asia. Equipped with paper maps, we began in China and improvised each day’s “itinerary” on the go. A gap year for grown-ups, I called it, although I scarcely qualified as one.

Nearly 40 years later, we are new retirees with the same wanderlust. We wondered: Could we recapture the thrill of winging it, enduring rough roads and cheap hotels?

We could and did, but for 2½ months instead of 12. We mapped out a route that would take us up Africa’s east coast and then—who knows where? Here’s how we rolled and five important lessons we learned on a 6,000-mile trip.

Kenya: Live large by day

Our first stop was the tiny, car-free island of Lamu, well-known for its high-profile visitors, from Kate Moss to the Obamas. This low-key getaway offered white-sand beaches, dhows — boats you can rent for day cruises and snorkelling — and lots of donkeys, the main mode of transport.

We considered the beachside Peponi Hotel in Shela, a hot spot since the 1960s (Mick Jagger bunked there). But room rates start at $250, far above our per-night budget of $70 or less. When contemplating almost 100 nights of travel, price matters.

So we chose a villa in the dunes called Amani Lamu, $61 per night for an en suite room with a private terrace and shared plunge pool.

We still had a cool Peponi moment come sunset: On the hotel’s whitewashed veranda, we sipped Pepotinis and plotted our next day’s interlude at the Majlis, Lamu’s fanciest resort (from $580).
With a $20 day pass, we could lounge around its pools and beach bars like proper resort habitués.

Lesson learned: Live like billionaires by day and frugal backpackers by night.
Must-go: Across the bay on Manda Island, bunk a night in a thatched-roof bungalow on stilts at Nyla’s Guest House and Kitchen (from $48 with breakfast).
After a dinner of doro wat, a spicy Ethiopian chicken stew and rice, the sound of waves will lull you asleep.

Egypt: Ask. Politely.

From Lamu, we flew to Aswan in Egypt. Our “plan”: Cruise down the Nile to Luxor, then take a train to Cairo, and venture to Giza’s pyramids.

Turns out it’s the kind of thing one really should book in advance. But at our Aswan hostel, the proprietor, who treated us like guests deserving white-glove service, secured a felucca, a vessel manned by a navigator and captain-cum-cook. Since we’d booked fewer than 24 hours in advance and there were no other takers, we were its sole passengers for the three-day trip.

One day, we stopped to tour ancient temples and visit a bustling camel fair, but otherwise, we remained on board watching the sunbaked desert slide by. We slept on futons on the deck under the stars. The cost: about $100 per night per person, including three meals.

Lesson learned: Ask for help. We found Egyptians kind and unfazed by our haplessness, especially when we greeted them respectfully with assalamu alaikum (“Peace to you”).
Must-go: For buys from carpets to kebabs, don’t miss Cairo’s massive Khan el-Khalili bazaar, in business since 1382. We loved the babouche, cute leather slippers, but resisted as our packs were full.

Turkey: Heed weather reports

Next stop Tunisia, via a cheap flight on EgyptAir. We loved Tunisia, but left after six days because the weather got chilly.

Fair enough, it was January. We hopped continents by plane and landed in Istanbul, where it snowed. Fortunately, two of Istanbul’s main pleasures involve hot water. We indulged in daily hammams, or Turkish baths, ranging from $30 to $60 for services that included, variously, a massage, a scrub-down and a soak.

Beneath soaring ceilings at the temple-like Kılıç Ali Paşa Halamı, brisk workers sternly wielded linen sacks to dowse my body in a cloud of hot foam.
In between visits to Ottoman-era mosques and the city’s spice markets, we staved off the chill by drinking fruity pomegranate tea and sampling Turkish delight and baklava at tea salons.

A favourite salon: Sekerci Cafer Erol in Kadıköy, a ferry-ride away on the “Asian” side of Istanbul, where the city adjoins Asia.

Lesson learned: Pay attention to the weather gods. We foolishly took the concept of travelling off-season too far.
Must-go: Don’t miss the Istanbul Modern, the Renzo Piano-designed art museum in the historic Beyoğlu district.

Cambodia: Chill out

After a long flight from Istanbul, we spent two weeks in Laos and then hopped another plane to Cambodia, specifically Koh Rong Sanloem, another car-free island.

Like vagabonds, we lolled by the warm, super-blue water of Sunset Beach, steps from our bungalow at Sleeping Trees (from $54 per night).
A caveat: You have to sweat to get to this island paradise. We took a bus, a ferry and then hiked for 40 minutes up and down a steep hill and through a jungle. You’ll find only a handful of “resorts”—simple bungalow complexes like ours. There’s nothing much to do. I’ll be back.

Lesson learned: Until our week in Cambodia, we’d been travelling too much and too fast, prioritising exploration over relaxation. This island taught us the pleasures of stasis.
Must-go: Spend one day in Cambodia’s capital city, Phnom Penh, to delve into its sobering history. Tour the Choeung Ek Genocidal Centre, site of a Killing Field, where nearly 9,000 Cambodians died.

Thailand: Be a frugal hedonist

We spent our last two weeks on the island of Ko Samui, where season three of “The White Lotus” was shot.

We went there for its astounding beauty, not the luxury resort experience that comes with too many boisterous lads on vacation, snake farms and traffic jams in town.

Truth be told, we flouted our budget rules to book an Airbnb with a pool (from $300) in the hills of Lipa Noi on the island’s quiet side. We joined the nearby Gravity Movement Gym to work out, but cooked our own meals to keep our final tabulation of expenses within reach.

Lesson learned: Pinching pennies feels restrictive, no matter how lush the surroundings. And it leads to bickering, as partners tally up who squandered how much on what.
With the end in sight, we splurged on the villa and even bought souvenirs, knowing we’d lug them for days, not weeks.
Must-go: Take the 30-minute ferry to sister island Ko Pha Ngan for its peace, love and yoga vibe and, once a month, full-moon parties.
Via Airbnb, we bunked at a Thai house called Baan Nuit, run by the Dear Phangan restaurant proprietors.

We sampled steamed dumplings, white fish in a Thai basil sauce and spicy noodles for a mere $15 apiece.
Hey, indulge in that “White Lotus” moment if you dare!