Skip to Content

AutoX Has Built a Self-Driving Car That Navigates with a Bunch of $50 Webcams

The startup wants to make autonomous vehicles cheap enough for everyone to use.
March 28, 2017

While companies like Uber and Google build self-driving cars replete with pricey sensors, Jianxiong Xiao is doing the same thing with some $50 webcams.

Previously an assistant professor of computer science at Princeton, Xiao is the founder and CEO of AutoX, a startup working to make autonomous transportation accessible to everyone. Speaking at MIT Technology Review’s EmTech Digital conference in San Francisco this week, Xiao credited his past with inspiring the startup’s goal: as a child growing up poor in a small town in China, he wasn’t able to visit the ocean until he was 18, despite the fact that it was just 20 miles away.

Now, Xiao said, he imagines a future where kids don’t have to depend on their parents to drive them around, for instance, and can instead summon a self-driving ride.

“Autonomous driving should not be a luxury,” he said.

To make this happen in a way that’s economical, AutoX is eschewing the standard self-driving car sensors like inertial measurement units, lidar, and differential GPS, which can add up to hundreds of thousands of dollars. Instead, over the past six months the company built software—the brains of a self-driving car, essentially—that harnesses seven Logitech webcams that are mounted on the front of a car, arranged to get a 360-degree view.

At EmTech, Xiao showed off some videos of the technology at work, letting a car navigate city streets on its own during the day, in light rain on a winding road, on a cloudy night, and at night on the highway. Another video showed how it can drive in varying lighting conditions, too—such as when going under a freeway overpass in San Jose. These kinds of conditions are still tricky even for self-driving cars bristling with state-of-the-art sensors.

Xiao says AutoX hopes to work with carmakers as well as companies, like Lyft and Uber, that are already helping people get rides.

Keep Reading

Most Popular

still from Embodied Intelligence video
still from Embodied Intelligence video

These weird virtual creatures evolve their bodies to solve problems

They show how intelligence and body plans are closely linked—and could unlock AI for robots.

conceptual illustration showing various women's faces being scanned
conceptual illustration showing various women's faces being scanned

A horrifying new AI app swaps women into porn videos with a click

Deepfake researchers have long feared the day this would arrive.

protein structures
protein structures

DeepMind says it will release the structure of every protein known to science

The company has already used its protein-folding AI, AlphaFold, to generate structures for the human proteome, as well as yeast, fruit flies, mice, and more.

Stay connected

Illustration by Rose WongIllustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.