Aicha Evans, CEO of Foster City, California-based Zoox, spoke during a keynote about the challenges facing self-driving cars and the promise they might hold if those challenges are overcome. And for the first time, she detailed a few of the hardware and software stacks underlying the company’s custom-designed vehicle platform — stacks which in part convinced investors to pledge more than $800 million in capital toward Zoox’s research and development.
Evans believes that three core competencies are required to build the driverless car of the future: artificial intelligence, fully autonomous driving, and a battery-powered electric frame designed for AI. “The new ear of mobility requires a radically different approach, not an incremental approach — one that was optimized for human drivers,” said Evans.
She didn’t reveal Zoox’s production vehicle design, but reiterated that it uses a combination of RGB cameras, and lidar (sensors that measure the distance to target objects by illuminating them with laser light and measuring the reflected pulses). Each vision sensor can see 270 degrees, and if one fails, the car retains a 360-degree view of the environment.
It isn’t your average self-driving electric, zero-emissions driverless car. Zoox’s model has four-wheel steering, which Evans claims allows it to follow trajectories with “much higher” accuracy than cars with two-wheel steering. It’s also got a dual power train and dual batteries, and the capacity of both batteries together is substantially larger than that of any single car battery today, Evans claims.
The idea is to reduce congestion through fleet management, and to minimize trips back to base stations for charging overnight. Zoox’s shuttle-like car — which is fully driverless — is designed to operate in a shared fleet in order to maximize efficiency and cut down on ride trip times.
It’s not unlike the self-driving ridesharing network Tesla CEO Elon Musk described earlier this year, during Tesla’s inaugural Autonomy Day .
“[When] you’re not using it, someone else is. This is a much better use of resources,” said Evans. “We believe this technology can solve the challenges facing cities around the world. With it, we can imagine a world where you can choose to live without owning or operating a car.”
Like other self-driving vehicles, Zoox’s use machine learning algorithms to contend with fraught environments they’ve never seen before, like a construction zone. They take in visual data and chart new courses around obstructions and obstacles, all without the need for human intervention.
Zoox is testing the bulk of its vehicles in San Francisco (and a few in Las Vegas), much like Cruise, its GM-owned rival. That’s a conscious choice: as Evans pointed out, San Francisco’s roads pose a challenge even for human drivers. “Our goal is to be safer than humans in developing this,” she said. “The multitude of challenges teach our way our basically how to be better and safer.”
Perhaps unsurprisingly, Zoox is using data collected from the real world to inform simulations that continuously improve its systems’ performance. Its cars drives by city blocks to create topologies, and its engineers use these topologies to create three-dimensional representations that AI agents traverse through endlessly and self-improve. .
Ultimately, when Zoox’s fleet deploys commercially, Evans believes it’ll save riders valuable time — and perhaps more importantly, give them control over their time. She pointed out that an estimated 400 billion annual hours are spent driving cars, and that drivers in San Francisco alone devote a collective 400,000 hours every day to commuting.
“In a more comfortable setting, we will be more productive, connect with friends enjoy music, and yes, some of us may work,” said Evans.
That’s not the only paradigm Zoox’s self-driving cars could change, Evans asserts. They might reduce the need for parking spaces and structures, minimize city congestion (a third of which is caused by drivers searching for parking), and reduce roughly a fourth of air pollution. On that last point, in fact, Evans said that entirely electric self-driving ridesharing fleets could lead to a reduction in CO2 metric tons 85,000 metric tons of CO2.
They could reduce accidents, moreover. About 94% of car crashes are caused by human error . In 2016, the top three causes of traffic fatalities were distracted driving, drunk driving, and speeding, and the National Safety Council pegs Americans’ odds of dying in a car crash at one in 114 .
“We believe in design from the ground up for autonomy, as opposed to [retrofitted] cars designed for human drivers is the right strategy to enable this new era,” said Evans. “But we think it’s a worthy one.”