How Wayve’s driverless cars will tackle one of their biggest challenges yet

Understanding why the model behaves the way it does tells Wayve what types of scenarios need additional help. Using a hyper-detailed simulation tool called PRISM-1, capable of reconstructing 3D street scenes from video footage, the company can generate tailor-made scenarios and run through them again and again until what she learns to manage them. How much retraining might the model need? “I can’t tell you the amount. It’s part of our secret sauce,” says Rus. “But it’s a small amount.”

The autonomous vehicle industry is known for its hype and over-promising. Over the past year, Cruise has fired hundreds of people after his cars caused chaos and injuries on the streets of San Francisco. Tesla is under federal investigation after its driver-assistance technology was blamed for several crashes, including a fatal collision with a pedestrian.

But the industry continues to move forward. Waymo said it now makes 100,000 robotaxi trips per week in San Francisco, Los Angeles and Phoenix. In China, Baidu says it offers some 287,000 rides in a handful of cities, including Beijing and Wuhan. Undaunted by claims that Tesla’s driver-assist technology is unsafe, Elon Musk announced his Cybercab last week with a timeline that would put these driverless concept cars on the road by 2025.

What should we think of all this? “Competition among robotaxi operators is intensifying,” says Crijn Bouman, CEO and co-founder of Rocsys, a startup that makes charging stations for autonomous electric vehicles. “I think we’re close to their ChatGPT moment.”

“The technology, the business model and the consumer appetite are all there,” says Bouman. “The question is which operator will seize the opportunity and emerge victorious. »

Others are more skeptical. We need to be very clear what we’re talking about when we talk about autonomous vehicles, says Saber Fallah, director of the Connected Autonomous Vehicles Research Lab at the University of Surrey, UK. Some Baidu robo-taxis still require a safety driver behind the wheel, for example. Cruise and Waymo have shown that fully autonomous service is viable in some locations. But it took years to train their vehicles to operate on specific streets, and expanding routes safely beyond existing neighborhoods will take time. “We won’t have robot taxis that can go anywhere anytime soon,” Fallah says.

Fallah is of the extreme opinion that this won’t happen until all human drivers hand in their licenses. For robo-taxis to be safe, they must be the only vehicles on the road, he says. He believes that current driving models are still not good enough to interact with the complex and subtle behaviors of humans. There are simply too many edge cases, he says.

Wayve is betting its approach will prevail. In the United States, it will start by testing what it calls an advanced driver assistance system, technology similar to Tesla’s. But unlike Tesla, Wayve plans to sell this technology to a wide range of existing automakers. The idea is to build on these foundations to achieve full autonomy in the coming years. “We will have access to scenarios encountered by many cars,” Rus explains. “The path to fully autonomous driving is easier if you progress level by level.”

But cars are just the beginning, Rus says. What Wayve is actually building, he says, is an embodied model that could one day control many different types of machines, whether they have wheels, wings or legs.

“We are an AI boutique,” ​​he says. “Driving is an important step, but it’s also a stepping stone. »