Wayve CEO participates in his main components to expand independent driving technology

Alex Kendall, co -founder of Wayve CEO, sees a promise to provide the starting technology of his independent car to the market. That is, if Wayve holds its strategy to ensure that the automatic driving program is cheap for its operation, inappropriate devices, and it can be applied to advanced drivers assistance systems, robotics, and even robots.

The strategy, which Kendall developed through Nvidia GTC ConferenceIt begins with the data -based learning approach from end to end. This means that what the system sees “sees” through a variety of sensors (such as cameras) translates directly to how it (such as making brakes or moving to the left). Moreover, this means that the system does not need to rely on HD maps or rules -based programs, as happened in previous versions of AV Tech.

The approach attracted investors. Wayve, which was launched in 2017 and has it I raised more than $ 1.3 billion Over the past two years, you plan to license its self -driving programs for car and fleet partners, such as Uber.

The company has not yet announced any car partnerships, but an official spokesman told Techcrunch this way in “strong discussions” with several original equipment manufacturers to integrate its programs into a group of different types of vehicles.

Its cheap software field is very important in obtaining these deals.

Kendall said that OEMS puts the ADAS system in new production vehicles that do not need to invest anything in additional devices because technology can work with current sensors, which usually consist of surrounding cameras and some radar.

Wayve is also “Silicon-Agnostic”, which means that he can run his program on any GPU of his OEM partners already in their cars, according to Candal. However, the current development fleet of ON-A-CBP is used.

Kendall said at the theater on Wednesday: “Entering Adas is really very important because it allows you to build sustainable work, build a large scale distribution, and to be exposed to data to be able to train the system until level 4.”

(The 4 -level driving system means that it can move in an environment on its own – under certain circumstances – without the need for human intervention.)

Wayve plans to market its system at AdAS first. Therefore, start starting the AI’s operating program to work without Lidar – Discovering light and radar that measures the distance using laser light to create a very accurate 3D map in the world, which is most of the companies that develop level technology 4 main sensors.

The path of the road to autonomy is like Tesla, which is It also works on a deep to end -tip learning model to operate its system and constantly improve its self -driving program. Since Tesla is trying to do, Way Reflects to take advantage of the ADAS wide -ranging process to collect data whose system will help reach full self -rule. (Tesla “full self” can perform some of the automatic driving tasks, but it is not completely independent. Although the company aims to launch the Robotaxi service this summer.)

One of the main differences between Wayve’s and Tesla approaches from the technology point of view is that Tesla depends only on cameras, while Wayve is happy to integrate Lidar to reach full self -rule in the short term.

“In the long run, there is definitely an opportunity when you build reliability and the ability to verify the validity of a level of size to reduce this (sensor suite) further.” Kindle said. “This depends on the product experience you want. Do you want the car to go faster through fog? Then you may want other sensors (such as Lidar). But if you are ready to understand Amnesty International, the restrictions of the cameras and to be defensive and conservative as a result?

Kendall also disturbed the GAIA-2, the latest global model of Wayve specifically designed for independent driving, which trains its operating program on huge quantities of both realistic and artificial data through a wide range of tasks. The model processes video, text and other procedures together, which Kendall says is more adaptive and like a person’s AI driver in his driving behavior.

“What really exciting for me is the driving behavior that looks like a person you see appears,” Kendall said. “Of course, there is no manually encrypted behavior. We do not tell the car about how to act. There is no infrastructure or high -resolution maps, but instead, the emerging behavior depends on data and enables driving behavior that deals with very complex and varied scenarios, including the previously previously scenarios during training.”

Wayve shares a philosophy similar to the start of transporting independent trucks, which also follows a comprehensive educational system. Both companies emphasized the scaling data that depends on data Artificial intelligence models that can be circulated Through different leadership environments, both depend on AI gynecological simulation devices To test and train technology.