Wayve Advances AV 2.0 with Generative AI on NVIDIA Platforms

London-based startup Wayve is spearheading a new era in autonomous vehicle (AV) technology, termed AV 2.0, by leveraging generative AI and NVIDIA's advanced platforms, according to the NVIDIA Blog. This innovation marks a significant leap from the traditional AV 1.0, which primarily focused on refining a vehicle's perception capabilities using multiple deep neural networks.

Generative AI and Unified AI Models

AV 2.0 is characterized by large, unified, end-to-end AI models capable of managing various aspects of the vehicle stack, including perception, planning, and control. Wayve's approach integrates these comprehensive in-vehicle intelligence systems to drive decision-making in dynamic, real-world environments.

The startup is developing autonomous driving technologies built on NVIDIA DRIVE Orin and its successor, NVIDIA DRIVE Thor, which utilizes the NVIDIA Blackwell GPU architecture. These platforms are designed for transformer, large language model (LLM), and generative AI workloads, providing the computational power needed for advanced autonomous driving.

Wayve’s Innovations and NVIDIA Collaboration

As a member of the NVIDIA Inception program for cutting-edge startups, Wayve specializes in developing AI foundation models for autonomous driving. These models equip vehicles with a “robot brain” that can learn from and interact with their surroundings. Alex Kendall, co-founder and CEO of Wayve, highlights the critical role of NVIDIA in their development process, stating, “NVIDIA has been the oxygen of everything that allows us to train AI. We train on NVIDIA GPUs, and the software ecosystem NVIDIA provides allows us to iterate quickly — this is what enables us to build billion-parameter models trained on petabytes of data.”

Generative AI also plays a key role in Wayve’s development, enabling synthetic data generation so AV makers can use a model’s previous experiences to create and simulate novel driving scenarios. This capability is crucial for advancing autonomous driving technologies, as it allows for safe and extensive testing in a controlled environment.

Embodied AI and Future Plans

Wayve is building embodied AI, a set of technologies that integrate advanced AI into vehicles and robots to transform how they respond to and learn from human behavior, enhancing safety. Recently, the company announced its Series C investment round, with participation from NVIDIA, to support the development and launch of the first embodied AI products for production vehicles. As Wayve’s core AI model advances, these products will enable manufacturers to efficiently upgrade cars to higher levels of driving automation, from L2+ assisted driving to L4 automated driving.

As part of its embodied AI development, Wayve launched GAIA-1, a generative AI model for autonomy that creates realistic driving videos using video, text, and action inputs. Additionally, the company introduced LINGO-2, a driving model that links vision, language, and action inputs to explain and determine driving behavior.

“One of the neat things about generative AI is that it allows you to combine different modes of data seamlessly,” said Kendall. “You can bring in the knowledge of all the texts, the general-purpose reasoning and capabilities that we get from LLMs and apply that reasoning to driving — this is one of the more promising approaches that we know of to be able to get to true generalized autonomy and eventually L5 capabilities on the road.”