Revolutionizing Autonomous Vehicles: Waabi CEO Raquel Urtasun’s Vision for a Driverless Future
Raquel Urtasun, a renowned expert in artificial intelligence and computer science, is on a mission to transform the autonomous vehicle industry. As the founder and CEO of Waabi, Urtasun has challenged the conventional approach to developing self-driving cars, instead focusing on creating an interpretable and verifiable end-to-end system. This innovative approach has garnered significant attention and investment, with Waabi securing over $200 million in backing from industry giants like Uber, Nvidia, and Porsche.
Courtesy of Waabi
Urtasun’s vision for Waabi is to deploy fully driverless trucks in Texas by the end of 2025, a timeline that would put the company years ahead of its rivals. This ambitious goal is driven by Waabi’s next-generation AV 2.0 approach, which leverages an end-to-end interpretable and verifiable A.I. model powered by the industry’s most realistic neural simulator.
Challenging Conventional Wisdom
Urtasun believes that the prevailing assumption that progress in A.I. is simply a matter of scaling bigger models with more data and compute is misguided. Instead, she argues that the focus should be on developing more efficient and effective A.I. systems that can reason and generalize to new situations. This approach has the potential to unlock significant breakthroughs in autonomous vehicle technology, enabling the safe and efficient deployment of self-driving cars and trucks.
According to Urtasun, the current “more is more” development approach is not only inefficient but also has significant environmental and societal costs. She emphasizes the need for sustainable A.I. development that prioritizes the well-being of humanity and the planet. By focusing on developing more interpretable and verifiable A.I. systems, Waabi aims to create a new generation of autonomous vehicles that are not only safer but also more efficient and environmentally friendly.
A New Era for Autonomous Vehicles
Waabi’s next-generation AV 2.0 approach has the potential to revolutionize the autonomous vehicle industry. By leveraging an end-to-end interpretable and verifiable A.I. model, Waabi’s system can safely generalize to all possible scenarios on the road, including those that are very different from what the model has seen during training. This breakthrough has significant implications for the logistics industry, enabling the safe and efficient deployment of self-driving trucks and transforming supply chains and economies.
Urtasun’s vision for Waabi extends beyond the autonomous vehicle industry, with the potential to empower robots with different form factors, from other types of vehicles to humanoids and warehouse robotics. This could completely reimagine the physical world, unlocking unprecedented efficiencies, enhancing safety, and creating new opportunities for humans.
Proving Safety and Efficacy
As Waabi prepares to deploy its driverless trucks in Texas, the company must demonstrate the safety and efficacy of its system to regulators and the public. Urtasun emphasizes the importance of rigorous testing and validation, including mixed-reality testing and on-road testing. Waabi’s digital twins have achieved a 99.7 percent accuracy score for training scenarios, providing evidence of the system’s safety and reliability.
Urtasun acknowledges that the last 0.3 percent of edge cases could be life-or-death scenarios, but she is confident that Waabi’s approach can address these challenges. By leveraging the industry’s most realistic neural simulator and conducting rigorous testing, Waabi can ensure that its system is safe and ready for deployment.
Image Source: observer.com
