Waabi

Waabi is an artificial intelligence company working on developing self-driving technology. The company's technology is designed to build a pathway to commercially viable, scalable, autonomous driving, with an artificial intelligence approach using deep learning, probabilistic inference, and complex optimization to create software that is trainable, interpretable, and capable of complex reasoning in a closed-loop simulator for common driving scenarios and edge cases necessary for driver safety.

Waabi was founded in 2021 by CEO Raquel Urtasun and is headquartered in Toronto, Canada. Prior to founding Waabi, Raquel Urtasun ran Uber's Advanced Technology Group in Toronto. The company is focused on driverless trucks, rather than robotaxis, because of the shortage of truck drivers and the ability to deploy fully autonomous trucks on the highways, which tend to be considered simpler than complex city streets.

An introduction to Waabi

The Waabi Driver uses the company's artificial intelligence-based autonomy stack as software with sensors and compute as hardware, which work together to form a solution for factory-level OEM integration, commercialization, and safe deployment. The use of artificial intelligence is used to make a trainable system that can learn from data, speeding up deployment and allowing the system to learn and increase its ability to make complex decision-making necessary for operating safely and keeping drivers around the truck safe. The approach also includes Waabi World, a school for self-driving that works to expose the Waabi Driver to a diversity of scenarios to develop better driving skills. This includes common driving situations and some edge cases to ensure the self-driving vehicle can respond safely to road conditions.

Waabi World is a driving simulation system for the Waabi Driver system that works to expose the AI-based Waabi Driver to a diversity of experiences to help develop its driving skills. The simulator is an immersive and reactive environment intended to design tests, assess skills, and teach the self-driving Waabi Driver to learn on its own and reduce the need to drive, testing miles in the real world while still producing safer and more affordable results. Waabi World is built around four core capabilities:

• Building digital twins of the world from data
• Performing near real-time high-fidelity sensor simulation to help test the entire software stack in an immersive and reactive manner
• Creating scenarios to stress-test the Waabi Driver
• Teaching the Waabi Driver to learn from its mistakes and learn to drive without human intervention