Waabi has announced a significant advancement in autonomous trucking after successfully deploying its self-driving software on a completely different truck platform without requiring any additional engineering work, retraining, simulation, or new real-world data.
The company revealed that its Waabi Driver system, originally trained exclusively on a Peterbilt 579, was able to take control of a Volvo VNL Autonomous truck and safely operate it on highways as well as complex urban roads from its very first mile.
According to Waabi founder and CEO Raquel Urtasun, the achievement represents more than just a company milestone.
“This was a massive announcement. It was massive for Waabi, but for the industry and Physical AI in general,” she said during an interview with FreightWaves.
The demonstration was carried out in partnership with Volvo Autonomous Solutions. Despite the significant differences between the Peterbilt 579 and the Volvo VNL Autonomous—including sensor placement, vehicle design and driving controls—the software required no modifications before operating the new vehicle.
“The sensors are in very different locations. The shape of the truck is very different. The way you control the Volvo VNL is also very different, and it feels very different from driving a Peterbilt,” Urtasun explained. “Yet we required zero changes. It was directly plug-and-play. Same stack. Same model. Same everything.”
During its first drive, the Volvo VNL Autonomous successfully performed lane changes, navigated traffic lights, executed right turns, crossed three-way intersections and completed U-turns using the existing Waabi Driver software.
Historically, transferring autonomous driving systems between different vehicle platforms has required more than a year of engineering, data collection and validation. Urtasun believes this latest demonstration changes that assumption.
“Up to this point, whenever anyone in the industry wanted to move from one vehicle platform to another, it typically required more than a year of engineering work,” she said. “The system generalized without needing anything.”
Nils Jaeger, President of Volvo Autonomous Solutions, described the test as an important milestone in the collaboration between the two companies.
He said the successful road test demonstrates the scalability of Volvo’s autonomous truck platform, which is designed to support multiple vehicle models and virtual driving systems across a broad range of applications, while bringing commercial autonomous transport closer to reality.
For carriers and truck manufacturers, the breakthrough could significantly accelerate the deployment of autonomous fleets. Waabi identifies two critical forms of generalization needed for large-scale adoption: operating across different driving environments, from highways to urban streets, and functioning across entirely different vehicle platforms. The Volvo demonstration validated the latter.
Urtasun added that the technology is not limited to heavy-duty trucks. The same AI architecture could eventually be applied to medium-duty trucks, robotaxis and other autonomous vehicle categories without requiring entirely new development.
She also noted that the approach would allow companies to adopt new sensor technologies more quickly, avoiding the repeated retraining that has traditionally accompanied hardware upgrades.
According to Waabi, the achievement stems from a different AI architecture than those commonly used across the industry.
Rather than relying primarily on increasing computing power, larger datasets and bigger data centers, the company says its system interprets its surroundings, reasons through potential actions, evaluates their consequences and then selects the safest maneuver in a way that more closely resembles human decision-making.
Urtasun described this reasoning-based approach as a new generation of autonomous driving technology and said Waabi is currently the only company capable of demonstrating these capabilities.
Although Waabi World, the company’s simulation platform, played a role in developing the Waabi Driver system, she emphasized that the Volvo test itself required neither simulation nor additional training.
Looking ahead, Urtasun said the same autonomous “brain” is intended to power multiple vehicle types, including future robotaxi programs.
She believes the company’s experience operating autonomous Class 8 trucks on highways and completing more than a year and a half of door-to-door trucking operations provides a competitive advantage, particularly because highway driving remains one of the biggest challenges for many robotaxi developers.
According to Urtasun, the primary challenge is no longer developing the technology itself but scaling it commercially.
She added that the system delivered strong performance immediately after deployment and that the company is now focused on expanding commercialization while maintaining its safety standards.
Waabi also confirmed that its upcoming driverless launch will operate on a fully validated OEM platform.
“Driverless is almost here,” Urtasun said. “The future of self-driving is truly here.”
