TIER IV and CMU partner to pioneer "Level 4+" autonomy through Safety21: Advancing beyond the limits of embodied AI

TOKYO, April 24, 2025 /PRNewswire/ -- TIER IV, the pioneering force behind the world's first open-source software for autonomous driving, is proud to announce a strategic collaboration with Carnegie Mellon University (CMU), widely regarded as a birthplace of autonomous vehicles in 1984, to realize the new Level 4+ autonomy concept. Together, they aim to advance scalability, explainability, and safety through a hybrid architecture that combines data-centric AI approaches with the best practices in robotics, while also unlocking the potential of embodied AI to improve transparency and traceability in decision-making.

This collaboration is further strengthened through Safety21, the US Department of Transportation's National University Transportation Center for Safety, led by CMU Professor Raj Rajkumar. TIER IV has joined Safety21's Advisory Council, promoting the value of open-source software through Autoware*, which serves as the foundation for state-of-the-art research and development that addresses the trade-offs between safety and user experience in autonomous driving systems.

Background

Traditional Level 4 autonomy has been built on robotics methods such as probabilistic estimation and machine learning, relying on hand-crafted behavioral rules, predefined high-definition maps, and localized data sets to coordinate core functions such as sensing, localization, perception, planning, and control. Autoware originated from this architecture and has been successfully deployed in autonomous driving systems around the world.

The new Level 4+ autonomy concept, advocated through this collaboration, represents an intermediate step between SAE J3016 Level 4 and Level 5. It remains within the Level 4 classification in terms of human roles, but incorporates key aspects of Level 5 system features. As a result, the vehicle can operate under virtually all conditions by flexibly expanding its operational design domains (ODDs) to cover previously unencountered scenarios.

The Level 4+ system features do not require the human to take over dynamic driving tasks (DDT). However, they may leverage additional information provided from outside the system, as part of strategic functions, to dynamically respond to environmental changes within the target operational domain (TOD). Meanwhile, the system continues to control tactical and operational functions. In this framework, the system retains full responsibility for safety assurance, even when external strategic input influences its behavior. For example, a human may provide guidance that adjusts waypoint planning at runtime to help the system align its behavior with both the defined ODD and the TOD.

Emerging end-to-end AI models, a key variant of data-centric AI approaches, are promising for realizing Level 4+ autonomy, particularly when integrated with rule-based systems and human-in-the-loop strategies. However, they also present critical challenges, including high data requirements, limited explainability in decision-making, and difficulties in establishing robust safety assurance. Because it is often unclear how such models generalize learned behaviors or what influences their outputs, ensuring trustworthy real-world deployment remains a key hurdle.

Strategic collaboration scope

To realize Level 4+ autonomy, TIER IV and CMU will jointly develop next-generation Autoware-based autonomous driving systems, with a particular focus on advancing  scalability, explainability, and safety through a hybrid architecture that combines data-centric AI approaches, including refined end-to-end AI models, with the best practices in robotics.

The project will center on modularizing end-to-end AI models to coexist with state-of-the-art robotics methods and safety enforcement mechanisms. Key challenges include incorporating intermediate representations that expose internal reasoning processes, as well as establishing contextual awareness features that allow the system to transition into minimum risk maneuver (MRM) mode when faced with unexpected circumstances. As a whole, this hybrid architecture is designed to support more transparent and traceable embodied AI decision-making, while addressing practical deployment concerns such as system accountability, safety assessment, and regulatory compliance.

The collaboration is structured as a three-year initiative. In the first year, TIER IV and CMU will develop a reference vehicle powered by Autoware, with early deployments planned in both Tokyo and Pittsburgh. This vehicle will also serve as a platform for real-world data collection, closed-loop verification and validation, and on-road experimentation. Insights from this phase will be reflected back into Autoware and the broader open-source software community, accelerating innovation at CMU and across the global research and developer ecosystem

In the second and third years, the focus will shift toward developing a comprehensive safety enforcement mechanism for the hybrid architecture. This mechanism will support the safety assessment processes required for vehicle certification and public road approval, and will underpin scalability, explainability, and safety necessary for trustworthy real-world deployment of Autoware-based autonomous driving systems.

"This collaboration marks a major milestone in uniting the strengths of AI and robotics to build autonomous driving systems that are safer, more scalable, and more explainable," said Shinpei Kato, founder and CEO of TIER IV. "Partnering with CMU enables us to further advance Autoware's capabilities and deepen our contribution to the open-source software community, driving the future of mobility through collective innovation."

"The integration of modular end-to-end AI models with traditional Level 4 systems represents a powerful advancement for the new Level 4+ autonomy concept," said Raj Rajkumar, George Westinghouse Professor in the Department of Electrical and Computer Engineering at Carnegie Mellon University. "This collaboration offers a unique opportunity to deepen our understanding of AI in autonomous driving systems, ultimately enabling safer, more scalable, and more effective deployments."

*Autoware is a registered trademark of the Autoware Foundation.

About TIER IV

TIER IV stands at the forefront of deep tech innovation, pioneering Autoware, the world's first open-source software for autonomous driving. Harnessing Autoware, we build scalable platforms and deliver comprehensive solutions across software development, vehicle manufacturing, and service operations. As a founding member of the Autoware Foundation, we are committed to reshaping the future of intelligent vehicles with open-source software, enabling individuals and organizations to thrive in the evolving field of autonomous driving.

About Carnegie Mellon University

Carnegie Mellon University, based in Pittsburgh, Pennsylvania, USA, is a private, global research university and stands among the world's most renowned educational institutions. It is considered to be a global leader in computer science, artificial intelligence, machine learning, robotics and drama, among other fields. The Tartan Racing team from Carnegie Mellon won the 2007 DARPA Urban Challenge that triggered the creation of today's global autonomous vehicle industry.

About Safety21

Funded by the U.S. Department of Transportation as the National University Transportation Center for Safety and led by Carnegie Mellon University, a birthplace of automated vehicles, Safety21 actively seeks to enable and accelerate a safer, more efficient transportation network. By leveraging new technologies and revolutionary trends in transportation, Safety21 aims to research, develop and deploy cutting-edge technologies and policies, and develop workforce and educational programs that directly address the challenges of integrating autonomous, connected, electric and shared vehicles with a transformative focus on safety, innovation, and economic growth.

About Professor Raj Rajkumar

Raj Rajkumar is the George Westinghouse Professor in the Department of Electrical and Computer Engineering at Carnegie Mellon University where he directs the US DOT Safety21 National University Transportation Center and its Metro21 Smart Cities Institute.  He is considered to be a pioneer of connected and autonomous vehicle technologies.

Media Contact
[email protected]

SOURCE Tier IV, Inc.