POSE: Phase I: CarDreamer: Towards an Open-Source Ecosystem for World-Model Based Self-Supervised Learning in Autonomous Driving

Award ID: 2449194

This Pathways to Enable Open-Source Ecosystems (POSE) project will enable CarDreamer, the first open-source ecosystem (OSE) focused on resources for the autonomous driving community of researchers. CarDreamer is a bridge that fosters collaboration across communities working to advance artificial intelligence (AI) technologies for autonomous driving. It addresses the needs of its stakeholders, ensuring a cohesive research and development pipeline. Through a variety of scoping and planning activities carefully designed for ecosystem discovery and community building, the CarDreamer OSE will engage potential users and content contributors and provide a suite of tools enabling them to develop their own AI functionalities and seamlessly integrate with other modules, thus catalyzing the sustainable, long-term growth of the OSE. This POSE project will develop CarDreamer as a comprehensive educational resource for embodied AI and the autonomous driving community. The project will help equip a workforce with the skills required to navigate the evolving challenges and opportunities presented by embodied AI and self-supervised learning (SSL) for autonomous driving. The flexible and modular design of the CarDreamer OSE enhances a concerted effort across communities in academia and industry, addressing the needs of different stakeholders in the autonomous driving community and ensuring an integrated research and development pipeline for world-model-based SSL. By providing hands-on experiences, the project invites a broad community, including K-12 students and educators, to engage with complex technological concepts. Engagements, such as hosting summer interns from community colleges, will position to the project to prepare the next-generation workforce in AI technologies for autonomous driving. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Florence Rabanal