IEEE CASE 2025 Workshop

Generative Physical AI for Robotics

August 21, 2025 - Los Angeles, California, USA

Announcements

17/06/2025 the Call for Papers is now open!

Call for Papers

The Generative Physical AI for Robotics workshop will explore the convergence of advanced generative AI models and physical robotic systems, with a focus on learning-driven adaptability, knowledge-aware reasoning, and real-world deployment. The event will feature invited talks, a poster session, and spotlight paper presentations to foster exchange across academia and industry. Topics include but are not limited to:

Generative modeling for control and behavior synthesis: real-time behavior prediction, simulation-to-reality transfer, stochastic modeling of actions and states, etc.
Knowledge-aware systems: structured representations, semantic grounding, domain-specific reasoning, contextual decision-making, etc.
Learning and adaptation: reinforcement learning integration, continual refinement through real-world feedback, adaptive policy generation, etc.
Applications in dynamic and complex environments: disaster response, autonomous navigation, human-robot interaction, precision manufacturing, healthcare robotics, etc.
Challenges and future directions: data efficiency, interpretability, safe deployment, interdisciplinary approaches bridging AI and physical systems, etc.

Paper Submission

About

The primary objective of the "Generative Physical AI for Robotics" half-day workshop at IEEE CASE 2025 is to foster an interdisciplinary dialogue and collaboration among researchers, practitioners, and industry experts at the cutting edge of artificial intelligence and robotics. The workshop seeks to explore how generative models and novel computational frameworks can inform the design, control, and adaptation of physical robotic systems, emphasizing both the theoretical underpinnings and practical implementations. Participants are expected to discuss how generative approaches can streamline robotic learning processes, optimize hardware performance, and enable adaptive responses in real-world scenarios.

Another central goal of the initiative is to cultivate innovative applications that merge data-driven techniques with physical system dynamics. By showcasing the latest research in generative physical AI, the workshop aims to challenge conventional robotics paradigms and inspire the development of new algorithms that integrate simulation, real-world experimentation, and machine learning. The session will include keynote talks, interactive demonstrations, and collaborative breakout sessions, each designed to highlight breakthroughs in generative methodologies that support self-adaptive behaviors and robust decision-making in complex environments.

Ultimately, the workshop aspires to serve as a catalyst for next-generation robotics innovation by bridging the gap between abstract generative models and their tangible implementations on physical platforms. It will provide a platform where emerging ideas, experimental results, and practical insights converge to redefine the future of robotic intelligence. Through this convergence, the event intends not only to enhance academic and industrial collaborations but also to pave the way for more resilient, efficient, and intelligent robotic systems capable of transforming a wide array of sectors from manufacturing to healthcare.