- This event has passed.
MAE Distinguished Lecture: Karthik Ramani
Title:
Shaping the Future: Towards Physical-AI for Understanding, Authoring, and Embodied Interaction
Abstract:
The convergence of sensors, spatial interfaces, and large visual-language AI models are transforming how we perceive, understand, and act in the physical world. Unlike traditional computing paradigms, emerging embodied systems share our viewpoint, context, and real-time spatial interactions. In this talk, I present three key themes from our research illustrating the future of Physical AI—where human-centered spatial computing blends digital intelligence with physical action.
First, I will explore our work on authoring environments that empower non-programmers to create immersive AR/VR applications. Our system, agentAR, enables subject-matter experts to author spatial learning experiences and guide physical workflows using multimodal inputs—voice, gesture, and spatial intent—enabling applications in training, production, and education. Second, I will highlight advances in AI-based perception that detect human actions and generate design responses in real time. Our platform, designfromX, integrates vision and language models to convert verbal prompts and sketches into 3D design alternatives, supporting co-creative workflows for product ideation and fabrication. Third, I will present applications of embodied Physical AI agents in performance and skill augmentation. The avaTTar system for table tennis training fuses biomechanics, visual feedback, and coaching intelligence—offering a glimpse into AI-augmented athletic training. We further demonstrate visuo-haptic interfaces for hands-on manufacturing training in the real-world.
Together, these systems point toward a future where Physical AI enhances how we design, train, and learn—expanding human potential across engineering, production, sports, and beyond. By bridging immersive interfaces and embodied intelligence, we aim to shape a new class of accessible, real-time, and spatially aware engineering systems.
Bio:
Karthik Ramani is the Donald W. Feddersen Distinguished Professor of Mechanical Engineering at Purdue University, with additional appointments in Electrical and Computer Engineering and a courtesy role in the College of Education. He leads the Convergence Design Lab, where his research brings AI into the physical world by blending human-centered AI with spatial intelligence to create immersive, real-time solutions for design, manufacturing, sports training, surgery, and hands-on learning. His work spans augmented spatial interactions, symbiotic human-AI collaboration, computational design thinking, and scalable upskilling platforms. A pioneer in Physical AI, he develops systems that perceive, understand, and act in real environments—extending human capacity through embodied and intuitive interfaces.
He has recently published in top-tier venues across computer vision (CVPR, ECCV, ICCV), human-computer interaction (ACM CHI, UIST), and AI (NIPS, ICLR), in addition to leading engineering design journals. He co-founded VizSeek, the world’s first commercial shape-based search engine for mechanical parts, and ZeroUI, a CES-awarded robotics startup. His educational innovations include Purdue’s Toy Design and Product-Process-Business Model Design courses. Prof. Ramani was a visiting professor at Stanford University and Oxford University and a research fellow at PARC (formerly Xerox PARC). He earned his B.Tech from IIT Madras, M.S. from The Ohio State University, and Ph.D. from Stanford—all in Mechanical Engineering. He currently also serves as coach of Purdue’s Table Tennis team, where research meets passion— in the emerging domain of Athletic AI.