David Held

David Held is an Associate Professor at Carnegie Mellon University's Robotics Institute, leading the Robots Perceiving And Doing (RPAD) lab. His work focuses on perceptual robot learning, integrating robotics, machine learning, and computer vision to enable robots to interact with complex environments. He holds a Ph.D. in Computer Science from Stanford University, an M.S. and B.S. in Mechanical Engineering from MIT, and conducted postdoctoral research at UC Berkeley. His research spans object manipulation, autonomous driving, and reinforcement learning, with a focus on robust perception and control in dynamic settings.

Research Interests: Developing methods for robots to manipulate novel objects, handle deformable materials, and operate in unstructured environments through deep learning and simulation-to-real transfer. He explores autonomous driving via self-supervised learning and semi-supervised techniques.

Notable Articles (2024–2025): Focus on articulated object manipulation, sim2real transfer, safety-aware policies, and perception in robotics. Recent work includes ArticuBot for universal manipulation policies and SplatSim for zero-shot transfer using Gaussian splatting.

• Awards: Google Faculty Research Award (2017), NSF CAREER Award (2021).
• Labs: RPAD Lab, CMU Center for Autonomous Vehicle Research.
• Teaching: Courses include Statistical Techniques in Robotics and Advanced Computer Vision.