Roy Xing
@royxing.bsky.social
Robotics PhD student at Dartmouth doing research in the A2R Lab [@a2r-lab.bsky.social]. Researching MPC+RL applied to locomotion and manipulation, all that fun stuff! @NSF GRFP Fellow
🤖
🤖
Reposted by Roy Xing
Did a fun lecture this week on data-drive control with hints of behavior cloning and diffusion policies, but all grounded in classical linear systems theory. Most of the ideas are drawn from the behavioral control theory literature: youtu.be/fnajY4Ip13w?...
Optimal Control (CMU 16-745) 2024 Lecture 24: Data-Driven Control and Behavior Cloning
YouTube video by CMU Robotic Exploration Lab
youtu.be
April 19, 2025 at 10:42 PM
Did a fun lecture this week on data-drive control with hints of behavior cloning and diffusion policies, but all grounded in classical linear systems theory. Most of the ideas are drawn from the behavioral control theory literature: youtu.be/fnajY4Ip13w?...
Is the difficulty the robotics-related infrastructure that needs to go into RL, whose core algorithms appear to be fairly consistent (e.g. variations of PPO)? Such as the accurate physics simulators, robot models, and the high compute?
June 18, 2025 at 6:17 PM
Is the difficulty the robotics-related infrastructure that needs to go into RL, whose core algorithms appear to be fairly consistent (e.g. variations of PPO)? Such as the accurate physics simulators, robot models, and the high compute?