Rishik Sathua
Rishik Sathua is an undergraduate student in Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign, where he conducts robotics and computer vision research under the supervision of Professor Saurabh Gupta. He is best known as a co-author of ARGUS, a pipeline that aligns robot scene geometry across shifting camera viewpoints using large 3D vision models to improve visuomotor policy learning. He has also co-authored work on precise mobile manipulation of small everyday objects published in IEEE Robotics and Automation Letters.
“ARGUS introduces a pre-processing pipeline that uses large-scale 3D vision models to reconstruct a scene from arbitrary camera angles and re-render it from a fixed, canonical viewpoint before passing it to a robot's visuomotor policy.”
Source→“The authors demonstrate that their method converges to high success rates 4–6x faster than previous methods like KYC and classical Diffusion Policy”
Source→“The hardware setup used for evaluating ARGUS relies on Intel RealSense D415 cameras, though notably used only for their RGB stream in the ARGUS pipeline”
Source→“The experiments were conducted using a 7-DOF Kinova Gen 3 robot arm, demonstrating the pipeline's applicability on standard industrial/research manipulators”
Source→“A Robotiq Hand-E gripper was used as the end-effector in the experimental setup”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.