Uploaded October 2023 | Updated September 2026, 6 hours ago
Maryland Robotics Center Seminar: Gore Robots: From Blood and Guts to Bits and Bytes
Juan Wachs
Adjunct Professor of Surgery
Purdue University
Robots can already solve sophisticated problems ranging from playing games, autonomous driving, and dancing—given enough observational of data for training. The core of such success resides in efficient algorithms, compliant hardware and robust computing, all implemented using carefully curated data collected before the training phase. Thus, robots learn in a “sterile” domain, under clean, controlled and to some extent supervised environments. As the target domain changes, however, moving to more quotidian scenarios, robots struggle to perform well. It is hard to think of an autonomous car trained in Silicon Valley being able to successfully navigate the crowded streets of New Delhi. Ideally, we would like to see robots that can learn while immersed in a non-sterile setting, while trying, exploring, manipulating and probing the environment as a learning strategy. To address this hurdle, my work in the area of robotics and autonomous systems focuses on transferring skills and knowledge from controlled settings to the wild. In this talk, I emphasize strategies and techniques to address fundamental challenges in emergency medicine. Specifically, I will discuss work related to surgical assistants, telesurgery, and skill augmentation. While medicine is the main domain of the research discussed, the outcomes and findings are applicable to the range of field robotics. Progress in these directions will contribute to the public purpose of creating the knowledge for developing robots that are more accessible, effective and sensitive to social needs.
For more information on the Maryland Robotics Center see:
https://robotics.umd.edu
Maryland Robotics Center Seminar: Gore Robots: From Blood and Guts to Bits and Bytes
Juan Wachs
Adjunct Professor of Surgery
Purdue University
Robots can already solve sophisticated problems ranging from playing games, autonomous driving, and dancing—given enough observational of data for training. The core of such success resides in efficient algorithms, compliant hardware and robust computing, all implemented using carefully curated data collected before the training phase. Thus, robots learn in a “sterile” domain, under clean, controlled and to some extent supervised environments. As the target domain changes, however, moving to more quotidian scenarios, robots struggle to perform well. It is hard to think of an autonomous car trained in Silicon Valley being able to successfully navigate the crowded streets of New Delhi. Ideally, we would like to see robots that can learn while immersed in a non-sterile setting, while trying, exploring, manipulating and probing the environment as a learning strategy. To address this hurdle, my work in the area of robotics and autonomous systems focuses on transferring skills and knowledge from controlled settings to the wild. In this talk, I emphasize strategies and techniques to address fundamental challenges in emergency medicine. Specifically, I will discuss work related to surgical assistants, telesurgery, and skill augmentation. While medicine is the main domain of the research discussed, the outcomes and findings are applicable to the range of field robotics. Progress in these directions will contribute to the public purpose of creating the knowledge for developing robots that are more accessible, effective and sensitive to social needs.
For more information on the Maryland Robotics Center see:
https://robotics.umd.edu
