{"id":44330,"date":"2026-09-22T15:51:44","date_gmt":"2026-09-22T19:51:44","guid":{"rendered":"https:\/\/www.bu.edu\/cise\/?p=44330"},"modified":"2026-09-22T15:52:07","modified_gmt":"2026-09-22T19:52:07","slug":"how-gentle-is-gentle-enough-building-robots-that-know-where-to-stop","status":"publish","type":"post","link":"https:\/\/www.bu.edu\/cise\/how-gentle-is-gentle-enough-building-robots-that-know-where-to-stop\/","title":{"rendered":"How Gentle Is Gentle Enough? Building Robots That Know Where to Stop"},"content":{"rendered":"<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">Imagine handing a surgeon a scalpel. You don\u2019t need to place it in an exact spot down to the centimeter\u2014you would just need to get it close and, above all, not stab anyone by accident. That distinction between precision and safety sits at the center of CISE Faculty Affiliate and Assistant Professor (ME, SE) <\/span><a href=\"https:\/\/www.bu.edu\/cise\/profile\/andrew-p-sabelhaus\/\"><span style=\"font-weight: 400;\">Andrew Sabelhaus\u2019<\/span><\/a><span style=\"font-weight: 400;\"> research on soft robots.<\/span><span style=\"font-weight: 400;\"><\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\"><img loading=\"lazy\" src=\"\/cise\/files\/2026\/09\/Drew-Sabelhaus.jpg\" alt=\"\" width=\"231\" height=\"231\" class=\"wp-image-44331 alignleft\" srcset=\"https:\/\/www.bu.edu\/cise\/files\/2026\/09\/Drew-Sabelhaus.jpg 600w, https:\/\/www.bu.edu\/cise\/files\/2026\/09\/Drew-Sabelhaus-150x150.jpg 150w, https:\/\/www.bu.edu\/cise\/files\/2026\/09\/Drew-Sabelhaus-550x550.jpg 550w, https:\/\/www.bu.edu\/cise\/files\/2026\/09\/Drew-Sabelhaus-300x300.jpg 300w, https:\/\/www.bu.edu\/cise\/files\/2026\/09\/Drew-Sabelhaus-100x100.jpg 100w\" sizes=\"(max-width: 231px) 100vw, 231px\" \/><\/span><span style=\"font-weight: 400;\">Sabelhaus directs <\/span><a href=\"https:\/\/sites.bu.edu\/srclab\/\"><span style=\"font-weight: 400;\">BU\u2019s Soft Robotics Control Lab<\/span><\/a><span style=\"font-weight: 400;\">, where his team builds robots from rubber and other pliable materials, then develops the mathematics that lets them move on their own. Unlike the rigid metal robots that dominate factory floors, soft robots are meant to touch things, bump into people, and operate in environments that can\u2019t be fully planned out in advance\u2014which raises questions that don\u2019t come up on an assembly line.<\/span><span style=\"font-weight: 400;\"><br \/>\n<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">\u201cThere are really deep questions about what it means for that robot to actually do its task right,\u201d Sabelhaus said, \u201cor what it means to even do a task in the first place.\u201d What does \u201csafe\u201d mean for a robot helping a doctor or lifting a patient out of bed? How much contact is too much? Sabelhaus\u2019 lab is trying to turn these doubtful, human questions into precise engineering answers.<\/span><span style=\"font-weight: 400;\"><br \/>\n<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">The lab\u2019s working hypothesis is that there is no single, universally accepted definition of \u201csafety\u201d or \u201csuccess.\u201d However, there is a language for specifying an explanation, case by case. Using a mathematical framework called temporal logic, engineers can encode rules like \u201cforce must stay under this threshold\u201d or \u201cthis action must happen within this time limit.\u201d From there, Sabelhaus\u2019 team works out how to translate that logic into an actual robot\u2019s motion\u2014essentially building an autonomous system that inflates or moves in exactly the way needed to satisfy the specification.<\/span><span style=\"font-weight: 400;\"><br \/>\n<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">It is a distinctly control-oriented approach that gives the lab its name. Other soft robotics labs, including several at BU, build robots that are far more mechanically sophisticated. However, Sabelhaus is more interested in how well the robot can make decisions. A complex soft robot with limited sensing and no real feedback logic, he explained, isn\u2019t much different from a factory arm: great at repeating the same motion, but unable to adapt when something changes. His lab instead tried to build simple robots, but well-instrumented enough to make nuanced decisions grounded in solid math.<\/span><span style=\"font-weight: 400;\"><br \/>\n<img loading=\"lazy\" src=\"\/cise\/files\/2026\/09\/HortonFingerTouch_1000wide_2022-11-11-1-636x254.png\" alt=\"\" width=\"499\" height=\"199\" class=\" wp-image-44333 alignright\" srcset=\"https:\/\/www.bu.edu\/cise\/files\/2026\/09\/HortonFingerTouch_1000wide_2022-11-11-1-636x254.png 636w, https:\/\/www.bu.edu\/cise\/files\/2026\/09\/HortonFingerTouch_1000wide_2022-11-11-1-768x307.png 768w, https:\/\/www.bu.edu\/cise\/files\/2026\/09\/HortonFingerTouch_1000wide_2022-11-11-1.png 1000w\" sizes=\"(max-width: 499px) 100vw, 499px\" \/><\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">The project Sabelhaus is most excited about right now is safety-verified control\u2014engineering a way to mathematically guarantee that a soft robot won\u2019t apply too much force, even while it\u2019s actively touching something. Last year, his lab demonstrated the first hardware example of a verifiably safe soft robot making contact with its environment, with provable bounds on how hard it pushes.<\/span><span style=\"font-weight: 400;\"><br \/>\n<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">The next step is applying that idea to real tasks, and one of Sabelhaus\u2019 favorite examples is agricultural: teaching a robot to pick a berry. Success means the berry actually detaches from the bush\u2014but squeeze too hard and the berry is crushed. It\u2019s a small, well-defined problem, which is exactly the point. Sabelhaus is cautious about the idea of robots making sweeping autonomous decisions in high-stakes settings like surgery. \u201cEverything that my group is working on is really taking really small pieces of the problem,\u201d he said\u2014mapping a narrow slice of human expertise into something precise enough for a robot to execute, before ever thinking about bigger decisions.<\/span><span style=\"font-weight: 400;\"><br \/>\n<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">When asked whether large language models and generative AI might eventually reshape this work, Sabelhaus was candid about the uncertainty. Safety-based robot control has relied on similar mathematical foundations for decades, and the rise of generative AI has \u201cturned all that on its head.\u201d Though exactly how the two will merge, if they do, remains an open question. In Sabelhaus\u2019s view, the largest defining challenge in soft robot control isn\u2019t hardware or algorithms; it is translation. Humans can describe what should or shouldn\u2019t happen using ordinary language, but converting that intuition into a precise engineering specification\u2014how hard is \u201ctoo hard,\u201d which finger should make contact, what the time limit should be\u2014remains largely unsolved. Bridging that gap between human intuition and formal specification is a problem his field will eventually have to confront directly.\u00a0<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">This semester, Sabelhaus is focused on teaching two courses that might look different on paper but share the same mathematical backbone: <strong>EK 103 (Computational Linear Algebra)<\/strong>, for first-year undergraduates, and <strong>ME 501 (Dynamical Systems Theory)<\/strong>, for first-year graduate students. In both, he enjoys showing students the classical version of a problem\u2014like positioning a rigid robot arm in space, a technique that\u2019s been well understood for decades\u2014before posing the harder question: what happens when the robot is soft and touches a person unexpectedly during a medical procedure? For Sabelhaus, that\u2019s where the real research begins, using tools that started in an introductory classroom and ended up back in the lab.<\/span><\/p>\n<p style=\"text-align: justify;\"><span style=\"font-weight: 400;\">Sabelhaus held a NASA Space Technology Research Fellowship with NASA Ames Research Center, then went on to an Intelligence Community Postdoctoral Research Fellowship at Carnegie Mellon University before arriving at BU. Both experiences shaped the kinds of questions he considers worth pursuing, pushing him to think about broader societal priorities rather than just what he personally finds interesting. That instinct to question his own assumptions feeds directly into his research on defining safety and success\u2014an exercise in staying open to the possibility that there isn\u2019t one obvious right answer. Sabelhaus received the NSF CAREER Award in 2024, and in 2026 he was named Professor of the Year in Mechanical Engineering at BU.\u00a0<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Imagine handing a surgeon a scalpel. You don\u2019t need to place it in an exact spot down to the centimeter\u2014you would just need to get it close and, above all, not stab anyone by accident. That distinction between precision and safety sits at the center of CISE Faculty Affiliate and Assistant Professor (ME, SE) Andrew [&hellip;]<\/p>\n","protected":false},"author":25938,"featured_media":44335,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[204,77],"tags":[],"_links":{"self":[{"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/posts\/44330"}],"collection":[{"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/users\/25938"}],"replies":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/comments?post=44330"}],"version-history":[{"count":3,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/posts\/44330\/revisions"}],"predecessor-version":[{"id":44347,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/posts\/44330\/revisions\/44347"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/media\/44335"}],"wp:attachment":[{"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/media?parent=44330"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/categories?post=44330"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.bu.edu\/cise\/wp-json\/wp\/v2\/tags?post=44330"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}