6/8 After training, a deep net can map sensor readings to contact locations and force. The detection of every single point is very precise. See the picture of the Minerva, where we stimulate every individual point. Left: weak forces, right stronger forces.

7/8 The paper contains lots of details and analyses. Many thanks to Intelligent Systems and Science Robotics for their support.

8/8 Fun fact: my first two papers in Science and Nature journals came out on the same day! Isn't that crazy (and with the same Ph.D. student Huanbo Sun) x.com/GMartius/statu…

Striving to improve touch sensing in #robotics, scientists published a paper in nature Machine Intelligence, presenting a thumb-shaped sensor with a camera hidden inside. This #research invention significantly improves a #robot haptic perception is.mpg.de/en/news/finger… Georg Martius ✈️ICLR’26🇧🇷

A paper in Nature Machine Intelligence presents a sensor design that offers accurate force sensation across a three-dimensional surface while being robust, low-cost and easy to fabricate. It can be transferred to a wide variety of robot parts. go.nature.com/3vp9LON

It's so disappointing to see a ETH professor defended his inappropriate slide like this.I am also attaching his reply to the original post: linkedin.com/posts/liule-ya… This is NOT acceptable and we need a statement. ETH Zurich Joël Mesot Günther Dissertori Sarah Springman #MentalHealthMatters

.Georg Martius ✈️ICLR’26🇧🇷 and Huanbo Sun apply taxel value sensor isolines theory and #MachineLearning to create #tactile sensing devices capable of super-resolution sensing. fcld.ly/fb17pzo

Last week, @davaexplorer presented the 3D Knit BioSuit™—a multidisciplinary, collaborative work involving groups across MIT and industry—at #MARS2022. Learn more about this new prototype, which is the first BioSuit™ to integrate smart sensing! media.mit.edu/posts/dava-new…

Making a #robot with a sense of touch is key if #robots are to be used in medical or care settings ensuring they do not injure the patient. Prof Katherine Kuchenbecker from the Intelligent Systems is part of the team behind this haptic #robotic finger. nature.com/articles/s4225…

Our tactile all-around sensor "Insight" (x.com/GMartius/statu…) is featured in the Nature Outlook article by Marcus Woo: nature.com/articles/d4158… alongside other cool haptic technology.

We are ready to present our sensor finger here at Intelligent Systems at the science days. #Insight

Our Insight sensor is “children” tested to be very robust Intelligent Systems Science & Innovation Days. Nature Machine Intelligence Cyber Valley

Our Solo robot learning to perform a backflip from rough demonstrations [sites.google.com/view/corl2022-…] made it into Video Friday (scroll down a bit) spectrum.ieee.org/video-friday-p…

Congrats Georg!!!! It’s amazingly impressive to see how you achieve this step by step along my Ph.D study. Well deserved.

Congratulations! Georg Martius ✈️ICLR’26🇧🇷, who leads the Autonomous Learning #Research Group Intelligent Systems in #Tübingen, is awarded an European Research Council (ERC) Consolidator Grant to develop autonomous #robots that learn from simulated experience: is.mpg.de/de/news/georg-… #Science #MaxPlanck #AI Max Planck Society

Huanbo Sun Huanbo Sun from Max Planck Institute gave a wonderful talk on various tactile sensors (from theory to applications) at MIT CSL seminar. If you are interested in tactile sensors, recommend checking out our recording. Youtube link: youtu.be/uFYSyKS0CEA

We need good touch simulators to make truly capable robots. Here is an excellent step!

👏👏Congratulations to the 2024 Alan T. Waterman Award recipients for their groundbreaking contributions in biomedical engineering, genetics and robotics. Their innovative approaches and visionary leadership are instrumental in driving forward the frontier of scientific

Last night, we celebrated the three Alan T. Waterman awardees whose hard work and dedication to advancing the progress of science is set to bring unparalleled benefits to society. Congratulations to Muyinatu A. Lediju Bell, Katrina G. Claw and Rebecca Kramer-Bottiglio!

For tactile sensing, why shear forces are less accurate to detect than normal forces: Isoline-based theory for tactile sensing explains shear force detection at superresolution. (Open Access Link: rdcu.be/eCQQ6)