UKRI Trustworthy Autonomous Systems (TAS) Hub ’s Post

For their research that has revolutionized the science of robotics while dramatically improving the quality of human life, three world leaders in robotics science will share the coveted 2024 John Scott Award. The Scott Awards, presented annually in Philadelphia to honor the scientific legacy of Benjamin Franklin, will this year recognized Dr. Daniela Rus, Director of the Computer Science and Artificial Intelligence Laboratory at MIT; Dr. Takeo Kanade, Professor of Computer Science and Robotics at Carnegie Mellon University; and Dr. Vijay Kumar, Dean of the School of Engineering and Applied Science at Penn. https://lnkd.in/e4wmNRzi

My dissertation at the University of London dove into cognitive robotics, exploring how humans interact with humanized or empathetic robots on a deeper, more intuitive level. I designed and ran experiments, collected user feedback, and coded the results to see where the boundaries between “human” and “machine” really blur. The final paper landed in the university library, marking a milestone in how we understand AI-driven systems in everyday life. That project taught me the full research cycle, from forming a solid hypothesis to translating raw data into actionable insights. Have you ever combined hands-on experiments with academic rigor? I’d love to hear about your journey!

Prajval Kumar Murali's PhD work has now culminated in this fantastic publication in The International Journal of Robotics Research (IJRR) which is a tour de force in beautiful Kalman math and AI doing visuo-tactile perception for the robust pose estimation and reconstruction. All done in Munich at Mohsen Kaboli's RoboTac lab at BMW. Prajval Kumar Murali, Bernd Porr, Mohsen Kaboli: Shared visuo-tactile interactive perception for robust object pose estimation. IJRR. https://lnkd.in/dZVNG2R2

At this year's IROS, we'll run a workshop on "Robot Safety in Times of AI: Data, Decision, and Multimodal Interaction" covering topics from safe HRI and safety verification to safe decision-making in robotic systems. If you are interested in submitting a paper: we'll be happy to receive _your_ paper, the deadline (20th of September) is approaching, and we have prizes for the best three contributions! Munich Institute of Robotics and Machine Intelligence (MIRMI) at the Technical University of Munich | IROS 2024

My First Master Project at HSLU! 🤖 During this semester, I had the great opportunity to work under Prof. Björn Jensen in the AI Robotics Research Lab at Lucerne University of Applied Sciences and Arts. 🎯 The goal was to build a system capable of detecting human poses in 3D for future human collision avoidance packages, as well as recognizing hand positions to enable seamless tool handovers. The use case is to assist surgeons during operations and free scrub nurses from stressful and dangerous tasks, helping to combat the shortage of skilled workers in healthcare. Over the course of three months, I delved into countless papers on HPE models, tested different methods to gather depth from monocular cameras, and compared various approaches for their precision and robustness. During the second half of the project, I developed a proof of concept and built a ROS package that estimates the 3D Cartesian pose of my hand, recognizes gestures, and controls a UR3e cobot. The system is designed to be modular — all you need, besides the cobot, is an RGBD camera and an ArUco board. The cobot can be controlled via the stock controller (using a URSim Python wrapper) or through the MoveIt package. This was also my first time working with ROS2, which brought plenty of challenges, especially when it came to integrating MoveIt with the UR drivers. It was a demanding process, with many long nights in the lab, but ultimately, it was incredibly rewarding. 🙏A big thank you to Stefan Siegler and Alex Petrovic for their valuable input and ideas. I had a fulfilling time in the lab and am looking forward to my next project here! Take a look at a quick demo of the system in action! 🦾 #Robotics #AI #HPE #ROS2 #UR3e #HealthcareInnovation #HSLU

The subtle gamification in Slooh sure motivates students to stay engaged! I've got just one item left to capture in the Slooh 100 myself... the Bow-Tie Nebula (NGC 40)... but only 243 of the Slooh 1000. There's a lot to discover in the night sky! #astroedu #astronomy #astronomyeducation #astrophotography #careerreadiness #careertechnicaleducation #cte #edtech #ngss #overvieweffect #scienceeducation #spaceeconomy #spaceeducation #spaceexploration #spacephilosophy #spacescience #spaceworkforce2030 #STEMeducation #STEAMeducation #workforcedevelopment

Exciting times in Munich at the Conference on Robot Learning (CoRL) 2024! I had the privilege of presenting our latest research on indoor localization and mapping, exploiting the distortion of the magnetic field. Connecting with so many brilliant minds in the robotics community was amazing and inspiring. For those curious, indoors, the magnetic field is distorted by nearby ferromagnetic materials in the walls, the furniture, etc. These distortions can be mapped and used as a unique signature to recognize a place. One of the challenges is to build such a map model that is probabilistic and differentiable without compromising computational speed. This challenge is the focus of my paper. We tackled it by developing our sparse and local Gaussian process regression approximation. I also want to thank my supervisors and advisors at CEA-List and Mines Paris - PSL , Vincent GAY-BELLILE, Steve Bourgeois, Cyril JOLY, and Alexis Paljic, for their help in getting this idea published. Check out the teaser video and accepted paper to learn more. https://lnkd.in/eHSskvJG

What is the relationship between mathematics, elephant’s trunk, nano-robots, and proteins? Alireza Mohammadi, UM-Dearborn professor of robotics, explains this, and a lot more, in his introduction video for the National Science Foundation (NSF) Award CMMI-2153744:

ACENTAURI team is thrilled to announce the arrival of Professor Rafael Murrieta-Cid ! 🌟 Professor Murrieta, from Centro de Investigación en Matemáticas (CIMAT), is joining the team as visiting researcher to work on his project on robot motion planning with sensory feedback and learning. His research focus on improving sampling-based motion planners, which are known to be the most efficient techniques for finding collision-free trajectories for complex robotics systems in environments with obstacles. However, those methods assume a known and static map and they don't offer trajectory correction online. To overcome these limitations, Professor Murrieta is investigating a joint approach including visual servoing, also including sensor feedback to release the constraint of an already known environment. Additionally, he will explore the possibility of merging that approach with learning-based techniques. It is an honor to welcome Professor Murrieta and to share our knowledge reciprocally to push the boundary of robotics together. If you want to join as well the team or you need more information, please visit our website https://lnkd.in/eJmhTJca. #Research #Robotics #MotionPlanning #VisualServoing