Michael Wood’s Post

Fully autonomous survey grade mobile SLAM mapping+remapping is only possible with cm level accuracy: CSIRO + AutoMap for the win!

NVIDIA's Project GROOT is a game-changer for humanoid robots! This new foundational technology allows robots to learn from human demonstrations and complete tasks in virtual environments. This is a huge step forward for the field of robotics, and I can't wait to see what the future holds! #NVIDIA #ProjectGROOT #Robotics #AI #FutureofTechnology

As #Modex 2024 approaches its grand finale, today marks the fourth and last day of an incredible journey showcasing the future of industrial automation. It’s your final chance to visit #Milvus Robotics at Booth A11705 and see our Autonomous Mobile Robots (#AMRs) in action. These robots are not just changing the game in load carrying for factories and warehouses; they’re setting a new standard for efficiency and safety. Don’t miss out on live demonstrations, in-depth discussions with our experts, and the opportunity to explore how our AMRs can revolutionize your operations. Let’s make this last day count by pushing the boundaries of innovation together. See you there! #MilvusRobotics #Modex2024 #LastDay #AMRs #FutureOfAutomation

Logistics and warehouse automation can be sprung in Industry 4.0 using intelligent devices. Autonomous Mobile Robots can deliver this performance by many different ways! https://lnkd.in/gGM2AHTZ

Exciting news for the future of robotics! NVIDIA is developing a new AI model called GR00T, specifically designed to help humanoid robots learn new skills much faster. GR00T uses a combination of techniques, including watching videos and imitating human demonstrations, all powered by the powerful Jetson Thor robotics chips from NVIDIA. I believe that this breakthrough in robotics will change the game. Robots won't just be for factories or large warehouses anymore! We can soon expect to see them helping out in homes, offices, and even our neighborhoods. Let's just wait for it. #airevolution #robotics

"Mobile ALOHA" revolutionizes Robotics! Imitation learning from human demonstrations has taken a giant leap beyond table-top manipulation. Here is a system for bimanual mobile manipulation tasks requiring whole-body control. Meet Mobile ALOHA—a low-cost, whole-body teleoperation system that combines ALOHA with a mobile base. Witness it autonomously completing complex tasks like sauteing, serving, cabinet handling, elevator usage, and pan rinsing with just 50 demos for each task. Website: https://lnkd.in/dam85Dv3 Github: https://lnkd.in/dJmqiy-7 #Robotics #Innovation #AI 

ICRA 2024 - Kick-off. 🚀 👉🏻 We are ready to welcome the thousands of visitors who have come to beautiful #Yokohama to talk about the robotics of the future. 📍Come to visit us at booth IC140 🔎You can also see qbrobotics devices at the Franka Robotics booth (IC050). #ICRA2024 #qbrobotics #softrobotics #softhand2 #research #softhandresearch

Pleased to share that a paper from our group on "Action2Code: Transforming Video Demonstrations into Sequential Robotic Instructions" is being presented at 21st International Conference on Ubiquitous Robots at 11 AM ET https://lnkd.in/g74tuZHT . Please plan to attend if you are attending the conference. Summary: The time-consuming nature of programming robotic motion and manipulation to achieve specific tasks is a key reason that robots have generally been restricted to repetitive tasks with little variation. Developers need to manually write specific code for each task, making it challenging to adapt the code. We developed a generative network to program robots through demonstration,  bringing more agility to the process. We propose a network, Action2Code, that takes video demonstration as an input and translates it to robotic instructions.

Unveiling the Robo Revolution! While the term 'robotics' became mainstream in the 1940s, did you know that Leonardo da Vinci sketched the first humanoid robot in the 15th century? His vision paved the way for the incredible field we know today. Ready to dive into the realm of robotics? Explore, innovate, and craft the future at UIT!  Sign up for our Robotics Classes at www.unicconacademy.com!

🎯 Goal: Autonomous vision-based grasp of unknown objects - a sort of Holy Grail for modern roboticists 💡 Approach: "Grasp it like a Pro" 1️⃣ Decompose the point cloud into basic shapes - based on MVBB by Danica Kragic and team 2️⃣ Grasp one of the approachable basic shapes like a human - yes the human is the Pro Step 2️⃣ fuses a descriptive - yet not huge - set of data ➕ a proper AI algorithm 🔎 details and results in the paper https://lnkd.in/dSubbR64 ⏭ What's Next? - LLM integration - Mobile Grasp - ... darko-project.eu iliad-project.eu Special thanks to Alessandro Palleschi Chiara Gabellieri Do Won Park, and Franco Angelini and to Lucia Pallottino Antonio Bicchi