Physics-based models bridge neuroscience and robotics! 🧠 Researchers from @CarnegieMellon University from NeuroMechatronics Lab started with a biomechanics model initially developed at West Virginia University to understand how coordinated muscle forces produce movement. The model examines forces and mechanics at the mechanical level, how neurons activate muscles, muscles generate forces, and forces move limbs. They used @MathWorks @MATLAB's Simscape Multibody to build and simulate these mechanical systems. Extended the approach to lower limbs. Investigated motor pathologies. Asked how biological neural networks organize themselves to perform sensorimotor computations, leading to work on Artificial Physics Engines. The human body is a mechanical system governed by physics, but movement is controlled through interacting layers. Neurons coordinate muscles. Muscles generate forces. Forces move limbs. To understand movement, you need to model the entire chain from neural activity to motion. Applied the same thinking to robotic control. Developed new neural network architectures for predicting and controlling motion in complex robotic systems. Physics-based simulations using Simscape Multibody train and evaluate these networks, providing the mechanical environments needed to test how neural controllers actually move real systems. EMG (electromyography) captures muscle activity signals. Decode those signals and you get intuitive control interfaces. A participant with spinal cord injury can use EMG from their arms to control a virtual hand, visualizing intended movement even when the limbs won't respond. The same approach extends to exoskeletal systems. EMG-based remote control. Movement intention becomes action through a device instead of paralyzed limbs. 🔗 That’s an awesome project, read more about it here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

5’s Warm-Up Drill Every rep is doing work on: 👣 Fast Footwork 💪 Core strength down-up ⚡ Plyometric 🧤 Clean, confident technique under control What I’m after is a controlled rehearsal of the biomechanics of the game. 🔁 Do two sets: one to your left, one to your right.

🇮🇹 🇺🇸 🤖 𝐈𝐭 𝐢𝐬 𝐭𝐢𝐦𝐞 𝐟𝐨𝐫 𝐦𝐞 𝐭𝐨 𝐰𝐫𝐚𝐩 𝐮𝐩 𝐥𝐚𝐬𝐭 𝐰𝐞𝐞𝐤’𝐬 𝐀𝐌𝐃 𝐀𝐝𝐯𝐚𝐧𝐜𝐢𝐧𝐠 𝐀𝐈 𝟐𝟎𝟐𝟔 𝐢𝐧 𝐒𝐚𝐧 𝐅𝐫𝐚𝐧𝐜𝐢𝐬𝐜𝐨. It was an intense and inspiring experience — and an honor to present #GENE01# together with @AMD CEO Dr. @LisaSu and South Korea Deputy Prime Minister and Minister Kyunghoon Bae 🇰🇷 ⚛️ At #CES26# in January, we joined Dr. Lisa Su on the AMD keynote opening cerimony as we unveiled our vision and the #GENE01# concept design. 🖐️ Six months later, that vision became physical. At #AdvancingAI#, we presented a fully functional humanoid platform that walks, interacts, and perceives the world through its full-body multimodal skin. During the presentation, we showcased what we believe is one of the key assets for human-centric Physical AI: a sensorized skin capable of perceiving touch and proximity across the robot’s body, enabling safer and more natural interaction with people. 🤝 This moment reflects the strategic relationship between @G_Bionics Generative Bionics and AMD, supporting key computing capabilities across our Physical AI platform. 🚀 But above all, it shows what becomes possible when deep Physical AI expertise, next-generation computing, and industrial execution come together. 🙏 A special thank you to Dr. Lisa Su for giving us the honor of presenting GENE.01, and to the entire #AMD# team for welcoming Generative Bionics at Advancing AI and supporting our first live U.S. appearance. Kudos to the entire Generative Bionics team for making this happen in just six months. And a special thanks to the team in San Francisco for the extraordinary work on site. 🚀 Our Physical AI is moving from concepts, models, and simulations into the real world very quickly, stay tuned. --- To know more. 📖 Press release: - 🇮🇹 - 🏴󠁧󠁢󠁥󠁮󠁧󠁿 📹 Full video: ℹ️ GENE.01 info 🤖 GENE.01 open source URDF 🔬 To learn about the science behind our #bodyascompute# computational approach for designing GENE.01 in such a short time, look at our Nature Machine Intelligence paper on the #Generative# Physical AI conditioned by #Biomechanics#: #GenerativeBionics# #GENE01# #PhysicalAI# #HumanoidRobotics# #Robotics# #AMD# #AdvancingAI# #IndustrialAI# #HumanRobotInteraction#

Grok Imagine image prompt: 1990s dark OVA anime style, retro cel animation, a young woman with blunt black bangs and pale blood-spattered face, head thrown back inside a glossy black biomechanical helmet, eyes wide, mouth slightly open, flooded in overwhelming emerald green light pouring through every seam and joint of her wet black armor, luminous green particles streaming around her face as if underwater, glowing green energy veins across the obsidian mech suit, ecstatic and terrified at once, girl and machine indistinguishable, intense bloom, sensory overload, harsh specular highlights, VHS film grain, painterly shading, Tsutomu Nihei and Yoshiaki Kawajiri inspired, transcendent, sublime.

See how Google Cloud AI is helping @TeamUSA freestyle skiers and snowboarders analyze their tricks like never before. From pose estimation to rotational velocity, we’re turning complex biomechanical data into actionable insights for the world's best athletes. #GoogleCloudNext#

China is entering an era in which everyone is a data collector launched China’s first Embodied AI Data Collection Community in Suqian,Jiangsu(the hometown of founder Liu Qiangdong) They plan to mobilize over 100,000 local citizens(especially stay-at-home moms )to wear a lightweight JoyEgoCam while doing daily chores. From wiping tables and folding clothes at home, to fruit picking, factory work, logistics, and elderly care… covering more than 100 real-life scenarios. These devices precisely capture key biomechanical data--such as upper-limb trajectories, force distribution, and hand-eye coordination,and feed it back to JD's data centers to train robotic models. OFC,Participants earn money while helping collect real-world data. The goal?Over 10 million hours in two years to solve robotics’ biggest problem--the severe shortage of authentic physical world data. Really interesting way to get everyday people involved in advancing AI humanoid robots.

The skeleton that slandered an entire species! August 1908. Three French brothers digging in a small limestone cave near La Chapelle-aux-Saints pulled a nearly complete human form out of the earth. Not scattered fragments. A body. Skull, spine, ribs, arms, legs. An old man laid in a shallow pit some 60,000 years earlier. He should of been a revelation. Instead he became a smear campaign. Marcellin Boule was one of the most powerful paleontologists in France and he spent years on those bones. When he published the reconstruction between 1911 and 1913, the world met a creature that could barely stand. Head thrust forward. Spine almost straight. Hips and knees bent. Big toe splayed like an ape’s. Boule called him uncouth, repellent, with only a “rudimentary psychic nature.” Newspapers and textbooks ran with it. Artists drew a hairy slouching brute shambling through Ice Age Europe with a club. That image stuck for half a century. Every cartoon caveman, every museum mural, every cheap insult about “Neanderthal thinking” traces back to this one skeleton. One old man. One scientists reconstruction. An entire species convicted of being primitive. Here is what Boule actually had on the table. The man was elderly by Ice Age standards, estimates now put him in his 50s or 60s, which is an extraodinary age for a hunter-gatherer. He had lost nearly every tooth. His jaw had resorbed. His left hip was a wreck of osteoarthritis. Neck and upper back locked up with degenerative joint disease. Healed rib fracture too. He almost certainly needed others to grind his food. He almost certainly needed help walking in those last years. Boule saw pathology and treated it as design. He saw an infirm elder and declared a species. The 1909 reconstruction that locked the stereotype in place showed a slouched figure next to an upright modern human, like the difference was evolutionary destiny instead of a bad hip. Popular illustrations went further. The “Man of Twenty Thousand Years Ago” got drawn as a shambling club-wielding ape-man coming out of a cave. That picture had almost nothing to do with how the living person actually moved. The skull itself is striking, massive brow, long low vault, huge nasal opening. Boule read those features as proof of a mind closer to an ape than to us. He was wrong about that too. The correction took decades. In 1957 William Straus and A.J.E. Cave went back over the bones and said the quiet part out loud. The hunched posture wasn’t typical Neanderthal anatomy. It was the posture of a man with severe deforming osteoarthritis. Take the disease away and the skeleton stands up. Later Erik Trinkaus argued Boule’s error wasn’t only medical. It was ideological. Boule didn’t want Neanderthals on the direct line to modern humans. A stooped dim-witted side branch was convenient. In 2019 Martin Haeusler and colleagues built a virtual reconstruction of the Old Man’s pelvis and spine. The numbers were blunt. His pelvic incidence predicted a lumbar curve of about 52 degrees, right in the modern human range. Wear patterns on the vertebrae showed he had the same S-shaped spine we use to walk upright. A hunched straight-backed Neanderthal is, as Haeusler put it, biomechanicly absurd. The Old Man did not slouch because he was primitive but because he was old, in pain, and still alive. A person who cannot chew solid food does not survive winters in Ice Age France without help. Somebody with a destroyed hip and a fused neck does not keep up with a mobile hunting band unless the band slows down for him. The cave itself looks like a burial, rectangular pit, body placed with care. Whether every claimed Neanderthal grave holds up is still argued. This one is among the strongest. The man who got used to prove Neanderthals were barely human is one of the best pieces of evidence that they were human, in the only sense that actually matters. They kept thier wounded. They fed their elderly. They put a body in the ground.