[ai-buttons]
📰 Stay Informed with Truth Mafia!
💥 Subscribe to the Newsletter Today: TruthMafia.com/Free-Newsletter
🌟 Join Our Patriot Movements!
🤝 Connect with Patriots for FREE: PatriotsClub.com
🚔 Support Constitutional Sheriffs: Learn More at CSPOA.org
❤️ Support Truth Mafia by Supporting Our Sponsors
🚀 Reclaim Your Health: Visit iWantMyHealthBack.com
🛡️ Protect Against 5G & EMF Radiation: Learn More at BodyAlign.com
🔒 Secure Your Assets with Precious Metals: Get Your Free Kit at BestSilverGold.com
💡 Boost Your Business with AI: Start Now at MastermindWebinars.com
🔔 Follow Truth Mafia Everywhere
🎙️ Sovereign Radio: SovereignRadio.com/TruthMafia
🎥 Rumble: Rumble.com/c/TruthmafiaTV
📘 Facebook: Facebook.com/TruthMafiaPodcast
📸 Instagram: Instagram.com/TruthMafiaPodcast
✖️ X (formerly Twitter): X.com/Truth__Mafia
📩 Telegram: t.me/Truth_Mafia
🗣️ Truth Social: TruthSocial.com/@truth_mafia
🔔 TOMMY TRUTHFUL SOCIAL MEDIA
📸 Instagram: Instagram.com/TommyTruthfulTV
▶️ YouTube: YouTube.com/@TommyTruthfultv
✉️ Telegram: T.me/TommyTruthful
🔮 GEMATRIA FPC/NPC DECODE! $33 🔮
Find Your Source Code in the Simulation with a Gematria Decode. Are you a First Player Character in control of your destiny, or are you trapped in the Saturn-Moon Matrix? Discover your unique source code for just $33! 💵
Book our Gematria Decode VIA This Link Below: TruthMafia.com/Gematria-Decode
💯 BECOME A TRUTH MAFIA MADE MEMBER 💯
Made Members Receive Full Access To Our Exclusive Members-Only Content Created By Tommy Truthful ✴️
Click On The Following Link To Become A Made Member!: truthmafia.com/jointhemob
Summary
Transcript
And the first of these breakthroughs is implicit stereo vision, because unlike its monocular predecessor, Helix S1 employs a stereo vision backbone with two cameras integrated via multi-scale feature extraction networks and a cross-attention transformer. This enables precise depth perception and a comprehensive environmental understanding. And in demonstrations, it exhibited a 60% throughput increase over non-stereo baselines. And on top of this, it can even generalize to untrained objects, showcasing its human-like spatial awareness which is critical for the subsequent feature, breakthrough number two, its multi-scale visual representation, which builds on the stereo vision to fuse inputs from both cameras into a multi-scale stereo network, generating compact visual tokens for the transformer without the added computational costs.
And this tool focus on fine details like shipping label orientation, as well as the broader context, like its workspace layout, enhances the robot’s overall reliability. In fact, this resulted in a significant boost in the robot’s effective throughput, which is a metric comparing human-robot package handling speed, ensuring maximum precision without sacrificing situational awareness. But the third improvement is even more important, which is learn visual proprioception. And this addresses the challenge of scaling a single policy across multiple humanoid robots. And this is because hardware variations like sensor or joint differences traditionally hinder policy transfer, and manual calibration is impractical for fleets because of its high cost.
So the S1 self-calibration system instead uses onward visual inputs to estimate 60 end effector poses without external tools. And the magic here is that this enables one policy trained on one unit to deploy across many other units with minimal performance loss. And this leads us to the fourth feature, dubbed sport mode, which enhances speed via a test time technique. And this works with action chunks, which are movement sequences generated at 200 Hertz, and then they’re resampled to compress execution time. For example, you can take a one second work trajectory and reduce it to 0.8 seconds.
And this would yield a 20% speed increase without retraining. And tests achieve up to 50% speedups with robots handling packages efficiently while retaining dexterity for tasks like label orientation. Notably though, beyond this 50% threshold, this precision drop necessitating resets. And with all four of these advances combined, they collectively improve the robots efficiency, adaptability and scalability in these real world BMW environments. But there’s another contender in the race for taking factory jobs. Entering number two, as unitary robotics just unveiled their brand new Dex five dexterous hand in two different versions, both putting the previous Dex three robot hand to shame when it comes to dexterity.
In fact, the Dex five one and Dex five one p both share a lot of core features. But the Dex five one p seems to be a serious step up in terms of capability. Both of these hands have 20 degrees of freedom with 16 active joints and four passive ones across all five fingers, which gives them near human level flexibility and dexterity. Plus both versions of the Dex five are built for smooth back drive ability, meaning that they can adjust to external forces without feeling stiff. And both can even swing their forefingers laterally by 22 degrees to grip curved objects better.
But the big difference between the two hands comes in the form of tactile sensors, as the Dex five one has an unspecified number of touch points per hand, according to unitary’s website, but the Dex five one p does specify the number of touch points as a total of 94 tactile sensors, suggesting this version is even more sensitive. And this kind of precision is extremely important for tasks that are ultra delicate. And this brings us to another feature, which is the drive system where both of the hands use high power density holocaust motors and high precision encoders for smooth motion.
But the Dex five one p is array resolution is broken down as having two by five across its palm, two by three on each of its five finger pads, two by three on each of its five fingertips and two by three sensors and two by three on its four finger roots. And this layout gives it a detailed feedback across the entire hand letting the robot control every millimeter and adapt to whatever it’s holding. And the Dex five one has similar components, but the phrasing implies the Dex five one p might pack more punch or efficiency, though it’s not super clear how big the gap is.
Meanwhile, the micro gap joint design is also mentioned for both which keeps movements fluid and reduces snags. So no difference there. And as for costs, the current most expensive version of the unitary’s G one robot is going for about $69,900. And this includes the Dex three hand with three fingers. So the upgraded version with the Dex five will likely fall somewhere in the $70,000 range with all things considered. And finally, a brand new robot you haven’t seen, and it’s called the X man R one from Keenan robotics. And it packs some impressive features for working alongside humans and other robots.
In fact, the robot features two five finger dexterous hands, and a 180 degree twisting waste with a knack for understanding complex tasks via speaking. And it’s built to think ahead, planning for long horizon tasks to make decisions to keep things running smoothly. And the X one is loaded with tech to make it aware of its surroundings with multiple cameras and lidar sensors being tucked into its face, waist and hands, giving it a sharp sense of what’s going on around it. And in a demo, you can even see the robot tidying up a space after being told to by a human, and it picks up trash, throws it away, organizes almost with casual precision, and its dexterous hands can grip and move objects with human like touch, with its service posture mimicking how a person might stand or bend while helping out.
But what’s really cool is how this fits into Keenan’s bigger picture, because the X man R one hemes up with its existing robot fleet like the S 100 and C 30, creating what they call a new embodied intelligence world. And in the video, this is showcased by handling hotel tasks like grabbing luggage, picking up a coffee order and making deliveries alongside the X man W three. And it’s not working solo because it’s part of the crew. And this is where you can see the 180 degree waste mobility come in handy, letting the robot twist and reach without awkward shuffling, while its sensors keep it from bumping into guests or furniture.
And the robots got a brain for planning too. And it doesn’t just react because it figures out how to store things efficiently, like deciding where to stack trays or stash supplies, so nothing is a mess later. And with its human like behavior, it’s supposed to make the robot feel less like a clunky machine and more like an actual cooperative helper. And with Keenan’s X man R one, they’re showing off this robot as being ready to work in the real world as of today. And this brings us to a bonus clip, because here’s a direct comparison between real world automation happening right now in China and in Germany.
So tell us which is more impressive in the comments below, and we’ll see you next time. [tr:trw].