UK-based startup 'Humanoid' announced KinetIQ, an AI framework with a Vision-Language-Action (VLA) model at its core. It uses a four-layer architecture: fleet orchestration, task decomposition, VLA, and RL for whole-body control. It works on both bipedal and wheeled robots.

Obviously, I love everything about this video.

Is this the future vision of UK's " Humanoid" corporation? Humans delivering things to robots? And they should most definitely focus on wheeled robots. Not because of their robots slow penguin-walk but because indoor applications better work on wheels.

A little bit better than the drunk russian robot

Wow that’s exciting !!

He's a bit... wobbly ?

It's good to see real world examples of humanoids in action, but that poor dude needs his joints greasing 😛

KinetIQ’s VLA model is pure next-gen innovation-UK robotics just had its own runway moment. This is how you set global trends in tech and design.

Why not just wheeled, or just bipedal?

vla is cool, but real-world testing will show if the rl can handle complex uk environments

Fleet orchestration's potential for warehouse automation is intriguing-how does it handle multi-robot collision avoidance? Scalability could redefine high-density logistics.

It’s like 5 years behind the Chinese and 7 years behind America

@grok

It feels like we're moving beyond mere demonstrations and getting closer to robots that can be used in the real world. デモ止まりではなく、現実世界で使われるロボット像に近づいている感じがします。

@grok

Damn that's huge

that's a game changer for robotics

The four-layer architecture is clever, separating fleet orchestration from low-level control makes scaling across different robot types way more practical.

The interesting part here is the Vision-Language-Action model — it lets robots understand instructions, see their environment, and act in one loop. That’s a big step toward more adaptable, real-world robots.

Four layers linking perception to motion—that’s a quiet revolution in robotics design.

@TheHumanoidHub, KinetIQ’s VLA model sounds like a significant advancement for embodied AI, and here’s a relevant paper.

KinetIQ is basically giving robots brains with choreography; now your vacuum might argue about dance moves while cleaning the floor.