Summary:

 

KinetIQ is an AI orchestration framework developed by Humanoid.ai (UK) for controlling entire fleets of humanoid and non‑humanoid robots. It provides a unified “physical AI” stack that manages robots across industrial, service, and home environments, coordinating everything from fleet‑level task assignment down to millisecond‑level joint control.

 

Editorial:

 

KinetIQ is one of the first serious attempts to build a general‑purpose control system for heterogeneous robot fleets. Instead of treating each robot as a standalone unit, KinetIQ treats the entire fleet as a distributed embodied system, where tasks, goals, and behaviors flow through a four‑layer cognitive architecture.  

 

This allows Humanoid.ai to run wheeled industrial robots, bipedal R&D humanoids, and future home robots under the same AI brain.  

 

The framework is built around cross‑timescale control, VLA‑based task execution, and reinforcement‑learning whole‑body control, making it one of the most advanced “physical AI” stacks in the Western robotics ecosystem.

 

Specifications:

 

Dimensions & Build

(Software system — no physical dimensions)  

- System Type: Multimodal robot‑control framework  

- Design Intent: Unified control of humanoids, wheeled robots, and domestic robots  

- Architecture: Four‑layer cross‑timescale cognitive stack  

- Developer: Humanoid.ai (United Kingdom)  

- Primary Platform: HMND‑01 humanoid robot  

 

Mobility

(Framework‑level capabilities — depends on robot hardware)  

- Locomotion Support:  

  - Bipedal walking  

  - Dynamic balance  

  - Whole‑body RL locomotion  

- Capabilities:  

  - Collision‑aware navigation  

  - Multi‑robot coordination  

  - Real‑time trajectory generation  

 

Note: No walking‑speed, incline, or gait‑stability metrics published (robot‑dependent).

 

Degrees of Freedom

(Framework‑level — varies by robot)  

- Total DOF: Not applicable  

- Supported Embodiments:  

  - Humanoids (e.g., HMND‑01)  

  - Wheeled industrial robots  

  - Future home robots  

 

Actuation

(Robot‑dependent)  

- Actuator Type: Not published  

- Capabilities:  

  - Millisecond‑level joint control  

  - Whole‑body reinforcement‑learning policies  

 

Note: No torque or joint‑speed data published.

 

Manipulation

- Capabilities:  

  - VLA‑driven skill execution  

  - Object handling  

  - Task‑level sequencing  

  - Multi‑step workflow execution  

 

Note: Payload, grip force, and hand DOF depend on the robot platform.

 

Perception & Sensors

KinetIQ fuses data from robot‑mounted sensors, including:  

- RGB cameras  

- Depth cameras  

- LiDAR  

- IMUs  

- Force/torque sensors  

- Tactile arrays (when available)  

 

Note: No camera resolution, FOV, IMU specs, or tactile‑sensor layout published.

 

Compute & AI

- AI System: Multimodal VLA + RL locomotion stack  

- Capabilities:  

  - Vision‑language‑action reasoning  

  - Task interpretation  

  - Skill‑level execution  

  - Fleet‑level orchestration  

  - Simulation‑to‑real transfer  

 

Note: No CPU/GPU/TOPS or memory specifications published.

 

Battery & Power

(Software system — no onboard power)  

- Power Source: Robot‑dependent  

- Runtime: Not applicable  

 

Connectivity

- Cloud‑connected orchestration  

- Local real‑time control loops  

- Industrial communication stack  

 

Note: No Wi‑Fi/Bluetooth/5G details published.

 

Applications

- Retail automation  

- Warehouse logistics  

- Manufacturing  

- Service robotics  

- Home robotics (future roadmap)  

- Multi‑robot fleet coordination  

 

Pricing

- Not published  

Bundled with Humanoid.ai enterprise deployments and robot platforms.

 

Humanoid.ai has NOT published:  

- Model sizes  

- Dataset details  

- Control‑loop frequency  

- API/SDK documentation  

- CPU/GPU/TOPS  

- Memory  

- Latency metrics  

- Safety certifications  

- Supported robot list beyond internal platforms  

 

Image: @VisionEffecttvTR

 

Company Source:

 

Humanoid.ai — Official Website

https://www.humanoid.ai

 

KinetIQ Announcement Page

https://www.humanoid.ai/kinetiq  

 

Key News & Coverage:

 

https://youtu.be/Y2DhzLPGdwY?si=QIkWqo6WfEZ0Oi8h