AEON
Hexagon

AEON

AEON is Hexagon’s 165 cm, 60 kg humanoid that sprints at 8.6 km/h, carries 15 kg, and runs an on-board NVIDIA Jetson Orin with hot-swappable batteries—purpose-built for fast, safe industrial inspection and logistics.

Description

AEON, developed by Hexagon Robotics—a division of the Swedish-Swiss technology giant Hexagon AB—is a groundbreaking humanoid robot unveiled in June 2025 at the Hexagon LIVE Global event. Standing at 165 cm tall and weighing 60 kg, AEON is engineered for industrial environments like automotive, aerospace, manufacturing, logistics, and warehousing, addressing acute labor shortages through superior agility, versatility, and spatial awareness. Unlike legged humanoids prone to balance issues in dynamic factories, AEON integrates wheels into its feet for 'ready-to-roll' mobility, achieving a top sprint speed of 2.4 m/s (8.6 km/h) while weaving through tight aisles under 15 kg payloads. Its 34 degrees of freedom enable fluid upper-body gestures, full-body balance, and dexterous manipulation with 10-fingered hands, supporting tasks from precise part inspection to heavy pallet handling. Architecturally, AEON fuses Hexagon's proprietary precision measurement technologies with advanced mechatronics: a lightweight aluminum alloy frame clad in composite shell for durability in harsh settings. High-performance electric actuators from maxon provide precision and load stability, powering locomotion across varied terrains. The multimodal sensor suite—integrating RGB cameras, depth sensors, LiDAR, force/torque feedback, and Hexagon's scanning tech—delivers centimeter-accurate 3D spatial mapping, enabling reality capture for digital twins via HxDR platform and Reality Cloud Studio. This feeds into AI-driven mission control for real-time environmental reasoning and task prioritization. At its core, AEON leverages NVIDIA's full-stack robotics ecosystem. Onboard NVIDIA Jetson Orin handles edge AI for motion planning, spatial reasoning, and LLM inference, with upgrades to NVIDIA IGX Thor for functional safety in human-collaborative ops. Training occurs via NVIDIA Omniverse on OVX servers, using Isaac Sim for photorealistic simulation of navigation/manipulation and Isaac Lab for reinforcement learning policy refinement. Hexagon is fine-tuning NVIDIA Isaac GR00T N1.5 foundation models for enhanced reasoning/policies, supplemented by GR00T-Mimic for synthetic motion from human demos. Microsoft Azure scales cloud training. This simulation-first approach slashed locomotion skill development from months to 2-3 weeks, ensuring sim-to-real transfer. Deployment history began immediately post-launch with pilots at Schaeffler (German industrial manufacturer) and Pilatus Aircraft (Swiss aerospace), testing manipulation, machine tending, part inspection, and reality capture. Results show 40% faster QC scans and 99.9% accuracy in production environments. Hot-swappable lithium batteries (4-hour runtime each) enable 24/7 operation via auto-swap docks, minimizing downtime. Commercial rollout targets early 2026, with early Jetson Thor access accelerating perception enhancements. Safe for human proximity (cobotic design), Ethernet/Wi-Fi 6 connectivity uploads data to cloud. Running Linux, AEON exemplifies 'physical AI' bridging sensors/actuators with industrial needs, positioning Hexagon as a leader in autonomous humanoids.

Key Features

Exceptional Agility

Combines dexterity and wheeled locomotion for 2.4 m/s speeds, high-accuracy tasks in dynamic factory environments, stable under 15 kg payloads.

Versatile Task Execution

Handles manipulation, part inspection, reality capture, and teleoperation without add-ons; end-to-end AI training for multi-skill adaptability.

Spatial Awareness & Reasoning

Multimodal sensor fusion (RGB, depth, LiDAR, force) with AI mission control builds cm-accurate 3D maps, adapts to changes in real-time.

Autonomous Power Management

Hot-swappable batteries for 4-hour runtime each, auto-swap mechanism ensures continuous 24/7 operation without human intervention.

Simulation-First AI Development

NVIDIA Isaac Sim/Lab and Omniverse enable rapid skill training (2-3 weeks), GR00T fine-tuning for reasoning, Jetson Orin/Thor for edge compute.

Industrial Durability

Aluminum/composite structure, maxon actuators, human-safe design for harsh automotive/aerospace settings.

Specifications

AvailabilityPrototype
NationalitySwedish
Websitehttps://robotics.hexagon.com/product/
Degrees Of Freedom, Overall34
Height [Cm]165
Manipulation Performance2
Navigation Performance2
Max Speed (Km/H)8.6
Strength [Kg]15
Weight [Kg]60
Runtime Pr Charge (Hours)4
Safe With HumansYes
Cpu/GpuNVIDIA IGX, Nvidia Jetson
ConnectivityEthernet, WiFi
Operating SystemLinux
Llm IntegrationHexagon explores fine-tuned Isaac GR00T foundation models for reasoning and policy learning
Motor TechHigh-performance electric actuators engineered for precision & load stability (details TBA)
Main Structural MaterialAluminium, composite shell over alloy frame
Number Of Fingers10
Main MarketIndustrial inspection, manipulation, reality-capture & logistics
VerifiedNot verified
ManufacturerHexagon
Height Cm165
Weight Kg60
Dof Overall34
Dof HandsDexterous with 10 fingers
Max Speed Mps2.4
Max Speed Kmh8.6
Payload Short Term Kg15
Payload Constant Kg8
Battery TypeHot-swappable lithium-ion
Runtime Hours4
ProcessorsNVIDIA Jetson Orin (current), NVIDIA IGX Thor (upcoming for safety)
OsLinux
ConnectivityEthernet, Wi-Fi 6
SensorsMultimodal: RGB cameras, depth sensors, LiDAR, force/torque sensors, Hexagon precision measurement scanners
ActuatorsMaxon high-performance electric actuators
MaterialsAluminum alloy frame, composite shell
Ai ModelsNVIDIA Isaac GR00T N1.5 (fine-tuned), GR00T-Mimic for motion; Isaac Sim/Lab for training
SimulationNVIDIA Omniverse, Isaac platform

Curated Videos

Video 1
Video 2
Video 3
Video 4
Video 5

Frequently Asked Questions

What are AEON's key dimensions and payload capacity?

AEON measures 165 cm in height and weighs 60 kg, with a human-proportioned form. It supports short-term payloads of 15 kg and constant carry of 8 kg, enabling stable handling of inspection tools and pallet boxes in industrial settings.

How does AEON achieve continuous operation?

Featuring hot-swappable lithium batteries with 4-hour runtime per charge, AEON uses an innovative auto-swap mechanism. One battery powers operation while another charges, eliminating downtime for true 24/7 autonomy in factories.

What AI and compute platforms power AEON?

Onboard NVIDIA Jetson Orin manages real-time motion, perception, and LLM reasoning; future IGX Thor adds safety. Training via NVIDIA Isaac GR00T, Omniverse/Isaac Sim for simulation-to-real, Microsoft Azure for scalable cloud development.

What sensors does AEON use for perception?

Multimodal suite fuses RGB cameras, depth sensors, LiDAR, force/torque feedback, and Hexagon's precision scanners, creating centimeter-accurate 3D digital twins for navigation, inspection, and adaptive task execution.

What is AEON's top speed and mobility?

AEON sprints at 2.4 m/s (8.6 km/h) via wheeled feet integration, navigating tight aisles while balancing 34 DoF for dexterous upper-body tasks, outperforming legged bots in industrial speed and stability.

What real-world deployments has AEON achieved?

Pilots with Schaeffler and Pilatus since June 2025 demonstrate 40% faster QC scans and 99.9% accuracy in manipulation/inspection. Commercial rollout planned for early 2026 across manufacturing sectors.

Is AEON safe for human collaboration?

Yes, designed as cobotic with functional safety upgrades via IGX Thor, proximity awareness from sensors, and compliant actuators, ensuring safe operation alongside workers in live production environments.

×