Spatial And Human Intelligence
For Physical AI.
Real spaces and real human behavior, captured to train robots for
factories, warehouses and stores.
Europe Needs Robots.
Robots Need Human and Spatial Data.
The labor gap
Operators cannot hire for warehouse, factory and overnight store work. They are asking for robots.
The data gap
Simulation lacks real human behavior. Ego video lacks exact world state. Robots learn slowly from either alone.
VLGE
Real spaces, real human behavior and exact world state in one dataset. Robots learn faster and deploy sooner.
Tens Of Thousands Of Proprietary 3D, Gaussian Splat And 360° Worlds,
From A Network Of Spatial Captures.
We Capture Spaces And Humans, And Deploy Robots.
01Capture the Space
Tens of thousands of proprietary 3D, Gaussian splat and 360° worlds, in the physical and the virtual world, captured at scale with objects.
02Capture the Humans
VLGE captures real-world human behavior, from everyday tasks to domain expertise.
03Fine-Tune, Train & Scale with AI Agents
VLGE’s agents learn from human behavior and produce new rollouts across tasks and conditions.
04Deploy Robots
Transfer expertise from human to robots
Spaces In The Physical And The Virtual World,
Captured At Scale, With Objects.
With Our Synthetic And Expertise Data,
Physical AI Is Solvable.
Physical AI data forces a tradeoff: natural human behavior or reliable world state.
Physics Simulation, Robot Teleop
Exact state, but scripted, policy-driven, or mediated through a robot.
Generated Video
Scalable visuals, but actions and state are inferred, not ground truth
Human Ego-Video and Physiology
Real humans, but occluded, one-shot, and disconnected from exact world state.
VLGE
Real human behavior + exact world state + repeatable as a distribution
Trained VLGE’s Agents On Our Real Human Behavior.
Human play trains the agent
run 1
run 2The agent scales the data
scales ×N · re-runs, keeps human-like data
HumanAgent
The agent generates the long tail
We Collect Real And Synthetic Human Behavior Data
To Train AI Agents, Test Their Performance On The Same Tasks,
And Prepare Them For Real-World Deployment In Robots.
We find exactly in which aspects the agent is failing, and improve it.
The Human Reference
Many people in the same scenario.
Trajectory, hesitation, decisions, attention.
The Agent
Run many times in the same scenario,
measured on the same signals.
Behavior Comparison
Compare how the agent behaves
with how people behave, signal by signal.
Where Your Clients Need Robots.
Manufacturing
Assembly, machine tending, quality checks.
Warehousing and logistics
Picking, packing, palletizing, inventory.
Automotive
Line-side handling, kitting, fine assembly.
Aerospace
Precision assembly and inspection in tight spaces.
Retail operations
Overnight inventory checks and store resets.
Retail and CPG design
Store digital twins. Packaging tested on real shopper behavior.
Two Ways To Work Together.
License
Capgemini licenses VLGE worlds and human behavior data to train and validate robots for client rollouts.
Capture
VLGE captures a client site and the people working in it, trains agents on it and benchmarks them before a robot ships.
and people
VLGE Unlocks The Next Era Of Physical AI.
- 3D, splat and 360° worlds
- Human behavior in simulation
- Real-world human data
- Digital twins
- The benchmark
- Robotics: VLA and policies
- Agentic AI
- World models
- Manufacturing
- Warehousing and logistics
- Automotive and aerospace
- Retail and CPG