Welcome to the Motion Lab

This is a live simulation of humans and robots sharing the same physical space.

You are the human.

Move around the room, cross paths with the robots, approach them, hesitate, change direction or step into their path. The robots will continue moving between stations and completing simple tasks around you.

The interesting part is not whether they can see you. It is whether they can understand how you are moving.

Motion Intelligence

Conventional autonomy can perceive where you are and plan a route around you.

Motion Intelligence adds another signal between those two layers: what your movement suggests you might do next.

SpatioTemporal compresses movement across space and time into Motion Tokens, a learned vocabulary of movement. Sequences of these tokens allow a machine to distinguish between patterns such as approaching, slowing, hesitating, turning away or committing to a path.

Press Space at any time to switch Motion Intelligence on or off.

Try creating the same encounter both ways and watch how the robots respond.

A small model with a big job

Everything you are seeing from Motion Intelligence comes from a model of around 30 MB, small enough to deploy directly onto a robot alongside its existing autonomy stack.

That is deliberate.

Humans do not consciously calculate trajectories every time someone crosses in front of us. Much of our ability to move safely around other people is fast, instinctive and almost invisible. We notice a hesitation, anticipate a step, adjust our path and keep moving.

Motion Intelligence is designed to provide robots with something similar: a small, fast layer concerned with movement, intent and what might happen next.

It does not replace perception or planning. It gives both better information at the moment it matters.

The result should be robots that move more naturally around us, anticipate our movements earlier and become easier for humans to understand and trust in shared spaces.

A small model helping robots read the room.

X-Ray Mode

Press X to see beneath the simulation.

X-Ray Mode exposes what the robots can observe and the motion information available to them as they navigate the room.

It is a view into the missing middle layer between perception and planning: not simply where you are, but how you are moving and what you might do next.

Controls

W  Move forward
S  Move backward
A  Turn left
D  Turn right

Q  Sidestep left
E  Sidestep right

Space  Motion Intelligence on / off
X  X-Ray Mode
C  Cycle camera
R  Reset simulation

Try this

Walk naturally at first. Then make yourself difficult to read.

Stop unexpectedly. Hesitate. Sidestep. Change direction. Cross a robot’s path, then change your mind.

You are creating the small ambiguities humans negotiate instinctively every day.

The Motion Lab lets you see what happens when robots begin to read them too.

Robots that read the room.