How does Physical AI work?
← Return to the report · page 7All explanations
Look at the warehouse illustration on page 7. A robot is carrying a package toward its destination. Then a person steps into its path. What happens next?
1. SEE / Notice something has changed
The robot uses cameras or other sensors to detect what is nearby. In this example, it detects that the space ahead is no longer clear. It needs to respond before continuing.
2. THINK / Choose a response
Its software checks what it is allowed to do. Should it wait for the person to pass? Is there another clear route? The word “think” in the diagram means choosing an action through software. It does not mean the robot thinks like a person.
3. DO / Carry out that choice
Suppose the robot stops and waits. Its software sends commands to the motors that move its wheels. The decision becomes a physical action: the robot stops moving.
4. ADJUST / Check again
The robot keeps checking its surroundings. When the person has passed and the route is clear, it can continue. If the route stays blocked, it may need to wait longer or request help. Its response depends on how it was designed.
Why the return arrow matters
The robot must keep checking as it moves. A route that was clear a moment ago may no longer be clear. The arrow returning to “See” shows this repeated checking.
The key idea: the machine changes its actions when conditions change. That does not necessarily mean it learns a new skill. Cameras and repeated checking also appear in older automation; modern AI can help interpret more varied situations.
This is an example to explain the illustration, not a report of a particular robot test. Different systems work differently. For technical background, see sources [1, 2] in the report’s sources.