QT100Traffic Management Robot
The all-round auxiliary officer on city streets
Talks in 28 languages and signals all eight national-standard traffic gestures, directing intersections, correcting violations and patrolling on its own.
Learn more →The all-round auxiliary officer on city streets
Talks in 28 languages and signals all eight national-standard traffic gestures, directing intersections, correcting violations and patrolling on its own.
Learn more →
Human–machine collaboration that cuts cost and lifts output
Staffs the fire-control room around the clock: responds to alarms within 5 seconds, resets false alarms itself, and logs every action and shift change.
Learn more →
The campus safety officer
An automotive-grade chassis, 11 cameras and 3 LiDARs for full-coverage perception, and map-free software that deploys in hours, from closed campuses to open city roads.
Learn more →
Technology
A separated dual model, not one end-to-end network: a cerebrum for perception and decision-making works with a cerebellum for motion control, replacing the single end-to-end VLA that dominates the field.
The cerebrum handles scene understanding and task planning; the cerebellum generates motion through a diffusion model. Each corrects the other.
Task analysis happens at the algorithm layer, rather than executing a point-to-point command–action–feedback loop.
From enclosed indoor spaces to rain, snow, fog, night, mixed traffic and GPS-denied environments.
News



Roadmap
2026
2027
2028