DEV KUMAR / APPLIED ROBOTICSLOCAL INFERENCE · NO API

MODEL → EXPERIMENT → EVIDENCE

UR7e Policy Lab.

A frozen robot policy. A classical controller. The same trajectory.

6 AXES · KINEMATIC SIMULATIONDrag to orbit · Scroll to zoom
TargetEnd-effector pathLive deterministic inference

REFERENCE EXPERIMENT

600 CPU episodes.
Both sides of the result.

100 paired trajectories × 2 controllers × 3 noise levels. No retraining. One frozen checkpoint.

Measured Python evaluation · no added noise
ControllerStrict completionTracking RMS*
Frozen SAC93 / 1005.21 mm
Fixed DLS90 / 1002.38 mm

*Mean per-episode RMS over acquired trajectories, including failures. DLS tracks more accurately in this experiment. The completion difference does not establish statistical superiority. All conditions and metadata ↗