Results.
Contact labels measured against sensors and tested in robot training.
Reliable contact labels.
Study details
REASSEMBLE: 83 recordings checked against wrist force-torque measurements. Contact precision uses a 95% interval.
Contact precision: 0.93 (95% interval 0.91 to 0.94). No-contact precision: 0.997 (lower bound 0.996). Coverage: 76% of windows, including half of contact windows and four fifths of no-contact windows; the rest are unlabeled.
At its shipped operating point, the impact channel matched 17 of 20 strikes on an axe clip within 500 ms: 187 ms median error and one unmatched record. Five non-impact clips returned one impact record.
Better robot training.
Study details
Pre-registered robomimic Can simulation, with the robot's own contact signal at execution.
Haptera contact labels raised success from 40.0% to 49.8% (+9.8 points; 95% interval 2.0 to 17.7). Shifted labels reached 36.1%. Each arm ran 30 seeds with 30 rollouts per seed: 900 rollouts per arm.
On the 63 pre-registered hard cases that failed all three baseline attempts, our labels recovered 39 (62%); the sealed simulator-sensor reference 40 (63%); the width-matched sensor reference trained alongside ours 44 (70%); shifted labels 23 (37%). Recovery showed no detectable difference between ours and either sensor reference. Correctly timed labels significantly outperformed shifted labels.
Contact at execution
The same contact channel with unlabeled training reached 32.4%, versus 42.3% for its control (interval minus 21.7 to plus 1.6 points).
Training with labels beat the channel-only policy by 18.6 points (interval plus 9.6 to plus 27.7).
Simulation, robomimic Can. Labeled policies used the robot's own contact signal at execution.
Fewer missed contact starts.
Study details
Adapting the video model on force-labeled recordings cut missed onsets by a third at the same false-alarm rate, on recordings it had never seen. Labels teach the model, not only the policy.
Difference -0.111 to -0.076 (95% interval), over 6 seeds.
From footage to labels.
Validation details
Before delivery, we validate your robot, camera, task and timeline against robot force, torque or joint signals, or a short capture with temporary instruments. Completed recordings are processed in batches.
Camera moves, task changes or measured drift trigger fresh validation.
Contact records
Haptera records contact and impacts from robot video, with timestamps, confidence and source evidence.
The record
Versioned event files on your recording timeline.
Contact state and impact times, with confidence and coverage.
Contact starts: automatic timestamps with joint-effort logs; reviewed intervals from video.
Recording IDs, model version, calibration and event evidence.
Recording inputs
30 fps video of the interaction.
Robot-state logs, if recorded. Joint efforts enable automatic contact-start timing at your chosen false-alarm rate within the measured held-out range.
Name the interaction before setup: gripper to object or tool to workpiece, for example.
- Your footageCamera, task and timeline.
- ValidateCheck against robot telemetry or a reference sensor.
- Get labelsTimestamped labels matched to your recordings.