Learning contactfrom motor current

In free space, a camera shows where an object is and joint positions show the position of the arm in space. Contact-rich manipulation is harder because the state that matters may be physical, and no longer geometric. A gripper closes without securing an object. A peg reaches the hole and presses against its rim. Two nearly identical camera images precede a smooth insertion in one case and a jam in the other.

The SO-101 used in this study is fitted with Waveshare ST3215 servos. Their control table specifies a read-only motor-current register. Reading that register adds no sensor and no mechanical modification to the arm.

Visual occlusion

This loss of direct visual evidence near contact is a central observability problem in contact-rich imitation learning.

1. 1

   Images and joint positions provide strong information about appearance, configuration, and motion. They do not directly measure interaction effort.

2. 2

   Contact-aware policies add tactile, force, or torque measurements to vision and kinematics. These measurements reveal contact onset, load, deformation, or slip, and they usually require added hardware, calibration, or a robot that already gives reliable torque estimates.

3. 3

   The SO-101 used in this study is fitted with Waveshare ST3215 servos, whose control table specifies a read-only motor-current register. Reading that existing register adds no sensor or mechanical modification to the study arm.

4. 4

   Motor-side current is affected by pose, motion, friction, and the high-ratio transmission. It does not specify the direction or location of an external force.

Three questions make the investigation operational. The three claims are kept apart because they are easy to confuse. The signal may contain task-relevant information. A trained model may use that information. Using it may improve closed-loop behaviour.

Characterisation of the current signal

Before motor current can be added to a policy, it must first be shown to carry a repeatable and task-relevant signal.

Dataset

A fair comparison between a vision-based policy and a current-aware policy requires more than recording both modalities. The policies must learn from the same demonstrations, images, joint positions, and action targets, while the additional current input is introduced through a reproducible transformation.

Evaluation

Binary task success is the primary endpoint. A trial succeeds when the peg is fully seated and flush with the fixture surface before the timeout. Every other outcome is a failure. One operator resets and scores every trial.

Results

The joint-selective temporal ACT comparison has the largest favourable point estimate: success rises from 7/30 to 12/30, a 71.4% relative increase and a 16.7-percentage-point absolute gain. Residual-conditioned Diffusion Policy also favours the current-aware system, with a 33.3% relative increase from 9/30 to 12/30. Raw and normalised direct ACT instead have relative decreases of 57.9% and 11.8% in their original completed sessions.

These findings support a differentiated interpretation of the tested current-aware settings: direct fusion has adverse observed differences, while learned temporal and residual routes have favourable but imprecise estimates. The unchanged ACT-V checkpoint varies across sessions, so the within-session relative gains do not establish a ranking of current representations across days. No independent replication of the temporal ACT comparison was completed before the mechanical stop. The normalised comparison was repeated in a second session (10/20 versus 6/20), retained as a separate robustness check with the same negative direction.

Conclusion

Existing motor-current readings on the evaluated SO-101 contain contact-related information and can influence learned action predictions. The strongest favourable task result comes from joint-selective temporal ACT: success increases from 23.3% to 40.0%, a 71.4% relative improvement over its contemporaneous vision-and-position control. This observation supports further investigation of the current-aware setting, while its uncertainty leaves superiority unresolved.

Citation

@mastersthesis{ranpariya2026contact,
  author = {Jaydeepkumar Ranpariya},
  title  = {Learning Contact Representations from Zero-Cost
            Proprioceptive Effort Signals for Contact-Rich
            Visuomotor Imitation Learning},
  school = {Friedrich-Alexander-Universit{\"a}t Erlangen-N{\"u}rnberg},
  year   = {2026},
  note   = {In cooperation with Fraunhofer IIS},
  url    = {https://contact-from-current.pages.dev/}
}