Residual Learning on a Physics Pivoting Controller
Faster Search, and Where the Quasi-Static Model Breaks
DOI:
https://doi.org/10.31224/8542Keywords:
Reinforcement learning, Gravitational Dynamics, motor current sensing, in-hand manipulation, Evolutionary systems, safety-critical systems, MuJoCoAbstract
One of three seeds stayed clearly worse with noise-model sensing; on validation objects, a release-rate cap added after training made every seed at least as accurate as the unguarded policy; and on heavier out-of-range objects the residual and model controllers degraded alike while pure RL degraded more.
Downloads
Download data is not yet available.
Downloads
Posted
2026-10-15 — Updated on 2026-10-21
Versions
- 2026-10-21 (2)
- 2026-10-15 (1)
License
Copyright (c) 2026 Clement Aidoo

This work is licensed under a Creative Commons Attribution 4.0 International License.
Version justification
Version 2 corrects a coding error in the rate-cap validation experiment (Section on the model-based guard). The cap improves rather than worsens accuracy. All other results are unchanged. See the correction note in the paper.