DeepSoMo: Deformable Object Motion Estimation under Proprioceptive Partial Observability via Predictive Latent Dynamics
DeepSoMo: Deformable Object Motion Estimation under Proprioceptive Partial Observability via Predictive Latent Dynamics
This research introduces DeepSoMo, a framework that enforces predictive structure in latent dynamics through self-supervised constraints to estimate motion of deformable bodies from proprioceptive sensing. Evaluated in high-fidelity finite-element simulations, the approach demonstrates consistent improvements in trajectory estimation, orientation recovery, shape reconstruction, and contact inference compared to recurrent baselines.
The paper proposes DeepSoMo, a learning-based framework using self-supervised latent prediction constraints to estimate motion of deformable bodies in soft robotics under proprioceptive partial observability, evaluated via high-fidelity finite-element simulation; claims of consistent improvements over recurrent baselines are limited to this simulated setting.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
rvised latent prediction constraint. We evaluate the approach in a canonical deformable dynamics setting using high-fidelity finite-element simulation as a controlled source of ground truth. Results demonstrate consistent improvements in trajectory estimation, orientation recovery, shape reconstruction, and contact inference over architecture-matched recurrent baselines.
Open source S4
arXiv:2611.11242v1 Announce Type: new Abstract: Estimating the motion of deformable bodies in soft robotics from proprioceptive sensing is challenging due to partial observability, as internal observations provide only indirect access to global motion, contact, and shape evolution. In such settings, instantaneous observations are insufficient, and accurate estimation requires temporally consistent latent representati
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org