OpenAlex ·
SleepWalking: Privileged Representation Shaping for End-to-End Blind Locomotion in Legged Robots
- Published
- Source
- OpenAlex
- DOI
- 10.48550/arxiv.2608.30883
- Original
- Read it at the publisher
- Topics
Abstract
Partially observable locomotion requires a policy to act when task-relevant properties of the robot--environment state are not fully specified by instantaneous observations. Existing approaches often address this challenge by explicitly estimating missing physical variables or processing extended observation histories through structured architectures. We take a different view: partial observability is fundamentally an information-retention problem. The decisive question is not how task-relevant information enters the network, but whether the policy's internal state retains it. Guided by this perspective, we propose SleepWalking for Robot Locomotion (SWAQ), a one-stage end-to-end framework that uses next-step privileged physical reconstruction to shape what a recurrent history representation retains during policy learning, while the deployed actor uses only a direct history-to-action pathway. Under aligned training settings, SWAQ achieves a 15.0\% higher peak mean terrain level than DWAQ, the strongest non-exteroceptive baseline, while using 44.4\% fewer inference MACs per control step. Layerwise probes further show that information associated with the reconstructed physical variables remains linearly decodable through the policy head up to the layer preceding the action output. Complementary theoretical analysis relates privileged-variable recoverability to the achievable-return gap between history-based and privileged-information policy classes. These results suggest that semantic objectives can structure learning without requiring a corresponding architectural decomposition of the deployed controller.
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Artificial Intelligence
OpenMUSTANC (MUltiple Scattering Theory At Nanoplasmonic Cavities): A MATLAB toolbox for the simulation of Plasmonic Sphere Aggregates
Direct Optimization of a 3D Finite-Source Reflector via Neural-Network Parameterization
Emergence and suppression of phonon vortices in two-dimensional crystals: Interplay of lattice symmetry, heavy impurities, and shear
One Print, Many Moves: Monolithic Origami-inspired Folding Actuator for Composable Soft Multi-DoF Systems
Self-Diffusion of Water through Thermally Activated Membranes
Get the few papers worth your time in Artificial Intelligence every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.