Residual Control for Fast Recovery from Dynamics Shifts

• URL: http://arxiv.org/abs/2603.07775v1
• Date: Sun, 08 Mar 2026 19:33:33 GMT
• Title: Residual Control for Fast Recovery from Dynamics Shifts
• Abstract summary: Unobserved dynamics shifts can experience substantial transient performance degradation in robotic systems.<n>We propose a stability-aligned residual control architecture in which a reinforcement learning policy trained under nominal dynamics remains fixed at deployment.<n>We show that the proposed method consistently reduces recovery time relative to frozen and online-adaptation baselines while maintaining near-nominal steady-state performance.
• Score: 1.9446811955039964
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Robotic systems operating in real-world environments inevitably encounter unobserved dynamics shifts during continuous execution, including changes in actuation, mass distribution, or contact conditions. When such shifts occur mid-episode, even locally stabilizing learned policies can experience substantial transient performance degradation. While input-to-state stability guarantees bounded state deviation, it does not ensure rapid restoration of task-level performance. We address inference-time recovery under frozen policy parameters by casting adaptation as constrained disturbance shaping around a nominal stabilizing controller. We propose a stability-aligned residual control architecture in which a reinforcement learning policy trained under nominal dynamics remains fixed at deployment, and adaptation occurs exclusively through a bounded additive residual channel. A Stability Alignment Gate (SAG) regulates corrective authority through magnitude constraints, directional coherence with the nominal action, performance-conditioned activation, and adaptive gain modulation. These mechanisms preserve the nominal closed-loop structure while enabling rapid compensation for unobserved dynamics shifts without retraining or privileged disturbance information. Across mid-episode perturbations including actuator degradation, mass variation, and contact changes, the proposed method consistently reduces recovery time relative to frozen and online-adaptation baselines while maintaining near-nominal steady-state performance. Recovery time is reduced by \textbf{87\%} on the Go1 quadruped, \textbf{48\%} on the Cassie biped, \textbf{30\%} on the H1 humanoid, and \textbf{20\%} on the Scout wheeled platform on average across evaluated conditions relative to a frozen SAC policy.

Related papers

• TTSD-FAR: Test-Time Self-Distillation with Fisher-Anchored Restoration for Missing-Modality Emotion Recognition in LVLMs [83.78708832901194]
  Large video-language models (LVLMs) have shown remarkable performance on multimodal tasks like multimodal emotion recognition (ER) in the wild.<n>However, real-world deployment remains challenging: modalities may be missing or noisy at test time.<n>We propose a solution in which a frozen teacher guides an adaptive low-rank student via self-distillation.<n>Experiments on MELD, DFEW, and BAH under 0%-50% missing modalities show that this unified adaptation-restoration design consistently outperforms entropy-based adaptation, RAG, and perplexity-based generation.
  arXiv  Detail & Related papers  (2026-08-18T23:40:28Z)
• PhaseLoRA: Control-Regime-Conditioned Low-Rank Adaptation for Continuous-Action Vision-Language-Action Policies [74.07190331264488]
  We propose a lightweight LoRA parameterization that conditions adaptation at each action-chunk prediction step.<n>PhaseLoRA modulates the LoRA left factor in the action expert, allowing the effective low-rank update direction to vary over time.<n>Results establish within-trajectory conditioning as an effective lightweight PEFT axis for continuous-action VLA policies.
  arXiv  Detail & Related papers  (2026-08-15T15:32:08Z)
• Score-Based Stabilization for Time-Dependent Problems [5.288977149942441]
  We propose a score-based stabilization framework for numerical simulation of partial differential equations.<n>We show that the stabilization operator acts as a contraction toward this manifold, yielding a correction mechanism with basin-conditional stability.
  arXiv  Detail & Related papers  (2026-07-27T22:26:54Z)
• ReactiveBFM: Reactive Closed-Loop Motion Planning Towards Universal Humanoid Whole-Body Control [54.267632784131415]
  ReactiveBFM is a real-time-loop planning-control framework for humanoids.<n>It mitigates fatal cumulative exposure bias via a scheduled prefix sampling curriculum.<n>It demonstrates unprecedented physical agility across a vast repertoire of text-conditioned closed-loop motions.
  arXiv  Detail & Related papers  (2026-06-29T14:27:27Z)
• Stabilizing Temporal Inference Dynamics for Online Surgical Phase Recognition [42.76777398447746]
  Online Surgical Phase Recognition (SPR) models can reach high frame-wise accuracy, yet their predictions often lack temporal stability.<n>We show that this instability is not random noise but arises from two mechanisms: early misclassifications corrupt temporal feature states and propagate forward to form error cascades.<n>We propose a unified Train-Inference-Evaluation framework that explicitly stabilizes temporal inference dynamics using model-agnostic, plug-and-play components.
  arXiv  Detail & Related papers  (2026-05-11T16:56:15Z)
• Diffusion Controller: Framework, Algorithms and Parameterization [54.82539154511621]
  We present a unified control-theoretic view that casts reverse diffusion sampling as state-only control within (generalized) linearly-solvable Markov Decision Processes.<n>Under this framework, control acts by reweighting the pretrained reverse-time transition kernels, balancing terminal objectives against an $f$-divergence cost.<n>Experiments on Stable Diffusion v1.4 show consistent gains in preference-alignment win rates and improved quality-efficiency trade-offs.
  arXiv  Detail & Related papers  (2026-03-07T01:49:59Z)
• When Sensors Fail: Temporal Sequence Models for Robust PPO under Sensor Drift [64.37959940809633]
  We study robustness of Proximal Policy Optimization (PPO) under temporally persistent sensor failures.<n>We show Transformer-based sequence policies substantially outperform, RNN, and SSMs in robustness, maintaining high returns even when large fractions of sensors are unavailable.
  arXiv  Detail & Related papers  (2026-03-04T22:21:54Z)
• Modeling and Controlling Deployment Reliability under Temporal Distribution Shift [1.4323566945483497]
  We propose a deployment-centric framework that treats reliability as a dynamic state composed of discrimination and calibration.<n>We show that selective, drift-triggered interventions can achieve smoother reliability trajectories than continuous rolling retraining.
  arXiv  Detail & Related papers  (2026-03-01T17:18:44Z)
• Cerebellar-Inspired Residual Control for Fault Recovery: From Inference-Time Adaptation to Structural Consolidation [2.0421986354783437]
  We introduce an inference-time, cerebellar-inspired residual control framework that augments a frozen reinforcement learning policy with online corrective actions.<n>The framework instantiates core cerebellar principles, including high-dimensional pattern separation via fixed feature expansion.<n>Experiments on MuJoCo benchmarks show improvements of up to $+66%$ on textttHalfCheetah-v5 and $+53%$ on textttHumanoid-v5 under moderate faults.
  arXiv  Detail & Related papers  (2026-02-06T22:16:00Z)
• AdaptNC: Adaptive Nonconformity Scores for Uncertainty-Aware Autonomous Systems in Dynamic Environments [7.201566646241765]
  Conformal Prediction methods maintain target coverage by adaptively scaling the conformal threshold.<n>We show that this fixed geometry leads to highly conservative, volume-inefficient prediction regions when environments undergo structural shifts.<n>We propose textbfAdaptNC, a framework for the joint online adaptation of both the nonconformity score parameters and the conformal threshold.
  arXiv  Detail & Related papers  (2026-02-02T04:41:35Z)
• Avoiding Premature Collapse: Adaptive Annealing for Entropy-Regularized Structural Inference [1.7523718031184992]
  We identify a fundamental mechanism for this failure: textbfPremature Mode Collapse.<n>We propose textbfEfficient Piecewise Hybrid Adaptive Stability Control (EPH-ASC), an adaptive scheduling algorithm that monitors the stability of the inference process.
  arXiv  Detail & Related papers  (2026-01-30T14:47:18Z)
• Automatic Stability and Recovery for Neural Network Training [1.9544213396776273]
  Training modern neural networks is increasingly fragile, with rare but severe destabilizing updates often causing irreversible divergence or silent degradation.<n>Existing optimization methods rely on preventive mechanisms embedded within stability probes, offering limited ability to detect and recover from instability.<n>We introduce a supervisory framework that treats optimization as a controlled runtime process.
  arXiv  Detail & Related papers  (2026-01-24T15:14:54Z)
• GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping [63.33669214116784]
  GRPO-Guard is a simple yet effective enhancement to existing GRPO frameworks.<n>It restores a balanced and step-consistent importance ratio, ensuring that PPO clipping properly constrains harmful updates.<n>It substantially mitigates implicit over-optimization without relying on heavy KL regularization.
  arXiv  Detail & Related papers  (2025-10-25T14:51:17Z)
• KCRL: Krasovskii-Constrained Reinforcement Learning with Guaranteed Stability in Nonlinear Dynamical Systems [66.9461097311667]
  We propose a model-based reinforcement learning framework with formal stability guarantees. The proposed method learns the system dynamics up to a confidence interval using feature representation. We show that KCRL is guaranteed to learn a stabilizing policy in a finite number of interactions with the underlying unknown system.
  arXiv  Detail & Related papers  (2022-06-03T17:27:04Z)
• On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control [47.71156648737803]
  Reinforcement learning is a framework for interactive decision-making with incentives sequentially revealed across time without a system dynamics model. We characterize a defined defined chain, identifying that policies associated with Levy Processes of a tail index yield to wider peaks.
  arXiv  Detail & Related papers  (2021-06-15T20:12:44Z)
• Stabilization via feedback switching for quantum stochastic dynamics [0.0]
  We propose a new method for pure-state and subspace preparation in quantum systems, which employs the output of a continuous measurement process and switching dissipative control to improve speed. We prove that the proposed closed-loop strategy makes the desired target globallyally stable both in mean and almost surely, and we show it compares favorably against a time-based and a state-based switching control law.
  arXiv  Detail & Related papers  (2020-12-16T02:54:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.