arXiv subjects

Samuel A. Moore

Publications and source records attributed to Samuel A. Moore.

7 recordsLinked to original sources

arXiv2026

Online Sim-to-Real Adaptation via Closed-Loop System Modeling

Sim-to-real transfer has made substantial progress, but can still produce controllers that remain stable and functional on hardware while suffering from degraded tracking accuracy due to residual dynamics mismatch. Correcting these errors typically requires identifying the underlying system dynamics, adapting the control policy, or returning to simulation for additional training and finetuning, all of which can require substantial data and computation. We propose OSRAM (Online Sim-to-Real Adaptation via Closed-Loop System Modeling), a framework that instead adapts the reference commands provided to an existing controller. OSRAM treats the deployed robot and its policy as a unified closed-loop dynamical system and learns its task-level command-response behavior directly from tracking observations. A closed-loop dynamics model is meta-trained across randomized dynamics in simulation and rapidly finetuned after deployment using limited real-world interaction. The adapted model is then used to optimize future reference commands while leaving the underlying control policy unchanged. We evaluate OSRAM on bipedal velocity tracking and loco-manipulation in simulation and on hardware. Results show that closed-loop modeling improves prediction and tracking accuracy under unseen dynamics, while online reference adaptation reduces residual sim-to-real tracking errors across different control objectives and hardware configurations. These results demonstrate that adapting the behavior of the robot-policy closed loop provides a practical alternative to finetuning the policy or identifying the full physical dynamics for sim-to-real transfer. More information can be found at http://generalroboticslab.com/OSRAM.

cs.RO↗

arXiv2026

The period map from commutative to noncommutative deformations

We study the period map from infinitesimal deformations of a qcqs derived scheme $X$ over a field $k$ to those of the associated $k$-linear $\infty$-category $\mathrm{QC}(X)$. We identify the corresponding map on tangent fibres with the dual HKR map $\mathrm{R}Γ(X, \mathbf{T}_{X/k})[1] \longrightarrow \mathrm{HH}^{\bullet}(X/k)[2]$ and study its injectivity on homotopy groups. As applications, we show liftability of qcqs schemes along square-zero extensions to be a derived invariant, at least when $\mathrm{char}(k) \ne 2$, and exhibit cases where the entire classical deformation functor of $X$ is a derived invariant; this partially answers a question of Lieblich.

math.AG↗

arXiv2026

Sym2Real: Symbolic Dynamics with Residual Learning for Data-Efficient Adaptive Control

We present Sym2Real, a fully data-driven framework for highly data-efficient adaptation of low-level controllers. Although symbolic regression is data-efficient, its role in real-world control has been limited due to its sensitivity to measurement noise, which corrupts the equations and leads to model degradation when fitted directly on real-world data. Sym2Real addresses this limitation by 1) learning first from low-fidelity simulation, where noise-free trajectories allow symbolic regression to identify the underlying dynamics, and 2) using a small amount of real-world data for targeted residual adaptation to bridge the sim-to-real gap. Using only about 10 trajectories, we achieve robust control of both a quadrotor and a racecar in the real world, without expert knowledge or simulation tuning. Through experimental validation on both platforms, we demonstrate consistent data-efficient adaptation across 6 out-of-distribution sim2sim scenarios and successful sim2real transfer across 5 real-world conditions. More information can be found at http://generalroboticslab.com/Sym2Real

cs.RO↗

arXiv2026

Learning Legged MPC with Smooth Neural Surrogates

Deep learning and model predictive control (MPC) can play complementary roles in legged robotics. However, integrating learned models with online planning remains challenging. When dynamics are learned with neural networks, three key difficulties arise: (1) stiff transitions from contact events may be inherited from the data; (2) additional non-physical local nonsmoothness can occur; and (3) training datasets can induce non-Gaussian model errors due to rapid state changes. We address (1) and (2) by introducing the smooth neural surrogate, a neural network with tunable smoothness designed to provide informative predictions and derivatives for trajectory optimization through contact. To address (3), we train these models using a heavy-tailed likelihood that better matches the empirical error distributions observed in legged-robot dynamics. Together, these design choices substantially improve the reliability, scalability, and generalizability of learned legged MPC. Across zero-shot locomotion tasks of increasing difficulty, smooth neural surrogates with robust learning yield consistent reductions in cumulative cost on simple, well-conditioned behaviors (typically 10-50%), while providing substantially larger gains in regimes where standard neural dynamics often fail outright. In these regimes, smoothing enables reliable execution (from 0/5 to 5/5 success) and produces about 2-50x lower cumulative cost, reflecting orders-of-magnitude absolute improvements in robustness rather than incremental performance gains.

cs.RO↗

arXiv2025

LAPP: Large Language Model Feedback for Preference-Driven Reinforcement Learning

We introduce Large Language Model-Assisted Preference Prediction (LAPP), a novel framework for robot learning that enables efficient, customizable, and expressive behavior acquisition with minimum human effort. Unlike prior approaches that rely heavily on reward engineering, human demonstrations, motion capture, or expensive pairwise preference labels, LAPP leverages large language models (LLMs) to automatically generate preference labels from raw state-action trajectories collected during reinforcement learning (RL). These labels are used to train an online preference predictor, which in turn guides the policy optimization process toward satisfying high-level behavioral specifications provided by humans. Our key technical contribution is the integration of LLMs into the RL feedback loop through trajectory-level preference prediction, enabling robots to acquire complex skills including subtle control over gait patterns and rhythmic timing. We evaluate LAPP on a diverse set of quadruped locomotion and dexterous manipulation tasks and show that it achieves efficient learning, higher final performance, faster adaptation, and precise control of high-level behaviors. Notably, LAPP enables robots to master highly dynamic and expressive tasks such as quadruped backflips, which remain out of reach for standard LLM-generated or handcrafted rewards. Our results highlight LAPP as a promising direction for scalable preference-driven robot learning.

cs.RO↗

arXiv2024

Automated Global Analysis of Experimental Dynamics through Low-Dimensional Linear Embeddings

Dynamical systems theory has long provided a foundation for understanding evolving phenomena across scientific domains. Yet, the application of this theory to complex real-world systems remains challenging due to issues in mathematical modeling, nonlinearity, and high dimensionality. In this work, we introduce a data-driven computational framework to derive low-dimensional linear models for nonlinear dynamical systems directly from raw experimental data. This framework enables global stability analysis through interpretable linear models that capture the underlying system structure. Our approach employs time-delay embedding, physics-informed deep autoencoders, and annealing-based regularization to identify novel low-dimensional coordinate representations, unlocking insights across a variety of simulated and previously unstudied experimental dynamical systems. These new coordinate representations enable accurate long-horizon predictions and automatic identification of intricate invariant sets while providing empirical stability guarantees. Our method offers a promising pathway to analyze complex dynamical behaviors across fields such as physics, climate science, and engineering, with broad implications for understanding nonlinear systems in the real world.

cs.LG↗

arXiv2022

A Model-Free Sampling Method for Estimating Basins of Attraction Using Hybrid Active Learning (HAL)

Understanding the basins of attraction (BoA) is often a paramount consideration for nonlinear systems. Most existing approaches to determining a high-resolution BoA require prior knowledge of the system's dynamical model (e.g., differential equation or point mapping for continuous systems, cell mapping for discrete systems, etc.), which allows derivation of approximate analytical solutions or parallel computing on a multi-core computer to find the BoA efficiently. However, these methods are typically impractical when the BoA must be determined experimentally or when the system's model is unknown. This paper introduces a model-free sampling method for BoA. The proposed method is based upon hybrid active learning (HAL) and is designed to find and label the "informative" samples, which efficiently determine the boundary of BoA. It consists of three primary parts: 1) additional sampling on trajectories (AST) to maximize the number of samples obtained from each simulation or experiment; 2) an active learning (AL) algorithm to exploit the local boundary of BoA; and 3) a density-based sampling (DBS) method to explore the global boundary of BoA. An example of estimating the BoA for a bistable nonlinear system is presented to show the high efficiency of our HAL sampling method.

cs.LG↗