Planning to Learn: Real-Time Robust Adaptive Control on the GPU with Guaranteed Future Learning
Abstract
Robust adaptive control is vital for keeping robots safe while actively reducing uncertainty in unfamiliar settings. However, traditional methods are often conservative since they fail to account for guaranteed future parameter learning during planning and struggle to scale to high-dimensional systems. In this work, we present A-SLS, a learning-aware robust adaptive model predictive control (RAMPC) framework based on system-level synthesis. A-SLS forecasts future parameter uncertainty and preserves correlations between process disturbances, parameter error, and estimate updates to construct tight, guaranteed reachable sets that account for future information gain. We also develop a GPU-accelerated A-SLS solver to enable real-time performance. Across nonlinear robotic systems, A-SLS reduces reachable-tube conservativeness, maintains real-time performance, and scales to whole-body control for a quadruped with 61 states, 8 uncertain parameters, and 12 control inputs. We further demonstrate the approach on hardware, where a physical Crazyflie drone learns an uncertain dynamics parameter online to stabilize on the fly.
Related
- Non-Fragile Guaranteed Cost Control for Uncertain Fractional Delay Systems of Variable Order with Time-Varying Control Input Delays: A Possibly Stability Analysis
- Robust Iterative Learning Control for Multi-Agent Systems with Random Impulsive Disturbances
- Learning Optimal Robust-Pathwise Control of Elliptic Equations with Random Coefficients
- Model-Based Galerkin Lifting with Exact LTI Decomposition for Guaranteed ℋ∞ Output Feedback Control of Nonlinear Systems
- Sharp Critical Minimax Laws and No-Learning Thresholds in Continuous-Time Adaptive Control
BibTeX
@misc{srinivasan2026planning,
title = {{Planning to Learn: Real-Time Robust Adaptive Control on the GPU with Guaranteed Future Learning}},
author = {Anutam Srinivasan and Siddhant Panse and Glen Chou},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.20925},
eprint = {2610.20925},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.20925},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Planning to Learn: Real-Time Robust Adaptive Control on the GPU with Guaranteed Future Learning
AU - Anutam Srinivasan
AU - Siddhant Panse
AU - Glen Chou
PY - 2026
DA - 2026/10/14
PB - arXiv
DO - 10.48550/arxiv.2610.20925
UR - https://arxiv.org/abs/2610.20925
AB - Robust adaptive control is vital for keeping robots safe while actively reducing uncertainty in unfamiliar settings. However, traditional methods are often conservative since they fail to account for guaranteed future parameter learning during planning and struggle to scale to high-dimensional systems. In this work, we present A-SLS, a learning-aware robust adaptive model predictive control (RAMPC) framework based on system-level synthesis. A-SLS forecasts future parameter uncertainty and preserves correlations between process disturbances, parameter error, and estimate updates to construct tight, guaranteed reachable sets that account for future information gain. We also develop a GPU-accelerated A-SLS solver to enable real-time performance. Across nonlinear robotic systems, A-SLS reduces reachable-tube conservativeness, maintains real-time performance, and scales to whole-body control for a quadruped with 61 states, 8 uncertain parameters, and 12 control inputs. We further demonstrate the approach on hardware, where a physical Crazyflie drone learns an uncertain dynamics parameter online to stabilize on the fly.
ER -