aiqbench analysis
This research bridges classical disturbance-accommodating control with modern representation learning, introducing a statistically principled hard-EM procedure to model latent disturbance dynamics. By enforcing uniform contractivity in the learned representations, the method achieves predictive capability and provable convergence for adaptive tracking under complex, coupled disturbances. Experimental validation on robotic systems demonstrates improved robustness over fixed-decay baselines, marking a significant step toward more resilient autonomous control architectures.
Statistical Learning of Contractive Dynamical Representations for Composite Adaptive Control
arXiv:2609.35758v1 Announce Type: cross Abstract: We present a representation-learning framework for composite adaptive tracking control under dynamically coupled disturbances. The framework connects classical disturbance-accommodating control (DAC) to recent last-layer adaptive disturbance-rejection methods. Specifically, we introduce a statistically principled hard expectation-maximization (hard-EM) procedure, with a Kalman smoother in the hard E-step, to identify dynamical representations of disturbance whose latent evolution is uniformly contractive. The learned representation evolves a latent disturbance-excitation state from measured plant features and control inputs and decodes that state into the time-varying disturbance acting on the nominal plant, thereby extending prior "fixed-decay" last-layer adaptive methods to a learned, predictive DAC-style formulation. Combined with Bayesian filtering of the learned latent state, this representation yields a composite adaptive tracking controller with predictive capability and provable exponential convergence to a bounded neighborhood. We validate our approach experimentally on a slippery ground vehicle carrying a liquid-sloshing tank and a pendulum load, and we further assess its robustness on a system of coupled Duffing oscillators. Across both settings, the method achieves accurate disturbance prediction and improved overall tracking performance relative to fixed-decay representation-learning ablations, LTI disturbance-accommodating baselines, and model-based PD baselines.
Abstract:We present a representation-learning framework for composite adaptive tracking control under dynamically coupled disturbances. The framework connects classical disturbance-accommodating control (DAC) to recent last-layer adaptive disturbance-rejection methods. Specifically, we introduce a statistically principled hard expectation-maximization (hard-EM) procedure, with a Kalman smoother in the hard E-step, to identify dynamical representations of disturbance whose latent evolution is uniformly contractive. The learned representation evolves a latent disturbance-excitation state from measured plant features and control inputs and decodes that state into the time-varying disturbance acting on the nominal plant, thereby extending prior "fixed-decay" last-layer adaptive methods to a learned, predictive DAC-style formulation. Combined with Bayesian filtering of the learned latent state, this representation yields a composite adaptive tracking controller with predictive capability and provable exponential convergence to a bounded neighborhood. We validate our approach experimentally on a slippery ground vehicle carrying a liquid-sloshing tank and a pendulum load, and we further assess its robustness on a system of coupled Duffing oscillators. Across both settings, the method achieves accurate disturbance prediction and improved overall tracking performance relative to fixed-decay representation-learning ablations, LTI disturbance-accommodating baselines, and model-based PD baselines.
9 pages, including an additional one-page appendix in this arXiv version. Accepted to the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)
Systems and Control (eess.SY); Machine Learning (cs.LG); Robotics (cs.RO)
arXiv:2609.35758 [eess.SY]
(or arXiv:2609.35758v1 [eess.SY] for this version)
https://doi.org/10.48550/arXiv.2609.35758 arXiv-issued DOI via DataCite (pending registration)
Full text under the publisher's open licence · 232 words
