Aug 2026· International Conference on Advanced Mechatronic Systems· pp. 13-18· 0 citations· 21 references
Abstract
Generalist Controllers (GCs) have recently emerged as a promising paradigm for controlling heterogeneous dynamical systems using a single learned policy. However, existing GC approaches typically require prior knowledge of the active system, which limits autonomous deployment and prevents fully self-contained operation. This paper proposes a soft supervisory control framework that employs attention-based trajectory analysis to automatically infer the system currently under control and provide this information to a switching controller. The proposed architecture processes histories of closed-loop states and control actions using state and action encoders, dual-timescale Long Short-Term Memory (LSTM) networks, and a multi-head attention mechanism to extract discriminative temporal features. A supervisory decision layer then estimates the most likely system identity together with a confidence measure which enables automatic controller conditioning. The framework is trained using expert trajectories generated by Linear Quadratic Integral (LQI) controllers and evaluated on five heterogeneous systems spanning stable, unstable, non-minimum-phase, linear, and nonlinear dynamics, which is unique. The obtained results show an overall supervisory classification accuracy of 98.93%, with precision, recall, and F1-scores exceeding 0.97 for all systems. Furthermore, the proposed approach remains effective under measurement noise, successfully identifies systems during transitions, and generalises to different reference signals not encountered during training. These results indicate the potential of attention-based soft supervisory mechanisms to enable autonomous operation in both GCs and classical switching controllers.
This work uses SHallow REcurrent Decoder networks-based Reduced Order Modeling (SHRED-ROM) to synthesize a real-time closed-loop controller for high-dimensional and parametric dynamics, relying solely on limited state sensor readings, alleviating the curse of dimensionality.
Matteo Tomasetto, Francesco Braghin, J. Kutz et al.· 0 citations
Experimental results in simulation and on real hardware demonstrate that the decentralized–supervised architecture can achieve comparable or improved aggregate tracking performance relative to a centralized policy, while preserving decentralized proposal generation and enabling execution-time supervisory coordination under partial observability.
A. Bozzi, Matteo Aicardi, E. Zero et al.· IEEE Access· 0 citations
Repetitive maritime operations can be effectively learned using the Imitation Learning (IL) paradigm, which transfers human expertise directly to Unmanned Surface Vehicle (USV) control systems. Dynamical Systems (DS) are widely used to model non-linear human demonstrations while offering inherent stability guarantees. However, real-world execution under persistent marine perturbations reveals a critical trade-off: standard DS-based IL approaches prioritize global target convergence at the expense of localized trajectory reproduction fidelity. To address this limitation, we present a hybrid learning-control architecture that integrates a DS-based IL reference generator with a neuroadaptive controller. Our approach introduces a control action that drives the USV back to the demonstrated path following exogenous disturbances, enabling dynamic human-like reactive alignment-termed behavioral tracking. The proposed methodology is validated using the Marine Systems Simulator (MSS) toolbox. Simulation results confirm that the framework generalizes complex maneuvering tasks while substantially improving trajectory tracking fidelity under disturbances compared to alternative control strategies.
Yeyson A. Becerra-Mora, José Ángel Acosta· 0 citations
Reinforcement learning (RL) enables the synthesis of control policies directly from data, making it highly appealing for complex cyber-physical systems (CPSs) and robotics. A persistent challenge, however, is ensuring strict, hard safety constraints during the active learning phase. In real-world physical systems, violating mechanical limits can cause irreversible damage, necessitating that exploration remains strictly within safe operational regions. We propose a generalized framework that combines the adaptive, high-performance nature of deep reinforcement learning (DRL) with the formal safety guarantees of model predictive control (MPC). Using a mathematical model of the system dynamics, offline MPC computations define a feasible state-action space, representing all safe combinations of system states and control inputs that guarantee constraint satisfaction. During training and deployment, the RL agent's instantaneous actions are projected onto this globally verified feasible set via a safety filter. We systematically evaluate our generalized approach on a non-linear 1-DoF laboratory testbed, demonstrating successful exploration and stable policy convergence on physical hardware.
George Schafer, Jakob Rehrl, Stefan Huber et al.· 0 citations
Simultaneous input-gain variation and large additive disturbance create a control problem in which a fixed observer or nominal controller may be unable to reproduce the performance of a regime-aware design. We study a training--deployment asymmetry: during simulation or commissioning, an expert controller is allowed to use the known gain and disturbance, whereas the deployed controller can use only the reference and measured states. Directly imitating expert actions is generally unsafe because the same instantaneous student observation may correspond to different privileged regimes and hence different expert actions. We propose a mechanism-guided transfer route rather than a new neural architecture. An exact sampled-data identity removes the additive disturbance from the expert law and reduces learning to a task-relevant inverse input gain inferred from causal state history. The latent target is reconstructed from expert actions and deployment-visible trajectories, so the true plant parameter is not required as a student label. A common-quadratic certificate is derived for the actual augmented sampled recursion, followed by explicit residual, coverage, switching, noise, and saturation qualifications. A parameter-regime scan shows that the nominal observer's error grows sharply as $a$ decreases and that the next gain above the best non-failing tuning diverges for every tested $a<1$. Direct action networks also fail in closed loop despite moderate offline error, whereas the structured student remains close to the privileged expert and reduces tracking RMSE by about 69\% relative to the tuned observer in unseen 60-s trials. The contribution is an interpretable design perspective for turning privileged multi-regime control knowledge into a deployable adaptive controller, together with conditions under which the transfer is meaningful.
Xitong Niu, Peifeng Hui, Zheyong Jiang et al.· 0 citations
Latent-conditioned adaptive policies can control robots across changing dynamics, but their learned latents remain internal representations of the policy rather than physical models that can be inspected, rolled out, or used by other control modules. This limits closed-loop analysis, diagnosis, and further improvement of a fixed policy. A direct mapping from latent to physical parameters is also under-specified, because multiple systems can induce similar closed-loop behavior. We therefore decode each operational latent into a distribution of quadrotor models using conditional flow matching. The decoded distribution enables two downstream uses without modifying the policy: online predictive tuning of a high-level controller around the fixed low-level policy, and robustness analysis under specified disturbances. Under perturbed actuator dynamics, decoded-model predictive tuning reduces position tracking RMSE by $23\%$ and heading RMSE by $45\%$ relative to fixed gains. Under Gaussian force disturbances, decoded-model ensembles closely predict the lateral tracking-error evolution. Together, these results show that control latents can be converted into physical model ensembles for tuning, robustness analysis, and diagnosis of frozen adaptive policies.
Dingqi Zhang, Ruiqi Zhang, Mark W. Mueller· 0 citations