Papers

We experimentally demonstrate enhanced sensitivity of an atom-based Rydberg radio frequency (RF) receiver integrated with a gradient refractive index (GRIN) Luneburg-type metamaterial lens. By analyzing the electromagnetically induced transparency (EIT) effect in Cesium vapor, we compare receiver performance with and without the GRIN lens under a 2.2~GHz and a 3.6~GHz far-field excitation. Our measurements reveal a significant amplification of the EIT window when the lens is introduced, consistent with the theoretical prediction that the local E-field enhancement at the vapor cell reduces the minimum detectable electric field and improves the microwave electric field measurement sensitivity of the Rydberg atom-based RF receiver over an ultrawide bandwidth of the lens. This experimental validation demonstrates the potential of metamaterial-enhanced quantum RF sensing for a wide range of applications, such as electromagnetic compatibility (EMC) testing, quantum radar, and wireless communication.

This work investigates the joint learning of pinching antenna (PA) positions and transmit beamforming for PA-aided integrated sensing and communication (ISAC) in the low-altitude wireless networks. By freely deploying antenna positions along waveguides, the pinching antenna system effectively mitigates the impact of path loss and thus enhances the capacities of sensing and communicating unmanned aerial vehicles (UAVs) that fly over a large range. We first model the problem of maximizing the sensing performance of multiple targets while satisfying the communication performance requirements of multiple users, where both the targets and users are UAVs. For mitigating in-waveguide attenuation and improving sensing performance, the segmented waveguide-enabled pinching antenna (SWAN) system is adopted. Furthermore, an alternative optimization (AO) algorithm for SWAN-based ISAC (SWISAC-AO) is developed, where the optimal structure of the transmit beamforming solution is derived. A graph neural network (GNN), termed SWISAC-GNN, is then proposed to jointly learn PA positions and transmit beamforming, with its alternative update procedure inspired by the SWISAC-AO algorithm. Numerical results show that the GNN achieves sensing performance comparable to or better than the AO algorithm while better satisfying communication requirements. Moreover, the SWISAC-GNN is with much lower implementation complexity, enabling real-time deployment.

This paper presents a combined sliding-mode control and subspace stabilization methodology for orbital stabilization of periodic trajectories in underactuated mechanical systems with one degree of underactuation. The approach starts with partial feedback linearization and stabilization. Then, transverse linearization along the reference orbit is computed, resulting in a periodic linear time-varying system with a stable subspace. Sliding-mode control drives trajectories toward this subspace. The proposed design avoids solving computationally intensive periodic LQR problems and improves robustness to matched disturbances. The methodology is validated through experiments on the Butterfly robot.

Considering a nonlinear system in Byrnes-Isidori form that is subject to unbounded perturbations, we apply Lyapunov redesign via feedback linearisation for trajectory tracking. Leveraging the ideas of tube-based geometric characterisation of the invariance properties of the closed loop, we generalise the classical stability criterion from the~literature from constant to nonconstant reference trajectories. The proposed analysis is tailored to the Lyapunov redesign and the tracking problem insofar as we incorporate the reference trajectory and the transient decrease of the tracking error enforced by the controller. In particular, we exploit that the Lyapunov function of the tracking error satisfies a differential inequality, thereby guaranteeing that the solution of the closed loop remains in a contracting tube along the reference trajectory.

Periodic operation often emerges as the economically optimal mode in industrial processes, particularly under varying economic or environmental conditions. This paper proposes a robust model predictive control (MPC) framework for uncertain systems modeled as polytopic linear differential inclusions (LDIs), where the dynamics evolve as convex combinations of finitely many affine control systems with additive disturbances. The robust control problem is reformulated as a convex optimization program by optimizing over configuration-constrained polytopic tubes and tracks a periodic trajectory that is optimal for a given economic criterion. Artificial variables embedded in the formulation ensure recursive feasibility and robust constraint satisfaction when the economic criterion is updated online, while guaranteeing convergence to the corresponding optimal periodic tube when the criterion remains constant. To improve computational efficiency, we introduce a quadratic over-approximation of the periodic cost under a Lipschitz continuity assumption, yielding a Quadratic Program (QP) formulation that preserves the above theoretical guarantees. The effectiveness and scalability of the approach are demonstrated on a benchmark example and a ball-plate system with eight states.

We propose a simple safety filter design for stochastic discrete-time systems based on piecewise affine probabilistic control barrier functions, providing an appealing balance between modeling flexibility and computational complexity. Exact evaluation of the safety filter consists of solving a mixed-integer quadratic program (MIQP) if the dynamics are control-affine (or a mixed-integer nonlinear program in general). We propose a heuristic search method that replaces this by a small number of small-scale quadratic programs (QPs), or nonlinear programs (NLPs) respectively. The proposed approach provides a flexible framework in which arbitrary (data-driven) quantile estimators can be used to bound the probability of safety violations. Through extensive numerical experiments, we demonstrate improvements in conservatism and computation time with respect to existing methods, and we illustrate the flexibility of the method for modeling complex safety sets. Supplementary material can be found at https://mathijssch.github.io/ecc26-supplementary/.

This paper introduces a direct comparative study of Physics-Informed Neural Networks (PINNs) and Long Short-Term Memory (LSTM) networks for adaptive steam temperature control in Heat Recovery Steam Generators (HRSGs), particularly under valve leakage faults. Maintaining precise steam temperature in HRSGs is critical for efficiency and safety, yet traditional control strategies struggle with nonlinear, fault-induced dynamics. Both architectures are designed to adaptively tune the gains of a PI-plus-feedforward control law in real-time. The LSTM controller, a purely data-driven approach, was trained offline on historical operational data, while the PINN controller integrates fundamental thermodynamic laws directly into its online learning process through a physics-based loss function. Their performance was evaluated using a model validated with data from a combined cycle power plant, under normal load changes and a challenging valve leakage fault scenario. Results demonstrate that while the LSTM controller offers significant improvement over conventional methods, its performance degrades under the unseen fault. The PINN controller consistently delivered superior robustness and performance, achieving a 54% reduction in integral absolute error compared to the LSTM under fault conditions. This study concludes that embedding physical knowledge into data-driven control is essential for developing reliable, fault-tolerant autonomous control systems in complex industrial applications.

Massive MIMO is a cornerstone of next-generation wireless communication, offering significant gains in capacity, reliability, and energy efficiency. However, to meet emerging demands such as high-frequency operation, wide bandwidths, co-existence, integrated sensing, and resilience to dynamic interference, future systems must exhibit both scalability and spectral agility. These requirements place increasing pressure on the underlying processing hardware to be both efficient and reconfigurable. This paper proposes a custom-designed spatial array architecture that serves as a reconfigurable, general-purpose core optimized for a class of wireless kernels that commonly arise in diverse communications and sensing tasks. The proposed spatial array is evaluated against specialized cores for each kernel using High-Level Synthesis (HLS). Both the reconfigurable and specialized designs are synthesized in a 32 nm process to assess latency, throughput, area, and power in realistic processes. The results identify conditions under which general-purpose systolic architectures can approach the efficiency of specialized cores, thereby paving the way toward more scalable and agile systems.

For a remote estimation system, we study age of incorrect information (AoII), which is a recently proposed semantic-aware freshness metric. In particular, we assume an information source observing a discrete-time finite-state Markov chain (DTMC) and employing push-based transmissions of status update packets towards the monitor which is tasked with remote estimation of the source. The source-to-monitor channel delay is assumed to have a general discrete-time phase-type (DPH) distribution, whereas the zero-delay reverse channel ensures that the source has perfect information on AoII and the remote estimate. A multi-threshold transmission policy is employed where packet transmissions are initiated when the AoII process exceeds a threshold which may be different for each estimation value. In this general setting, our goal is to minimize the weighted sum of time average of an arbitrary function of AoII and estimation, and transmission costs, by suitable choice of the thresholds. We formulate the problem as a semi-Markov decision process (SMDP) with the same state-space as the original DTMC to obtain the optimum multi-threshold policy whereas the parameters of the SMDP are obtained by using a novel stochastic tool called dual-regime absorbing Markov chain (DR-AMC), and its corresponding absorption time distribution named as dual-regime DPH (DR-DPH).

Recent works have demonstrated the viability of utilizing over-fitted implicit neural representations (INRs) as alternatives to autoencoder-based models for neural video compression. Among these INR-based video codecs, Neural Video Representation Compression (NVRC) was the first to adopt a fully end-to-end compression framework that compresses INRs, achieving state-of-the-art performance. Moreover, some recently proposed lightweight INRs have shown comparable performance to their baseline codecs with computational complexity lower than 10kMACs/pixel. In this work, we extend NVRC toward lightweight representations, and propose NVRC-Lite, which incorporates two key changes. Firstly, we integrated multi-scale feature grids into our lightweight neural representation, and the use of higher resolution grids significantly improves the performance of INRs at low complexity. Secondly, we address the issue that existing INRs typically leverage autoregressive models for entropy coding: these are effective but impractical due to their slow coding speed. In this work, we propose an octree-based context model for entropy coding high-dimensional feature grids, which accelerates the entropy coding module of the model. Our experimental results demonstrate that NVRC-Lite outperforms C3, one of the best lightweight INR-based video codecs, with up to 21.03% and 23.06% BD-rate savings when measured in PSNR and MS-SSIM, respectively, while achieving 8.4x encoding and 2.5x decoding speedup. The implementation of NVRC-Lite will be made available.

Vortex-Induced Vibrations (VIVs) of cylindrical structures present significant challenges in various engineering applications, including marine risers, tall buildings, and renewable energy systems. Hence, it is vital to control Vortex-Induced Vibrations of cylindrical structures. For this purpose, in this study a novel approach is introduced to VIV control, based on a model-based active control strategy integrated with a Neural Network (NN) in the presence of uncertainty modeling. The proposed method utilizes a closed-loop control system, where feedback from the system's dynamic state is used to generate adaptive control commands, enabling the system to respond to changing flow conditions and nonlinearities. Then, the controllability analysis is conducted to assess the efficiency of the control strategy in mitigating VIV. Two control approaches are implemented: simple learning and composite learning. Both strategies significantly enhance vibration suppression, achieving up to 99% reduction in vibrations despite uncertainties in the system. The results demonstrate the potential of the proposed method to enhance the efficiency, stability, and lifespan of structures subject to VIV.

Finite abstractions are discrete models of dynamical systems, such that the set of abstraction trajectories contains all system trajectories. There is a consensus that abstractions suffer from a scalability bottleneck: to obtain a sufficient “accuracy" (how closely the abstraction models the system), the required abstraction size often becomes very large, rendering computation infeasible. This is accentuated for complex, high-dimensional systems, due to the curse of dimensionality. Yet, after decades of research, there are no formal results on the size-accuracy tradeoff of abstractions. Here, we derive a statistical, quantitative theory of the size-accuracy tradeoff of abstractions of autonomous deterministic systems and uncover fundamental limits on their scalability, through rate-distortion theory—the information theory of lossy compression. Abstractions are viewed as encoder-decoder pairs, encoding trajectories of dynamical systems. Rate measures abstraction size, while distortion describes accuracy, defined as the spatial average deviation between abstract trajectories and system ones. We obtain a fundamental lower bound on the minimum achievable abstraction distortion, given the system dynamics and the abstraction size; and vice-versa a lower bound on the minimum required size, for given distortion. The bound depends on the complexity of the dynamics, through trajectory entropy and a constant depending on the trajectory manifold's geometry. We demonstrate its tightness on some dynamical systems. Finally, we showcase how this new theory enables constructing minimal abstractions, optimizing the size-accuracy tradeoff, through an example on a chaotic system.

Deep Image Prior (DIP) has recently emerged as a promising one-shot neural-network based image reconstruction method. However, DIP has seen limited application to 3D image reconstruction problems. In this work, we introduce Tada-DIP, a highly effective and fully 3D DIP method for solving 3D inverse problems. By combining input-adaptation and denoising regularization, Tada-DIP produces high-quality 3D reconstructions while avoiding the overfitting phenomenon that is common in DIP. Experiments on sparse-view X-ray computed tomography reconstruction validate the effectiveness of the proposed method, demonstrating that Tada-DIP produces much better reconstructions than training-data-free baselines and achieves reconstruction performance on par with a supervised network trained using a large dataset with fully-sampled volumes.

This paper presents a modular autonomous driving architecture for Formula Student Driverless competition vehicles operating in closed-circuit environments. The perception module employs YOLOv11 for real-time traffic cone detection, achieving 0.93 mAP@0.5 on the FSOCO dataset, combined with neural stereo depth estimation from a ZED 2i camera for 3D cone localization with sub-0.5 m median error at distances up to 7 m. State estimation fuses RTK-GNSS positioning and IMU measurements through an Extended Kalman Filter (EKF) based on a kinematic bicycle model, achieving centimeter-level localization accuracy with a 12 cm improvement over raw GNSS. Path planning computes the racing line via cubic spline interpolation on ordered track boundaries and assigns speed profiles constrained by curvature and vehicle dynamics. A regulated pure pursuit controller tracks the planned trajectory with a dynamic lookahead parameterized by speed error. The complete pipeline is implemented as a modular ROS 2 architecture on an NVIDIA Jetson Orin NX platform, with each subsystem deployed as independent nodes communicating through a dual-computer configuration. Experimental validation combines real-world sensor evaluation with simulation-based end-to-end testing, where realistic sensor error distributions are injected to assess system-level performance under representative conditions.

This paper investigates wireless systems aided by dual-polarized intelligent surfaces. We compare reconfigurable intelligent surface (RIS), which adjust their reflection matrices, with movable signals operating with fixed intelligent surface (FIS), which adjust the signal frequency while the surface properties remain fixed. For both RIS and FIS, we consider surfaces with a diagonal reflection matrix, named diagonal RIS/FIS, and surfaces with a reflection matrix not limited to being diagonal, named beyond-diagonal RIS/FIS. Movable signals with FIS always outperform RIS, achieving at least a fourfold gain. When transmitter and receiver polarizations differ, beyond-diagonal FIS further enhances performance.

Exoskeletons play a crucial role in assisting patients with varying mobility levels during rehabilitation. However, existing control strategies face challenges such as imprecise trajectory tracking, interaction torque oscillations, and limited adaptability to diverse patient conditions. To address these issues, this paper proposes an assist-as-needed (AAN) control algorithm that integrates a human-robot coupling dynamics model, a human torque-momentum observer (HTMO), and an adaptive parameter controller (APC). The algorithm first employs inverse dynamics to compute the joint torques required for the rehabilitation trajectory. The HTMO then estimates the torque exerted by the patient's joints and determines the torque error, which the exoskeleton compensates for via a spring-damper system, ultimately generating the target trajectory. Finally, the APC ensures adaptive assistive control. The proposed method is validated for its effectiveness in MATLAB/Simulink.

This paper presents a novel fault-tolerant control framework for steam temperature regulation in Heat Recovery Steam Generators (HRSGs) subject to actuator faults. Addressing the critical challenge of valve degradation in superheater spray attemperators, we propose a synergistic architecture comprising three components: (1) a Sliding Mode Observer (SMO) for estimation of unmeasured thermal states, (2) a Physics-Informed Neural Network (PINN) for estimating multiplicative actuator faults using physical laws as constraints, and (3) a one-sided Sliding Mode Controller (SMC) that adapts to the estimated faults while minimizing excessive actuation. The key innovation lies in the framework of closed-loop physics-awareness, where the PINN continuously informs both the observer and controller about fault severity while preserving thermodynamic consistency. Rigorous uniform ultimate boundedness (UUB) is established via Lyapunov analysis under practical assumptions. Validated on real HRSG operational data, the framework demonstrates effective fault adaptation, reduced temperature overshoot, and maintains steam temperature within 1{\deg}C of the setpoint under valve effectiveness loss. This work bridges control theory and physics-guided machine learning to deliver a practically deployable solution for power plant resilience, with extensions applicable to thermal systems subject to multiplicative faults.

The growing demand for robust, scalable wireless networks in the 5G-and-beyond era has led to the deployment of Unmanned Aerial Vehicles (UAVs) as mobile base stations to enhance coverage in dense urban and underserved rural areas. This paper presents a Multi-Agent Deep Reinforcement Learning (MADRL) framework that integrates Proximal Policy Optimization (MAPPO), Multi-Agent Deep Deterministic Policy Gradient (MADDPG), and Multi-Agent Deep Q-Networks (MADQN) to jointly optimize UAV positioning, resource allocation, Quality of Service (QoS), and energy efficiency through 5G network slicing. The framework adopts Centralized Training with Decentralized Execution (CTDE), enabling autonomous real-time decision-making while preserving global coordination. Users are prioritized into Premium (A), Silver (B), and Bronze (C) slices with distinct QoS requirements. Experiments in realistic urban and rural scenarios show that MAPPO achieves the best overall QoS-energy tradeoff, especially in interference-rich environments; MADDPG offers more precise continuous control and can attain slightly higher SINR in open rural settings at the cost of increased energy usage; and MADQN provides a computationally efficient baseline for discretized action spaces. These findings demonstrate that no single MARL algorithm is universally dominant; instead, algorithm suitability depends on environmental topology, user density, and service requirements. The proposed framework highlights the potential of MARL-driven UAV systems to enhance scalability, reliability, and differentiated QoS delivery in next-generation wireless networks.

Boolean matrix factorization (BMF) approximates a given binary input matrix as the product of two smaller binary factors. Unlike binary matrix factorization based on standard arithmetic, BMF employs the Boolean OR and AND operations for the matrix product, which improves interpretability and reduces the approximation error. It is also used in role mining and computer vision. In this paper, we first propose algorithms for BMF that perform alternating optimization (AO) of the factor matrices, where each subproblem is solved via integer programming (IP). We then design different approaches to further enhance AO-based algorithms by selecting an optimal subset of rank-one factors from multiple runs. To address the scalability limits of IP-based methods, we introduce new greedy and local-search heuristics. We also construct a new C++ data structure for Boolean vectors and matrices that is significantly faster than existing ones and is of independent interest, allowing our heuristics to scale to large datasets. We illustrate the performance of all our proposed methods and compare them with the state of the art on various real datasets, both with and without missing data, including applications in topic modeling and imaging.

Integrated sensing and communication (ISAC) is a promising paradigm for future wireless systems due to spectrum reuse, hardware sharing, and joint waveform design. In dynamic scenes, Doppler shifts degrade both sensing and communication, which is particularly critical for beam-sensitive orbital angular momentum (OAM) wavefronts. To address this, we propose a Doppler-robust ISAC framework, which first senses and then communicates. Specifically, in the sensing phase, multiple vortex modes are simultaneously transmitted via code-division mode-multiplexing (CDMM). To solve Doppler-induced inter-mode interference, we propose a velocity-consistency matching (VCM)-expectation maximization (EM) algorithm that jointly decodes the sensing matrix and estimates range, azimuth, elevation, and velocity for multiple moving targets. In the communication phase, the joint transmitter (Tx) beamforming and receiver (Rx) beam steering are configured from the estimated channel state information (CSI). We further quantify the sensing-communication allocation trade-off by evaluating how pilot length affects estimation accuracy, beam alignment, and spectral efficiency (SE). Simulation results show that the proposed VCM-EM and ISAC designs achieve higher sensing accuracy and communication SE than baseline schemes in dynamic scenarios.

We introduce fixed-initial-state constant-input dynamical system (FISCIDS) representations for nonlinear functions. A function argument serves as a constant input to an input-affine system with a fixed initial state, and the function value is the terminal output. Every nonsingular differentially algebraic function on an open set star-shaped at the origin admits an exact quadratic FISCIDS representation. At least one analytic non-differentially-algebraic function admits such a representation. These results establish exact representation guarantees for structured finite-dimensional parametric model classes.

This paper presents an auto-tuning framework for torque-based Nonlinear Model Predictive Control (nMPC), where the MPC serves as a real-time controller for optimal joint torque commands. The MPC parameters, including cost function weights and low-level controller gains, are optimized using high-dimensional Bayesian Optimization (BO) techniques, specifically Sparse Axis-Aligned Subspace (SAASBO) with a digital twin (DT) to achieve precise end-effector trajectory real-time tracking on an UR10e robot arm. The simulation model allows efficient exploration of the high-dimensional parameter space, and it ensures safe transfer to hardware. Our simulation results demonstrate significant improvements in tracking performance (+41.9%) and reduction in solve times (-2.5%) compared to manually-tuned parameters. Moreover, experimental validation on the real robot follows the trend (with a +25.8% improvement), emphasizing the importance of digital twin-enabled automated parameter optimization for robotic operations.

Optimization methods play a central role in signal processing, serving as the mathematical foundation for inference, estimation, and control. While classical iterative optimization algorithms provide interpretability and theoretical guarantees, they often rely on surrogate objectives, require careful hyperparameter tuning, and exhibit substantial computational latency. Conversely, machine learning (ML ) offers powerful data-driven modeling capabilities but lacks the structure, transparency, and efficiency needed for optimization-driven inference. Deep unfolding has recently emerged as a compelling framework that bridges these two paradigms by systematically transforming iterative optimization algorithms into structured, trainable ML architectures. This article provides a tutorial-style overview of deep unfolding, presenting a unified perspective of methodologies for converting optimization solvers into ML models and highlighting their conceptual, theoretical, and practical implications. We review the foundations of optimization for inference and for learning, introduce four representative design paradigms for deep unfolding, and discuss the distinctive training schemes that arise from their iterative nature. Furthermore, we survey recent theoretical advances that establish convergence and generalization guarantees for unfolded optimizers, and provide comparative qualitative and empirical studies illustrating their relative trade-offs in complexity, interpretability, and robustness.

The fundamental designs of wireless systems toward AI-Native 6G and beyond are driven by the need for ever-increasing demand of mobile data traffic, extreme spectral efficiency, and adaptability across diverse service scenarios. To overcome the limitations posed by feedback-based multiple-input and multiple-output (MIMO) transmission, we propose a novel frequency-domain Correlation-aware Feedback-free MIMO Transmission and Resource Allocation (CaFTRA) framework tailored for fully-decoupled radio access networks (FD-RAN) to meet the emerging requirements of AI-Native 6G and beyond. By leveraging artificial intelligence (AI), CaFTRA effectively eliminates real-time uplink feedback by predicting channel state information (CSI) based solely on user geolocation. We introduce a Learnable Queries-driven Transformer Network for CSI mapping from user geolocation, which utilizes multi-head attention and learnable query embeddings to accurately capture frequency-domain correlations among resource blocks (RBs), thereby significantly improving the precision of CSI prediction. Once base stations (BSs) adopt feedback-free transmission, their downlink transmission coverage can be significantly expanded due to the elimination of frequent uplink feedback. To enable efficient resource scheduling under such extensive-coverage scenarios, we apply a low-complexity many-to-one matching theory-based algorithm for efficient multi-BS association and multi-RB resource allocation, which is proven to converge to a stable matching within limited iterations. Simulation results demonstrate that CaFTRA achieves stable matching convergence and significant gains in spectral efficiency and user fairness compared to 5G, underscoring its potential value for 6G standardization efforts.

Policy gradient algorithms are widely used in reinforcement learning and belong to the class of approximate dynamic programming methods. This paper studies two key policy gradient algorithms, the Natural Policy Gradient and the Gauss-Newton Method, for solving the Linear Quadratic Regulator (LQR) problem in unknown stochastic linear systems. The main challenge lies in obtaining a consistent gradient estimate from noisy data due to errors-in-variables in linear regression. This issue is addressed by employing a primal-dual estimation procedure. Using this novel gradient estimation scheme, the paper establishes convergence guarantees with a sample complexity of order O(1/epsilon). Theoretical results are further supported by numerical experiments, which demonstrate the effectiveness of the proposed algorithms.

Infrared small target detection plays a crucial role in military reconnaissance and air defense systems. However,existing low-rank sparse based methods still face high computational complexity when dealing with low-contrast small targets and complex dynamic backgrounds mixed with target-like interference. To address this limitation, we reconstruct the data into a fourth-order tensor and propose a new infrared small target detection model based on bilateral tensor ring decomposition, called BTR-ISTD. The approach begins by constructing a four-dimensional infrared tensor from an image sequence, then utilizes BTR decomposition to effectively distinguish weak spatial correlations from strong temporal-patch correlations while simultaneously capturing interactions between these two components. This model is efficiently solved under the proximal alternating minimization (PAM) framework. Experimental results demonstrate that the proposed approach outperforms several state-of-the-art methods in terms of detection accuracy, background suppression capability, and computational speed.

Reinforcement learning (RL) offers transformative potential for robotic control in space. We present the first on-orbit demonstration of RL-based autonomous control of a free-flying robot, the NASA Astrobee, aboard the International Space Station (ISS). Using NVIDIA's Omniverse physics simulator and curriculum learning, we trained a deep neural network to replace Astrobee's standard attitude and translation control, enabling it to navigate in microgravity. Our results validate a novel training pipeline that bridges the simulation-to-reality (Sim2Real) gap, utilizing a GPU-accelerated, scientific-grade simulation environment for efficient Monte Carlo RL training. This successful deployment demonstrates the feasibility of training RL policies terrestrially and transferring them to space-based applications. This paves the way for future work in In-Space Servicing, Assembly, and Manufacturing (ISAM), enabling rapid on-orbit adaptation to dynamic mission requirements.

The US Naval Research Laboratory's (NRL's) Autonomous Planning In-space Assembly Reinforcement-learning free-flYer (APIARY) experiment pioneers the use of reinforcement learning (RL) for control of free-flying robots in the zero-gravity (zero-G) environment of space. On Tuesday, May 27th 2025 the APIARY team conducted the first ever, to our knowledge, RL control of a free-flyer in space using the NASA Astrobee robot on-board the International Space Station (ISS). A robust 6-degrees of freedom (DOF) control policy was trained using an actor-critic Proximal Policy Optimization (PPO) network within the NVIDIA Isaac Lab simulation environment, randomizing over goal poses and mass distributions to enhance robustness. This paper details the simulation testing, ground testing, and flight validation of this experiment. This on-orbit demonstration validates the transformative potential of RL for improving robotic autonomy, enabling rapid development and deployment (in minutes to hours) of tailored behaviors for space exploration, logistics, and real-time mission needs.

Probabilistic Graphical Models (PGMs) encode conditional dependencies among random variables using a graph -nodes for variables, links for dependencies- and factorize the joint distribution into lower-dimensional components. This makes PGMs well-suited for analyzing complex systems and supporting decision-making. Recent advances in topological signal processing highlight the importance of variables defined on topological spaces in several application domains. In such cases, the underlying topology shapes statistical relationships, limiting the expressiveness of canonical PGMs. To overcome this limitation, we introduce Colored Markov Random Fields (CMRFs), which model both conditional and marginal dependencies among Gaussian edge variables on topological spaces, with a theoretical foundation in Hodge theory. CMRFs extend classical Gaussian Markov Random Fields by including link coloring: connectivity encodes conditional independence, while color encodes marginal independence. We quantify the benefits of CMRFs through a distributed estimation case study over a physical network, comparing it with baselines with different levels of topological prior.

This paper presents a hybrid Sequential Convex Programming (SCP) framework for solving the unbalanced three-phase AC Optimal Power Flow (OPF) problem. The method combines a fixed McCormick outer approximation of bilinear voltage-current terms, first-order Taylor linearisations, and an adaptive trust-region constraint to preserve feasibility and promote convergence. The resulting formulation remains convex at each iteration and ensures convergence to a stationary point that satisfies the first-order Karush-Kuhn-Tucker (KKT) conditions of the nonlinear OPF. Case studies on standard IEEE feeders and a real low-voltage (LV) network in Cyprus demonstrate high numerical accuracy with optimality gap below 0.1% and up to 2x faster runtimes compared to IPOPT. These results confirm that the method is accurate and computationally efficient for large-scale unbalanced distribution networks.