arXane › Electrical Engineering and Systems Science › Electrical Engineering and Systems Science
Wednesday, 4 November 2026: 31 new papers, 14 cross-lists and 16 replacements in Systems and Control (eess.SY), in arXiv's order, with full abstracts. The same papers as arXiv's own listing, set as a newspaper; every title links to arXiv.
Millimeter-wave (mmWave) links can lose tens of decibels within milliseconds when a person or vehicle crosses the line of sight, and reactive handover recovers only after the outage has begun. Integrated sensing and communication (ISAC) promises to let the network see blockers coming and move users to another cell in advance. This paper asks how much such foresight is worth relative to the best reactive handover, and what sensing accuracy is needed to harvest it. We formulate cost-aware cell selection with knowledge of the future signal-to- noise ratio as a sequential decision problem, define the value of foresight through a policy-independent bound, and develop an ISAC-assisted handover framework in which a radio unit senses blockers with its own communication waveform and a simulated Open RAN xApp plans handovers. Ray-traced experiments in a 28 GHz street canyon, reported on held-out scenarios, show that, with a short look-ahead, perfect foresight closes most of the gap between service-aware (Layer-3) reactive handover and an instantaneous oracle. Harvesting this value, however, requires the user and nearby blockers to be localized to about a decimeter, well beyond what our radar tracker and meter- level user positioning provide; with realistic errors, even an uncertainty-aware planner largely defers to reactive handover.
Quantum kernel methods require estimating entries of a dense Gram matrix through repeated circuit measurements, making kernel construction expensive and statistically noisy. For a dataset of size $n$, dense construction requires $O(n^2)$ kernel queries, while existing approximation methods typically assume inexpensive access to kernel entries or use non-adaptive sampling. We study adaptive low-rank quantum kernel acquisition when the kernel is accessible only through a finite-shot quantum overlap oracle. Our framework, ICAQKL, uses residual-driven incomplete Cholesky construction to acquire informative kernel columns sequentially and form a positive-semidefinite rank-$r$ surrogate, reducing the training-kernel acquisition cost to $O(nrN_s)$ circuit executions for $r\ll n$, where $N_s$ is the number of shots per queried entry. The central difficulty is that finite-shot errors perturb both the queried kernel values and the adaptive pivot sequence that determines the low-rank factor. We develop a non-asymptotic perturbation framework for this stochastic adaptive process, establishing concentration over the realized queries, stability of noisy pivot selection and factor construction, and propagation of the coupled sampling and approximation errors to the QSVM objective, optimizer, prediction function, and classification margin. The resulting rank--shot conditions characterize when ICAQKL approaches the full QSVM induced by the same quantum feature map. Experiments on benchmark and biomedical datasets show favorable accuracy--query-cost tradeoffs, competitive performance with the full QSVM, and competitive or improved performance relative to the evaluated Nystr\"om baselines under comparable settings. ICAQKL is intended to enable measurement-efficient and theoretically controlled use of a selected quantum kernel, rather than to establish quantum advantage over classical kernels.
Advanced attitude determination and control (ADCS) techniques can be difficult for small satellite teams to adopt: tightly coupled implementations require substantial code changes to evaluate different actuators, sensors, orbits, mission goals, and control laws. Existing tools often emphasize orbit mechanics or flight software rather than closed-loop ADCS. This paper presents an open-source, modular ADCS framework in pure Python for rapid development, repeatable testing, and flight-oriented integration across heterogeneous spacecraft. The package supports torque- and command-level control, bound-respecting allocation, and full- and reduced-attitude objectives. Interchangeable estimation, control, planning, and actuator/sensor modules are reconfigured through configuration changes. Integrated simulation includes attitude and orbit propagation, sensor and actuator bias and noise, and environmental disturbances. An actuator-aware planner supports underactuated architectures, while hardware-in-the-loop (HIL) capability enables testing on mission-representative computers. Case studies include a 3U CubeSat with three magnetorquers and one reaction wheel, a magnetorquer-only 1U, and a 6U with three reaction wheels and thrusters. These demonstrate actuator and sensor failures, full- and reduced-attitude goals, and momentum management. HIL testing on a Raspberry Pi uses the same package. Adapting a published control law to a new actuator configuration requires fewer than five lines of configuration changes. The framework underpins flight software for an Earth-observing CubeSat with visible and long-wave infrared imagers, three magnetorquers, and one reaction wheel, without star trackers or propulsion.
The Musical Metaverse (MM) enables geographically distributed musicians, audiences, sound engineers, and AI agents to interact through synchronized audio, visual, kinematic, and haptic streams, imposing latency, synchronization, bandwidth, and scalability requirements that are difficult to satisfy with conventional bit-level communication. This paper argues that semantic communication, understood here as the transmission of musically meaningful information rather than raw audio-visual-haptic samples, provides a principled communication paradigm for the MM. We define the MM from a communication perspective and derive modality-specific require ments; propose a three-pillar framework for musical semantic communication comprising semantics extraction, semantic-aware orchestration, and cross-modal semantic synchronization; survey representative work across semantic communication, music AI and edge orchestration, identifying the gaps that prevent their integration; and propose a semantic-first reference architecture based on device-side semantic encoders, edge semantic routers, and cross-modal semantic synchronization. The resulting framework positions the MM as a concrete and demanding use case for 6G semantic communication.
Max-Pressure (MP) has emerged as a widely studied decentralized traffic signal control framework, thanks to its desirable theoretical properties and strong control performance. Early research on MP focused on improving control performance by exploring different variables of the traffic state for the definition of pressure. Subsequent studies expanded MP to incorporate the complexity of real-world transportation systems and emerging technologies. This paper provides a comprehensive review of these developments. The review first introduces the theoretical foundations of MP and examines early developments in movement weight definition. It then synthesizes major extensions for multimodal and multiscale network control, followed by recent developments in connected and automated vehicle (CAV) environments and the integration of MP with artificial intelligence (AI) and data-driven methods. Beyond summarizing existing approaches, this review examines their underlying assumptions, data requirements, implementation considerations, and practical limitations. Based on this comprehensive review, key research gaps and future directions are identified, including remaining limitations in the theoretical foundations of MP and the underexplored integration of autonomous vehicle control with the MP framework. Overall, this review provides a structured perspective on the evolution and current state of MP control and highlights opportunities for developing more proactive, robust, and practically deployable MP-based traffic control methods.
Capacity expansion models often dramatically simplify the physics of how power flows through a transmission network in order to preserve computational tractability. However, with the rapid growth of electrification, large loads, renewables, and storage, increased realism in planning future grids is imperative. We extend the open-source capacity expansion model GenX with AC optimal power flow physics, and analyze generation, storage, and transmission investment decisions for 20-zone, hourly models of Germany and Texas. We compare our results against those using two common approximations, the transport model and DC optimal power flow. We find that total costs and emissions largely agree across all formulations. However, we find substantial differences in outputs related to transmission, generation siting, storage, and reactive-power support between the approximate formulations and the true AC physics formulation. We identify where AC physics changes planning decisions and discuss how to potentially capture these effects without the full computational cost.
The letter aims to improve the math-to-real or math-to-sim transfer of a mission controller designed for a Temporal Reach Avoid Stay (T-RAS) specification. Our approach uses the control-theoretic concept of Incremental Input-to-State stability ($\delta$-ISS). We model the real-world system as an uncertain system, represented by a nominal mathematical model with an additive disturbance. Our method involves two steps: firstly, we make the nominal mathematical model and the real-world system $\delta$-ISS. Secondly, we design a mission controller for the mathematical model that satisfies a modified specification. We then prove that applying the same mission-level input sequence to the real-world system guarantees that the original specification is satisfied in the real world. Videos of real-world experiments are available at https://youtu.be/NvjkIZxtnPM.
This paper considers online intrusion-risk assessment and graded warning for unmanned aerial vehicles under bounded measurement attacks. To address this problem, we develop a monitoring framework that combines trusted state estimation, attack-aware risk assessment, and probabilistic signal temporal logic (PrSTL). Trusted observations support recursive state estimation, while potentially attacked observations inform integrity testing and intrusion-risk assessment. The key contribution is a confidence-set construction that combines normalized innovation statistics with physical attack bounds to characterize admissible measurement biases. The resulting family of bias-corrected Gaussian posteriors yields an upper bound on the probability of occupying a no-fly zone or its surrounding safety margin. Confidence guarantees and finite-cover numerical certificates quantify the reliability of this risk envelope, which is evaluated with or without an integrity alarm. The PrSTL monitor analyzes the resulting risk sequence to distinguish ordered, persistent, and repeated risk and produces graded warnings with explicit robustness margins and supporting windows. We validate the monitor in flight experiments with a Crazyflie quadrotor tracked by a Vicon motion capture system. Controlled comparisons quantify the contributions of individual monitoring components. Prescribed risk profiles verify the temporal semantics and illustrate how distinct risk patterns produce different warning levels.
Funnels, time-varying invariant sets around nominal trajectories, certify that a closed-loop system tracks its plan, and can be composed into verified motion plans. Their computation by sum-of-squares (SOS) programming is well established in formulation but not in practice: the result depends on implementation choices such as the initial guess, Lyapunov function normalization, and termination rule, which the literature documents only briefly. This paper presents a self-contained treatment of SOS-based funnel computation, implemented in the open-source package computeSOSFunnels. We propose a minimum-volume formulation with inlet and outlet constraints under which funnels with coinciding nominal endpoints compose by construction. Numerical experiments on a unicycle, with timing studies on a cart-pole and a 12-state quadrotor, characterize the pipeline: infeasible initial guesses are partly decidable before solving, the certificate shows no violations among 39,000 boundary samples per funnel under the true dynamics, and constructing the semidefinite program, not solving it, dominates computation time.
Can market notices read by a large language model (LLM) improve short-horizon electricity price spike forecasts? We study five regions of Australia's National Electricity Market. A local LLM extracted features once per notice, only notices visible at the forecast origin were used, and a gradient-boosting model produced onset probabilities. In an out-of-sample development back-test (July 2022 to July 2024), notice features did not lower the 60-minute Brier score of a numeric model including the market operator's next-hour forecasts. At 24 hours they did, but rule-based features did so almost equally and attribution to notice content was not established. A locally frozen protocol, not publicly deposited, governed a retrospective single-use holdout from 15 August to 17 October 2026. At 60 minutes, the hypothesis that LLM features lower the numeric model's Brier score was not established (H1; relative Brier skill -0.09%), so the planned comparison with rule-based features was not tested (H2). At 24 hours, the hypothesis that LLM features lower the score against a text-free control was not established (H3; +0.37%, 95% CI [-9.74, 1.39], inconclusive). By the frozen wording rule the non-significant 24-hour improvement comes mainly from recalibration and cannot be attributed to notice content, although post hoc the decomposition could not separate recalibration from resolution. The 24-hour base rate was 1.56% on the holdout against 14.0% in development. With 5 onsets on 3 days and 24-hour positives in 2 of 22 three-day blocks, the holdout neither established nor excluded a 24-hour effect.
Wheel-legged robots balance on two wheels while using articulated legs to change height, absorb impacts, and cross uneven terrain. On hardware, the balance controller must rely on noisy inertial and encoder measurements, and it must reject disturbances that are not known in advance, such as payloads and wind, without exceeding motor torque limits. This paper presents a moving horizon estimator (MHE) combined with an offset-free, constrained model predictive controller (MPC) for the sagittal balance of a 9.95 kg wheel-legged robot. The estimator uses a virtual-leg model augmented with a horizontal hip force, a body pitch torque, and a gyroscope bias. Its accelerometer model removes the predicted hip acceleration, and its process noise is increased when an impact is detected. We evaluate the approach in MuJoCo with simulated sensor noise, bias, quantization, and delay, and compare it with LQR baselines that use ground-truth state, a complementary filter, or the same MHE. In a driving, squatting, reversing, and bowing sequence, MPC with MHE reduced the RMS speed error to 0.11 m/s, compared with 0.24 m/s for LQR with ground-truth state. Under an unknown 2 kg payload and a 10 N headwind, the mean position error in wind was 6 cm, compared with 81 cm for LQR with ground-truth state. Across a 3 $\times$ 4 grid of payloads and headwinds, LQR with the same estimator fell in 3 of 12 combinations, compared with 1 of 12 for MPC. These results are limited to simulation on flat ground, and experimental validation on hardware remains future work.
We propose a Multi-technology AdopTion-opinion co-Evolution (MATE) model to describe the diffusion of $m \geq 2$ competing alternatives within a networked population whose opinions evolve under social influence and the feedback generated by adoption. Communities interact through two layers: a physical network governing exposure to adoption, and a social network shaping opinion evolution through social imitation. Individuals can adopt a technology, eventually becoming dissatisfied at a rate reflecting the quality of the technology, and switch to an alternative at a rate driven by their opinion of that alternative. For an arbitrary number of competing technologies $m \geq 2$, we characterize the equilibria and establish the existence of an adoption-diffused equilibrium in which all technologies coexist. For $m > 2$, the equilibrium configuration depends on the broader competitive environment: unlike the two-technology case, within-community relative adoption generally depends also on equilibrium opinions and on the other alternatives. Finally, we formulate finite-horizon opinion-shaping control problems, derive necessary optimality conditions via the discrete-time Pontryagin maximum principle, and compute numerical extremals using a forward-backward sweep method.
The growing autonomy of AI agents has enabled self-improvement, where agents iteratively refine their programs through execution feedback and performance evaluation. As these agents increasingly operate in physical environments and rely on wireless connectivity, communication conditions become an integral part of their improvement process. However, existing approaches largely assume ideal communication, overlooking how wireless conditions shape both the feedback that guides program modifications and the evaluation outcomes that determine which modifications survive. This challenge is further complicated by the coupling between agent behavior and wireless connectivity, as agents influence their own communication conditions through mobility and transmission decisions. To address this problem, we propose WiSI, a wireless-in-the-loop self-improvement framework that integrates spatiotemporal wireless channels, edge-assisted perception, and Large Language Model-driven program improvement. By incorporating wireless execution into the improvement loop, WiSI reveals how wireless execution experience shapes the generation and retention of program structures, and investigates the wireless conditions needed for robust improvement. Experiments show that agents improved under ideal links achieve only 40.0% task success under congestion, compared with 77.2% when improved under realistic wireless conditions. More importantly, wireless-aware feedback and selection jointly enable agents to discover and retain communication-efficient program structures that outperform parameter tuning and remain robust even with inaccurate run-time channel observations. We further find that moderate-load or mixed-network evaluation maintains strong performance across deployment conditions, while simplified models preserving link-rate statistics achieve comparable results without detailed spatial channel information.
The energy transition causes a growing number of distributed flexible energy resources in the low voltage grid. To utilize that flexibility, automation and control algorithms are needed, which typically are developed in theoretical research. We build a cyberphysical algorithm test bed to bridge the gap between theoretical research and practical application in cooperation with a local DSO in Walenstadt, Switzerland. To obtain suitable monitoring and data analysis capabilities from their existing data landscape, we propose tool-independent data harmonization measures in preparation for a digital twin of the grid. For data extraction from high-security network zones we developed a secure pipeline that implements a logical data diode ensuring one-way traffic only. Furthermore, we established a test community of voluntary prosumer households, whose flexible assets were made accessible through a common backend, giving us actuation capabilities. The presented publish-subscribe architecture is not only suitable to integrate various data sources into a central data collection, but also to manage access privileges of different stakeholders. This could support unbundling of an organization's data flows and the implementation of data space concepts.
Residual deployment can affect closed-loop tracking. We compare two ways to deploy one supervised acceleration-residual network in nonlinear model predictive control (NMPC) of a quadrotor. Cyclewise deployment repeats one correction across the horizon. Stagewise deployment evaluates shifted prediction nodes before the solve. Network parameters remain fixed during control. On prescribed force pulses, both schemes improve average tracking mainly after force removal, but worsen hover; some prediction checks also fail. Seed 1 reinforcement-learning benchmarks add no mean benefit under pulses or prescribed random forces. On 220 conditions, cyclewise and stagewise supervision reduce mean integral absolute position error (IAE) by 4.38% and 7.77%. On a separate 32-condition bank, fixed gain 1.5 lowers IAE by 27.47% relative to stagewise supervision but raises angular-rate root-mean-square error by 10.19%. In one synthetic computational fluid dynamics field, stagewise supervision lowers four-task mean IAE by 2.01% against a same-graph zero-residual control; some secondary metrics worsen. These simulations show that residual deployment affects tracking, with gains and limitations that depend on task and motion metric.
Integral quadratic constraints (IQCs) and scaled relative graphs (SRGs) are two prominent frameworks for analyzing feedback systems. Although both frameworks are rooted in the same graph separation principle, a general characterization of their relationship has remained unavailable. This paper establishes conditions under which the IQC- and SRG frameworks are equivalent. To this end, we introduce two new classes of systems, termed disjunctive and conjunctive systems, which are respectively characterized by unions and intersections of quadratic input-output constraints. Under structural assumptions, we prove equivalence between the IQC and SRG frameworks for these system classes. By exploiting this equivalence, we develop novel IQC- and SRG-based converse stability results for linear and nonlinear systems. Our results accommodate multiple quadratic constraints, extending well-known converse results in robust control beyond the single-constraint setting.
Large Language Model (LLM)-based agents can automate complex engineering workflows by interacting with tools, interpreting feedback, and making iterative decisions. This paper introduces TuneBench, a benchmark for evaluating Agentic AI in power system dynamic model calibration, where controller parameters are adjusted to reproduce target dynamic responses. The benchmark reflects a common Transmission System Operator (TSO) practice, in which equivalent models used for Dynamic Security Assessment (DSA) are calibrated against measurements from grid-code compliance tests when proprietary models are unavailable. Two cases of increasing complexity are considered: the governor and Automatic Voltage Regulator (AVR) of a synchronous generator and the control system of a Battery Energy Storage System (BESS), both tuned against active-power and voltage step tests within a budget of 15 simulations. LLM-based agents are compared with random search and Genetic Algorithm (GA) baselines over 10 independent runs. Only the most advanced models, Opus 5.5 and GPT-6, accurately reproduce all target trajectories, with median R2 above 0.995 in every test and tracking errors up to two orders of magnitude lower than the baselines. The agent logs show that these agents autonomously extract dynamic performance indices, such as settling time and damping, and relate them to specific control parameters, decomposing the tuning problem into physically interpretable sub-problems. The benchmark is publicly available at https://github.com/Power-Agent/PowerAgentBench/tree/main/benchmarks/dynamic/level3.
The effect of Tai Chi postures on the mechanical response of the body to external pushes is usually assessed with clinical balance scores, which say little about the underlying dynamics. We identify, for five subjects and eight postural conditions, linear models from the mechanical work transferred by a pendulum hitting the forearm to the change of the antero-posterior ground reaction force. Each trial contains 10 to 16 impacts, and each impact yields a transient of 0.55\,s only. Building on earlier analyses of these data, we improve the processing chain in three ways. The input is related to the energy lost by the pendulum and its repeatability is quantified. The output is defined component-wise, which makes it independent of the zeroing of the force plate. Finally, the impacts are stacked in a multi-experiment least-squares problem and validated on unseen impacts. Tenth-order ARX models then reach a median leave-one-impact-out fit of 65\%, against 72\% for the repeatability of the response itself, and never require stability enforcement, unlike estimation on averaged impacts. Closing the eyes lengthens the identified delay by about 70\,ms in all subjects, and the Tai Chi posture increases the antero-posterior force transmitted per joule in four of five subjects. Surface EMG envelopes show a tonic co-activation of the lower-leg muscles in this posture, a smaller reactive triceps burst and a fast habituation of the tibialis response. Together, they point to a stiffer, anticipatory postural strategy.
Embodied intelligent applications have expanded from air and ground into maritime environments, where missions such as coastal reporting require coordination among underwater sensing, surface gateways, and aerial relays. Yet field trials remain costly, and existing simulation tools make it difficult to study how cross-medium communication affects cooperation among these platforms. We present TriMedium, an integrated Air-Ground-Sea simulator for cross-medium embodied networking. TriMedium has three design elements. First, it provides a shared three-dimensional world where aerial, terrestrial, and underwater entities coexist, so underwater agent motion and visual observations stay geometrically consistent with air and ground nodes. Second, it uses a common simulation clock that preserves observation timestamps while scheduling delayed communication and computation, clarifying when the shore receiver obtains each observation. Third, modular quantization and compression interfaces accommodate heterogeneous telemetry and image pipelines, while integrity checks separate intentional lossy distortion from incomplete or corrupted delivery. Three case studies validate TriMedium. Multimodal data delivery tests telemetry and image pipelines, scene construction examines cross-domain spatial consistency, and positioning assesses geometric accuracy under delayed communication. Together, they show TriMedium enables embodied networking by quantifying cross-medium delivery dynamics and fidelity across heterogeneous payloads.
Mean-shift exploration enables assignment-free swarm shape assembly, but obstacle-constrained navigation additionally requires coordinated adaptation of the desired morphology and its physical execution. This paper presents an environment-adaptive navigation framework that integrates target-field planning, execution regulation, and morphology-preserving mean-shift exploration. A feasibility-aware graph search plans the position, orientation, and morphology of an unlabeled target field without specifying robot-specific goals. An execution-regulated reference manager adjusts field progression and morphology switching using swarm entering, coverage, and lag metrics. A moving-frame controller separates prescribed rigid motion from local mean-shift redistribution and incorporates neighbor interactions and obstacle responses. Simulations and hardware experiments demonstrate that the proposed framework coordinates morphology transitions and collective motion for assignment-free swarm navigation through cluttered environments.
We study finite-horizon safety verification for continuous-time stochastic systems modeled by stochastic differential equations. Given an open and bounded safe set, a compact initial set, a finite time horizon, and a prescribed violation-probability threshold, the objective is to certify that the exit probability from the safe set before the horizon remains below the threshold. Under uniform ellipticity of the diffusion matrix and regularity of the safe-set boundary, we develop a time-dependent barrier-like condition expressed as a set of functional inequalities, obtained by relaxing a parabolic initial-boundary value problem associated with the exit probability. If a function satisfies the condition, then the probability of leaving the safe set before the horizon is guaranteed to remain below the threshold. Conversely, when the exit probability is strictly smaller than the prescribed threshold uniformly over the initial set, the barrier-like condition is also necessary: there exists a smooth function satisfying it. Furthermore, when the dynamics are polynomial and the set descriptions are semialgebraic, the problem of verifying the condition admits a sum-of-squares formulation that is both sound and complete, yielding a complete degree hierarchy for finite-horizon stochastic safety verification.
Many emerging capacity expansion optimization formulations implicitly presume the convexity (or concavity) of power system adequacy risk (or capacity accreditation) metrics. These assumptions have yet to be formally justified. We present concise mathematical proofs validating the claim for certain metrics (expected unserved energy and effective load carrying capability), but also provide simple invalidating counterexamples for others (loss-of-load expectation and equivalent firm capacity). Formulations built on non-convex (non-concave) metrics structurally undervalue candidate portfolio adequacy, and so cannot claim to produce least-cost expansion plans.
This letter addresses the stability analysis of discrete-time linear systems with time-varying delay using a Lyapunov-Krasovskii functional (LKF) framework. An improved matrix-separation-based inequality (IMSBI) is developed to obtain tighter bounds for the summation terms arising in the forward difference of the LKF. Based on this result, an augmented LKF incorporating additional cross-terms is constructed to capture the interaction between system states and delay evolution. This leads to a less conservative delay-dependent stability criterion expressed as linear matrix inequalities (LMIs). Numerical results demonstrate improved stability margins and a reduced number of decision variables compared with existing methods.
Proactive intersection safety analysis depends on trajectory data whose reliability varies by sensing modality, yet infrastructure-perception and surrogate safety measure (SSM) research have only recently begun to converge. This paper synthesizes both literatures alongside control theory to clarify how sensing choices, safety metrics, and predictive control connect end-to-end, and where infrastructure-derived risk signals could inform autonomous vehicle decision-making. Searches covered light detection and ranging (LiDAR), cooperative perception, time-to-collision (TTC), post-encroachment time (PET), extreme value theory, model predictive control (MPC), and signal control. Screening considered trajectory-based safety and control relevance, excluding detection benchmarks without safety linkage. This structured narrative synthesis retained 56 sources. No single SSM is universally valid: one-dimensional TTC and modified TTC suit longitudinal conflicts, two-dimensional TTC extends time-based measurement to crossing geometries, and observed PET captures crossing and vulnerable road user (VRU) conflicts but only after the event. LiDAR consistently outperforms camera-only sensing for continuous three-dimensional trajectory extraction, illumination robustness, and privacy, though its range degrades in heavy rain, fog, and snow. Its output becomes control-ready only when paired with filtering--Kalman, Savitzky-Golay--that stabilizes noisy detections. This review complements it from the infrastructure side: it examines how roadside sensing affects that chain's reliability, classifies SSMs by conflict geometry and temporal availability, and maps risk components onto control objectives and constraints for infrastructure and connected and automated vehicle (CAV) applications.
Port-Hamiltonian models describe physical systems by separating stored energy, dissipation and external inputs. Learning such models from data requires a choice of representation and fitting method. We compare using neural networks, analytic expressions and combinations of the two for representing the different components of the port-Hamiltonian system, and compare fitting the models with gradient-based optimisation or sparse regression. Testing on simulations of mechanical and electrical systems, with and without added noise, we present prediction accuracy and computational cost of the different approaches.
Online dynamic security assessment and day-ahead planning require rapid identification of where disturbances will produce significant dynamic activity. Full time-domain simulations are computationally demanding, while existing machine-learning approaches predict system-level outcomes or require post-disturbance trajectories. This paper proposes a pair-aware Graph Isomorphism Network with Edge features (GINE) framework comprising two task-specific models for predicting bus-voltage and generator apparent-power activity from the operating point and contingency description. Physically aware KPI cuts convert simulated responses into ordinal activity targets for model training. Evaluations on Nordic 32 and the French transmission system with more than 6 000 buses show that pair-aware fusion and graph embeddings improve on a local baseline on Nordic 32, with GINE the lower-cost graph encoder, and that the framework remains effective at large scale. As a downstream application, the predicted activity maps guide Adaptive Model Simplification, reducing continuous and discrete model variables by 72% and 66%, achieving speedups of up to 3.16x, and preserving voltage trajectories when critical substations are retained. These results demonstrate scalable, pre-disturbance prediction of component-level dynamic activity.
Accurately predicting eddy-current and hysteresis losses in mild steel components such as the structural parts and tank of power transformers remains a challenge in transformer design. Conventional numerical approaches for eddy-current loss analysis require discretization at a fraction of the skin depth. While accurate, such fine volumetric mesh becomes computationally prohibitive for large power transformers. Moreover, hysteresis loss is not always included in this analysis. To investigate the trade-off between computational cost and accuracy, this work presents a comparative analysis of eight distinct modeling frameworks for predicting both eddy-current and hysteresis losses in mild steel. These include efficient methods avoiding meshing, such as standard impedance boundary conditions, and more accurate approaches retaining full volumetric meshing, namely complex permeability, nonlinear single-valued B-H curve, and Jiles-Atherton hysteresis modeling. To accurately represent the material's electromagnetic behavior, the properties of the mild steel used are measured directly from samples. Finally, loss measurements are conducted to benchmark all eight frameworks, demonstrating mean absolute percentage errors between predictions and measurements generally below 20%.
Large Language Models (LLMs) lack physical grounding and extrapolation. This paper formalises the Novelty Horizon as tasks needing extrapolation beyond training support, causal reasoning absent from learned correlations, or post-frontier information. A forensic analysis tests LLMs on unguided, guided, and ground-truth data from an underrepresented diode-clamped short-circuit experiment using temporal, spectral, and statistical validation against a first-principles Modified Ohm's Law. Both script-based approaches fail catastrophically with physical and spectral inconsistencies, extreme statistical pathologies, zero Generative Fidelity Score, and linear-regime outputs. Direct prompting corroborates distinct failure modes. This paper introduces the Past Information Boundary and physics-based authentication. LLM data generation cannot extrapolate past the Novelty Horizon essential for novel discovery. Hybrid neuro-symbolic systems must preserve human causal reasoning and experimental design.
OperAItor, an open-source framework for developing and evaluating large language model agents for supervisory fault management in chemical processes. The framework is demonstrated on a dynamic reactor-distillation process representing isopropanol dehydrogenation, with physically based thermodynamics and modified reaction kinetics to permit liquid-phase operation. The process is represented by a 113-variable differential-algebraic model and operates at a nominal feed rate of 100kmol/h, producing approximately 4t/h of acetone. The simulated plant includes 8 PI control loops, 22 process alarms, 5 duty-standby pump trains, and 11 independent safety instrumented functions. The agent architecture integrates multivariate statistical process monitoring, fault diagnosis, knowledge-graph retrieval, action planning, and supervisory action generation. Five action-validation configurations are investigated, comprising operation without validation, an LLM-based judge, digital-twin simulation, and deterministic admissibility checks with and without consequence projection. The framework is evaluated using four LLMs across 20 scenarios involving single and multiple simultaneous faults. Without agent intervention, 18 scenarios result in plant stops, corresponding to an economic loss of $399,000 relative to fault-free operation under the assumed benchmark economics. For GPT-5 nano, Qwen3 8B, and Mistral 7B, validated agent configurations reduce the number of stops to 5-8 and recover $265,000 of the loss. The study provides a reproducible benchmark for comparing agent architectures, language models, and action-validation methods for supervisory fault management in chemical processes.
A general mathematical model of experience formation is considered, in which the learning level is defined as the probability that the next value of the uncertainty factor has already been mastered and not forgotten. Within this model the shape of the learning curve is derived, and it turns out to be determined primarily by the distribution of situation occurrence: a uniform distribution yields an exponential curve, whereas a power-law distribution yields a power-law curve whose exponent equals the ratio of the distribution exponent to that same exponent increased by one. Both mechanisms generating power-law curves, established earlier in the analysis of debugging, assembly and repair activities, are shown to have been independently rediscovered in deep learning theory as the "frequency" and "spectral" mechanisms of scaling laws for artificial neural networks, while the exponential curves of machine learning (contraction operators, strongly convex optimization, stationary reinforcement) are generated by the same constancy of the contraction coefficient that was revealed half a century earlier in models of radioactive decay, the first-order lag element, and others. A solution is given for the problem of choosing a training distribution that differs from the deployment distribution. A methodological consequence is discussed: three distinct mechanisms generate curves indistinguishable in shape, so an observed learning curve does not allow one to judge the "internal" structure of the learning system.
This paper investigates iterative methods for solving feedback-type linear equations under small phase condition. We first develop an iterative solution scheme for linear equations with sectorial (semi-sectorial) matrices, i.e., matrices whose numerical ranges are contained in an open (closed) half complex plane. Exploiting the geometric properties of the numerical range, we introduce the notion of numerical-radius contraction as a convergence criterion. Such a scheme, with some extra effort, leads to an iterative method for the feedback-type equations under small phase condition, whose convergence condition is characterized in terms of phase information. This complements classical iterative methods for feedback-type equations based on the small gain condition. A case study on large-scale power networks is presented to illustrate the computational efficiency of the proposed method.
Synchronization is a fundamental collective phenomenon in many natural and engineered networks. The stable operation of oscillator networks, such as electric power grids, requires phase locking with phase differences across edges below a given threshold. A central question is whether a given network supports such a state. In this contribution, we extend a convex optimization formulation of the nonlinear fixed point equations of the Kuramoto model, that also describe lossless power grids. Based on this formulation, we derive rigorous and tight bounds on the error introduced by an arbitrary approximation of the original model, in particular linear approximations. The bounds yield edge-wise regions of trust around the approximation, within which the true nonlinear solution must lie. They further provide sufficient and necessary conditions for phase locking under prescribed phase difference limits. Numerical experiments show that the bounds are tight over a broad range of parameters.
In robotics, shape control, or shape servoing, refers to the automatic manipulation of a deformable object toward a desired shape. This article is the second of a two-part tutorial on data-driven, Jacobian-based shape control, in which the deformation Jacobian, a local model of how robot motion changes the measured shape, is estimated from measurements. Part I built the shape-servoing loop and derived a local stability condition. This paper starts by using the singular value decomposition of the deformation Jacobian to identify shape changes that can be produced by small robot motions, those that require large robot motions, and those that are locally unavailable. It then analyses how the choice of shape features determines how the controller regulates the object's shape and, therefore, how much the object deforms on its way to the target. It then describes three ways in which the shape control loop can fail: loss of controllability, loss of observability, and occurrence of mechanical instabilities. It closes with shape-control extensions to time-varying targets, secondary tasks, and three-dimensional objects.
We consider a UAV-assisted status-update system in which a rotary-wing unmanned aerial vehicle (UAV) collects updates from multiple sensors, stores the newest update of each sensor, and forwards selected buffered updates to a ground station (GS). We minimize a weighted sum of the time-average GS-side age of information (AoI) and UAV energy consumption over a finite horizon. The optimization is subject to half-duplex operation, sensor and UAV energy-causality constraints, transmission-rate and power limits, and a terminal-location constraint. Because collection changes the information age at the UAV but does not reduce the GS-side AoI until the buffered update is forwarded, mobility, collection, and forwarding decisions are coupled over time. We use a hierarchical AoI-energy actor-critic (HAE-AC) controller in which a coarse-timescale actor selects return-feasible UAV movements, while a deterministic per-slot scheduler constructs feasible collection or forwarding sets and assigns the minimum transmit power required for each scheduled transmission. In a 100-sensor deployment, HAE-AC reduces the average GS-side AoI by about 22% relative to an adaptive staleness-gated benchmark. Compared with forward-upon-collection, selective buffered forwarding reduces the number of downlink bursts by roughly 30% while increasing the number of collections per sensor by more than 23%.
The Musical Metaverse relies on distributed audio, visual, inference, and delivery functions deployed across heterogeneous edge-cloud resources. Queue-aware orchestration frameworks such as SPARQ jointly optimize service placement, routing, and resource allocation, but fix the execution regime of each resource in advance: it is either shared, favoring statistical multiplexing, or guaranteed, favoring performance isolation. This paper introduces MusiSPARQ, a subscription-aware control framework that promotes resource isolation from a fixed infrastructure attribute to a runtime decision. Each switchable resource is represented through coupled shared- and guaranteed-resource pools drawing from the same physical-capacity budget, and at each orchestration epoch the controller selects which class-specific service flows remain shared and which receive isolated capacity. The per-epoch decision jointly accounts for class-specific latency objectives, per-class maximum-latency limits, queueing stability, isolation cost, and switching overhead. In a dynamic virtual-concert scenario, MusiSPARQ reduces the Premium latency-target violation rate from 16.19% to 0.29% relative to fully shared execution, while reserving on average only 0.71% of capacity and, unlike static isolation, keeping Standard and Basic traffic within their latency targets.
Hamilton-Jacobi (HJ) reachability provides certified backward reachable tubes (BRTs) for safety-critical robotic systems, but the level-set computation scales exponentially in the state dimension and has historically been performed offline. As autonomy stacks (ADAS/AV, humanoid robotics, and vision-language-action models) increasingly rely on black-box, probabilistic policies, a deterministic safety layer with formal guarantees is an essential complement. We present a CUDA implementation of the canonical WENO5 + TVD-RK3 + Lax-Friedrichs level-set Hamilton-Jacobi-Isaacs solver for 2D, 3D, and 4D dynamics that is fast enough to run inside a robot's control loop. In 32-bit floating point (FP32) on a single workstation GPU (NVIDIA RTX 4000 Ada), it computes the 3D Air3D BRT at 112 Hz, a 4D pursuit-evasion BRT at 61 Hz, and a 2D wind-affected drifter BRT at 74-147 Hz, 5.9-17.5x faster than the JAX reference hj_reachability at matched precision and 1.7-16x faster than a torch.compile reimplementation of the same scheme. In 64-bit floating point (FP64) it agrees with hj_reachability on eight benchmark cases, reproduces Mitchell's Air3D BRT to within 0.04% of the reference cell count, and runs 2.0-4.6x faster. On an embedded NVIDIA Jetson AGX Thor it computes Air3D at 66 Hz (N=41) and the 2D drifter at 54 Hz, 3.8-8.9x faster than hj_reachability on the same device. As a multi-agent application, an in-kernel label-propagation extension recovers the K-agent capture-basin Voronoi tessellation from a single combined HJ solve, independent of fleet size; it runs at 4.0 ms per tick in a closed-loop wildfire-response simulation of six quadcopters under time-varying, non-uniform wind.
This paper presents an innovative guidance and control method for a space manipulator that actively maneuvers its base to hold a tumbling object stationary relative to the base during proximity operations. This simplifies motion planning, reduces collision risk, and enables impact-free capture using single- or multi-arm configurations, with or without predefined grasping interfaces. The proposed approach involves two phases. In the pre-grasping phase, a synchronization control aligns the target's center of mass (CoM) with the system's instantaneous center of rotation (ICR), virtually rigidizing the target-servicer system to keep the object stationary for reliable grasping. In the post grasping phase, a control strategy based on Hamiltonian dynamics and optimal control stabilizes the combined system by minimizing time, fuel, or energy, while respecting base torque limits and multiple grasping constraints. For structured targets, applied forces and torques stay within their bounds; for unstructured ones, friction constraints prevent slipping. A case study demonstrates the method's effectiveness in enabling both the subsequent grasping and detumbling phases.
Reactive synthesis is the problem of generating a program that produces outputs satisfying a logical specification for every input stream. We study distributed synthesis for two processes communicating in rounds over unreliable links, with at least one message delivered in every round. Despite this delivery guarantee, we show that synthesis from linear temporal logic specifications is undecidable, even when messages carry the processes' complete causal histories. More generally, we give a complete characterization of the communication restrictions that make synthesis decidable. Restrictions are described by a message adversary $A$, a set of allowed infinite graph histories: sequences of communication graphs, each specifying which messages are delivered in a round. Iterated chromatic subdivision associates these histories with points in the unit interval, yielding an image $\pi(A) \subseteq [0,1]$. For every $\omega$-regular message adversary $A$, synthesis is decidable if and only if $\pi(A)$ is nowhere dense. In the decidable cases, our procedure takes a deterministic parity automaton for $A$ and a specification, decides realizability, and constructs a realizing distributed algorithm whenever one exists. Further, our topological arguments shed new light on the fundamental requirements of synthesis. They show that, for $\omega$-regular message adversaries, synthesis is decidable if and only if consensus is repeatedly solvable in bounded time. This links two previously unrelated problems from formal verification and distributed computing through the processes' uncertainty about communication.
Decisions about the direction of research in power engineering determine which technologies mature and, through them, shape infrastructure that lasts decades, yet their consequences cannot be observed before they are taken, and existing forecasts of research fields extrapolate a single trajectory. Here we introduce research world models, which treat a research field as a dynamical system that can be acted upon, and establish one for power engineering from 1.1 million papers. A distilled language model reads each paper into role-typed concepts, two sophisticated, high-fidelity graph-temporal neural network cores learn how the resulting concept network evolves, and operators learned from natural experiments in the publication record estimate the effects of interventions. In temporally held-out tests, the model forecast years ahead which concepts would grow and which method-target pairs would be joined for the first time, outperforming tuned machine-learning baselines in most settings, and its estimated intervention effects tracked those realized in natural experiments it had not seen. Its training scheme and three of its operators replicated in two fields withheld from development. Questions posed in natural language are compiled by a domain-specialized 27-billion-parameter language model into executable actions and answered as paired simulations with and without the intervention. The answers were rated 4.1 of 5 for faithfulness by an independent language-model judge. By revealing the latent dynamics of a research field and its response to interventions, research world models offer a new way to simulate and test research decisions before they are made.
Multicopters are a type of aerial robot seeing extensive real-world applications but with range and endurance as critical bottlenecks. Optimization of its energy efficiency through energy-aware flight planning (e.g. cruise speed-scheduling) has hence been an important topic. The mass of the multicopter is an important factor to consider, as it has a significant impact on efficiency, and many common applications are subject to frequent mass variation due to payload change. This work derives closed-form scaling relationships between the energy-optimal steady horizontal flight behavior of a multicopter and its vehicle mass, and validates them experimentally on two multirotor platforms. Specifically, for mass-only variation, i.e., when payload changes the vehicle mass while the airframe geometry, aerodynamic characteristics, and rotor configuration remain unchanged, we show through a combination of numerical simulation, theoretical derivation, and experimental validation that (1) the energy-optimal velocity is proportional to the square root of vehicle mass; (2) the energy-optimal pitch angle is constant with respect to mass; and (3) the optimal energy per meter is proportional to mass. The analysis shows that these three relationships arise from a common mass-similarity property of the steady-state flight and rotor-aerodynamic equations. Furthermore, the obtained scaling relationships enable simple yet effective cruise-speed planning to optimize energy efficiency and extend the range and endurance of the multicopter.
Large language models (LLMs) on personal devices keep user data local, but limited memory restricts their general capability. Since most user-specific tasks do not require the breadth of frontier models, a practical solution is to adapt a shared on-device model into lightweight specialists. However, backpropagation is prohibitive in memory, and users rarely provide labeled data. We present MobileSpec, a system that trains specialists with forward passes alone and learns from automatically checked task outcomes, as in reinforcement learning with verifiable rewards (RLVR). Because forward-only updates are built from random perturbations, training is efficient only when these perturbations resemble the weight changes of RLVR, which largely preserve the singular values of pretrained weights and rotate their singular vectors. The core algorithm of MobileSpec, FuRA-S, produces such perturbations while training only the block-spectral coordinates of pretrained weights, namely the singular values of their column blocks. As we show theoretically and empirically, rescaling these blocks independently shifts their relative contributions to each weight matrix and thereby rotates its singular vectors. Since this update can be rewritten as a transformation of the layer input, FuRA-S also adapts INT4 models without modifying their weights. With specialists of only 1-5 MB that update 0.015-0.05% of the parameters, MobileSpec raises the success rate of a phone-use agent from 6-23% up to 89-100% and trains a personalized 8B model on an iPhone 16 Pro overnight. On RLVR benchmarks, FuRA-S is competitive with full-parameter evolution strategies (ES) and outperforms ES over LoRA adapters and quantized ES on INT4 models.
By harnessing the inherent commonsense capabilities of pretrained Large Language Models (LLMs) and Vision-Language Models (VLMs), current Vision-Language Action(VLA) models for legged robots excel at interpreting language-based navigation commands but lack the physical grounding and explicit safety assurance required for real-world operation. Therefore, this paper aims to cage Legged-Vision-Language Navigation (VLN) through a safety architecture that provides physical grounding and ensures navigation safety. The key idea is not to abandon the powerful capabilities of end-to-end VLA planning, but rather to augment and cage them. In particular, we integrate a high-level VLA planner with a low-level physical-grounding layer that leverages efficient state estimation, in-situ 3D spatial perception, and an Signed Distance Field (SDF) representation to generate executable trajectories. We provide a navigation planner with provable collision-avoidance guarantees under the stated assumptions by combining a low-dimensional Bezier-parameterized MPPI planner with Discrete Exponential Barrier Function (dEBF) constraints. The proposed Legged-VLN system is modular, making it compatible with various VLM/VLA models and agnostic to the robot platform. The proposed system is validated with extensive real-world experiments on humanoid and quadrupedal robots, demonstrating versatile, efficient, and safe vision-language driven navigation capabilities.
In this paper, we formulate a two-player zero-sum game under dynamic constraints defined in terms of a hybrid inclusion. The game consists of a min-max problem involving a cost functional that depends on the actions and corresponding (potentially non-unique) solutions, defined as functions of hybrid time and, hence, can flow or jump. We present sufficient conditions given in terms of Hamilton-Jacobi-Isaacs-like equations to establish a bound and to exactly evaluate the worst-case cost under the optimal strategy. Under additional conditions, we show that the proposed optimal state-feedback laws render a set of interest pre-asymptotically stable for the resulting hybrid closed-loop system. The results are illustrated in an application with sporadic jumps and linear quadratic hybrid dynamics.
Modern unit commitment (UC) scenarios, particularly those involving uncertainties from demand fluctuations and renewable energy sources, pose significant challenges to traditional optimization methods due to the large search spaces they encompass. This paper proposes a quantum computing-ready optimization pipeline for deterministic and stochastic UC, integrating classical, quantum-inspired, quantum annealing, and hybrid quantum-classical solvers across test systems, including the IEEE 118-bus system. A novel Quadratic Unconstrained Binary Optimization (QUBO) reformulation accommodates generator scaling and 24-hour scheduling horizons with compact Boolean logic penalties for startup, cooldown, and spinning reserve constraints. A divide-and-conquer strategy and Quantum Computing Monte Carlo (QCMC) simulation address scalability across 27,730 uncertainty scenarios, reducing compilation time by up to two orders of magnitude. Experimental results demonstrate the pipeline's ability to achieve efficient, scalable, and accurate UC optimization under uncertainty: the hybrid quantum method approach matches the classical baseline, reduces stochastic UC runtime from 40.76 to 14.79 hours, and characterizes pure quantum annealing as limited to small systems with 0.22% to 2.18% encoding errors and constraint violations. These results provide an analysis of the scalability and computational practicality of hybrid quantum optimization for generator scheduling in UC.
The feed-forward propagation through neural networks is crucial for training and inference. To trade off efficiency and accuracy, it has been proposed to consider intermediate outputs of hidden states as data-dependent regularization during training and, ultimately, as early-exits during inference. In this work, we analyze the joint optimization of loss functions computed for the output layer and intermediate layers using optimal control theory. We prove that ResNets trained in this joint fashion exhibit the so-called turnpike phenomenon, this implies that the number of layers required to reach a prescribed loss level is bounded from above independently of the overall network depth. This induces a similarity property across the internal representations of early-exiting networks of varying depths. We connect the turnpike to the overthinking phenomenon, and provide theoretical insights into the success of early-exit criteria derived from softmax probabilities, specifically entropy and max-probability. In a range of experiments, we demonstrate the existence and effect of turnpikes and the resulting similarity in the early-exit performance for ResNets of varying depths.