Search Results (583)

Reliable digital twins of complex nonlinear systems require not only high predictive accuracy but also physical consistency, robustness under degraded operating conditions, and explicit uncertainty handling. Purely data-driven models often suffer from poor generalization, unphysical behaviors, and limited interpretability when facing noisy measurements and unseen operating conditions. This paper introduces a knowledge-guided physics-informed hybrid learning framework that integrates recurrent neural networks with Unscented Kalman Filter (UKF) state estimation and embedded thermodynamic constraints within a unified uncertainty-aware architecture. The proposed PI-LSTM-UKF framework achieves competitive predictive accuracy and improved physical consistency relative to the residual-learning hybrids by tightly integrating physics-informed recurrent learning, thermodynamic constraints, and sequential UKF state estimation. While the UKF provides robust recursive correction under noisy measurements during closed-loop operation, the physics-informed Long Short-Term Memory (PI-LSTM) learns nonlinear corrections and long-term dynamics that cannot be captured by the linear model alone. The proposed framework is systematically benchmarked against a hierarchy of seven modeling approaches, including Dynamic Mode Decomposition with control (DMDc), Sparse Identification of Nonlinear Dynamics (SINDy), and residual-learning variants based on Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). High-fidelity Simscape simulations of a Rankine-cycle steam turbine system are used as a challenging simulation-based case study. Results show that the knowledge-guided hybrid approach achieves competitive predictive accuracy, improved physical consistency, and robust performance under an unseen load profile, severe thermodynamic degradation, valve hysteresis, and substantially elevated sensor noise. The framework provides a promising simulation-based foundation for uncertainty-aware digital twins of nonlinear thermal power systems. Validation using operational plant data remains necessary before its application to real-time monitoring and predictive maintenance. Full article

Retired lithium-ion batteries retain considerable residual value for second-life energy storage applications, but significant variations in health conditions and complex operating scenarios make accurate remaining useful life (RUL) prediction challenging. To address the limited availability of capacity measurements and the poor adaptability of conventional models to dynamically fluctuating degradation trajectories, a hybrid RUL prediction framework integrating Wavelet Packet Energy Entropy (WPEE), a Fractional-Order Grey Model (FGM), and an Unscented Kalman Filter (UKF) is proposed. WPEE extracted from discharge voltage signals is employed as a degradation indicator, while a Box–Cox transformation enhances its correlation with capacity. An Adaptive Mutation Particle Swarm Optimization (AMPSO) algorithm is used to determine the optimal fractional-order parameter, and the optimized FGM is incorporated into the UKF state-transition process for recursive state correction. Validation was conducted using four retired lithium-ion cells and two series-connected battery packs with different health conditions at prediction starting points of 20, 25, and 30 cycles. The results show that the proposed method effectively tracks degradation evolution, with RUL prediction errors within 7 cycles for retired cells and within 6 cycles for battery packs. Across all 18 prediction cases, FGM–UKF achieved an overall mean AE of 3.111 cycles, lower than those of FGM (5.111 cycles), GM(1, 1) (4.000 cycles), and LR (4.278 cycles). These results demonstrate the effectiveness and robustness of the proposed framework for lifetime assessment in second-life battery energy storage systems. Full article

An Adaptive-Parameter Maximum Correntropy Kalman Filter (APMCKF) algorithm is proposed to address state estimation degradation in semi-active suspensions caused by non-linear coupling between time-varying physical parameters and non-Gaussian noise. First, a time-varying dynamic model with non-linear damping is established via bench tests. A genetic algorithm (GA) globally optimizes key physical parameters to suppress model mismatch. Second, the APMCKF integrates an adaptive suspension parameter update mechanism. This closed-loop mechanism refreshes the system state matrix in real-time, effectively overcoming state-tracking lag. Concurrently, the maximum correntropy criterion (MCC) is embedded within the Sage–Husa recursive framework to dynamically reconstruct the observation noise covariance matrix, ensuring robust filtering under heavy-tailed noise. Simulations under ISO Class A–D random road profiles demonstrate that the APMCKF reduces the root-mean-square error (RMSE) by 62.33–81.24% compared to the adaptive Kalman filter (AKF). It also outperforms the adaptive-parameter Kalman filter (APKF), yielding a 27.49% accuracy improvement on Class D roads where non-Gaussian noise is most severe. Moreover, comparative evaluations against standard non-linear Bayesian filters demonstrate that the APMCKF successfully overcomes the truncation errors of the Extended Kalman Filter (EKF) and the tracking hysteresis of the Unscented Kalman Filter (UKF), reducing the average RMSE by up to 74.98% and 60.76%, respectively, under severe Class D non-Gaussian excitations. Furthermore, the algorithm exhibits excellent disturbance rejection under transient speed bump impacts and maintains stable error reduction across vehicle speeds of 10–25 m/s. Ultimately, the APMCKF delivers high-precision estimation and exceptional robust stability under variable speeds and non-Gaussian disturbances. Full article

The proliferation of space debris in Low Earth Orbit (LEO) poses a growing threat to operational satellites, requiring robust surveillance and tracking systems. Radar is a primary technology for monitoring these objects; however, standard tracking algorithms often degrade when measurements are sparse, corrupted by non-Gaussian noise, or available only as short tracklets. Traditional methods, such as the Unscented Kalman Filter (UKF), rely on explicit physical and statistical models that may struggle to converge under highly nonlinear dynamics, uncertain initialization, and non-ideal sensor perturbations. In this study, we propose a hybrid deep learning framework for learned one-step tracklet filtering of radar measurements. The architecture consists of a stateful Gated Recurrent Unit (GRU) layer followed by one-dimensional Convolutional Neural Network (1D-CNN) layers, complemented by variable-specific preprocessing strategies, including residual learning for range and relative pivoting for azimuth, to handle the scale disparities and heterogeneous behavior of radar observables. This design combines the ability of GRUs to model temporal dependencies with the effectiveness of CNNs in extracting local features for signal denoising. The method is validated on synthetically generated LEO trajectories with realistic orbital perturbations and tunable radar noise profiles, including Gaussian noise, impulsive spikes, transient degradation, and state-dependent perturbations. Compared with EKF- and UKF-based analytical baselines, the proposed model achieves lower filtering error and improved robustness under severe non-Gaussian disturbances. Stress testing further shows that the network can reject non-physical sensor anomalies without requiring long initialization warm-up phases, making it suitable for sparse short-tracklet processing in synthetic SST benchmark scenarios. Full article

A fractional-order equivalent circuit model (FOECM) provides a compact and physically interpretable representation of the memory-dependent polarization behavior of lithium-ion batteries. Leveraging this property, a fractional order model-guided residual-correction framework is proposed for state-of-charge (SOC) estimation, in which the FOECM, unscented Kalman filter (UKF), and long short-term memory (LSTM) residual learner are integrated into a unified estimation chain rather than treated as separate modules. In this framework, the FOECM is parameterized using Dynamic Stress Test (DST) data and incorporated into the UKF to construct the FOECM + UKF estimator. The LSTM learns the history-dependent SOC residual from sequences of measured operating signals and FOECM + UKF SOC estimates, and its output is added to the UKF estimate without replacing the fractional-order physical model. The proposed hybrid estimator is trained and configured using the available DST, Supplemental Federal Test Procedure (US06), and Federal Urban Driving Schedule (FUDS) data, and is independently evaluated on the US06 and FUDS profiles of Cell 008. Compared with the FOECM + UKF estimator, the proposed hybrid estimator reduces the SOC root mean square error (RMSE) from 2.70% to 0.90% on US06 and from 2.93% to 1.20% on FUDS, with mean absolute error (MAE) values of 0.66% and 1.03%, respectively. These results demonstrate the effectiveness of coupling fractional-order memory modeling with sequence-based residual correction under the tested dynamic operating profiles. Full article

Hybrid rocket motors are an attractive option for the upper-stages of low-cost small launchers, but are susceptible to variability in performance both in time and between firings. Moreover, key contributors to hybrid motors’ performance such as oxidizer-to-fuel ratio ($O/F$) are difficult to estimate, and by extension, to control. Four approaches were evaluated for the estimation and control of $O/F$ under system uncertainty, including through on-line estimation by an Unscented Kalman Filter (UKF). A Monte Carlo analysis was conducted of a simulated hybrid kick motor, where key sources of system uncertainty such as the characteristic velocity efficiency (${\eta }_{c^{*}}$), fuel regression coefficients, and oxidizer flow characteristics were allowed to be variable. Feedback control of $O/F$ informed by the UKF obtained 6.8% smaller control error than the best alternative approach. Yet the Monte Carlo analysis showed that among uncertainty sources considered, ${\eta }_{c^{*}}$ was the primary driver of performance variability, while $O/F$ regulation had a small influence. This was because the total and specific impulses were relatively insensitive to $O/F$ for the considered motor configuration and ranges of $O/F$ observed during the simulated burns—highlighting the importance of system uncertainty quantification when formulating performance-regulating interventions. Further, the proposed UKF observer provided data-informed estimates of combustion efficiency and propellant residuals in time, which are valuable for the planning and execution of accurate orbital insertions in a kick motor susceptible to performance uncertainty. The developed uncertainty quantification and control modeling framework can be used also during the design and assessment of other control interventions under system uncertainty. Full article

This paper proposes a trajectory-tracking algorithm for reentry vehicles based on KalmanNet with a time-varying observation matrix. First, a nonlinear state evolution model of the reentry vehicle and a radar measurement model are developed in the radar measurement coordinate system. Then, inspired by the computation process of the Kalman gain (KG) in the extended Kalman filter (EKF), the recurrent neural network (RNN) architecture of KalmanNet is improved. The gated recurrent unit (GRU) originally used to track process noise statistics is removed. Instead, the input features are redesigned to directly estimate the prior state covariance. Furthermore, another GRU is introduced to estimate the time-varying observation matrix, considering the nonlinear characteristics of radar measurements. The calculated observation matrix is fed into both the GRU responsible for estimating the covariance of the difference between the predicted observation and the observed value and the fully connected layer that computes the KG. Finally, the proposed method is compared with six representative algorithms, including EKF, particle filter (PF), unscented Kalman filter (UKF), convolutional neural network (CNN), Long Short-Term Memory (LSTM), and the original KalmanNet. Simulation results demonstrate that the proposed method achieves the highest estimation accuracy, while its computational time remains nearly the same as that of the original KalmanNet. Monte Carlo simulations under three model-mismatch conditions are conducted to validate the robustness of the proposed method. Full article

To address localization error accumulation in autonomous underwater vehicle (AUV) swarms due to underwater acoustic communication interruptions, this paper proposes a cooperative localization method that integrates Long Short-Term Memory (LSTM) prediction and factor graph optimization. During the real-time stage, each AUV uses a trained LSTM to predict observations, ensuring the Unscented Kalman filter (UKF) maintains continuous state estimation during interruptions and mitigates error accumulation. During the post-processing stage, a factor graph comprising motion model factors, cooperative observation factors, and LSTM prediction factors is constructed on the AUV swarm master node. By adaptively switching factor types based on communication status, global nonlinear optimization is performed on the AUV states. Simulation results show that compared with UKF + LSTM, the proposed method reduces the Average Localization Error (ALE) by 55% and the Root Mean Square Error (RMSE) by 60%; compared with the Rauch–Tung–Striebel (RTS) smoothing algorithm, it reduces the ALE by 36% and the RMSE by 44%. This fully verifies that the strategy combining real-time state maintenance and post-processing global optimization can more effectively correct AUV localization errors in communication-interrupted regions. Experiments under different communication interruption durations further confirm the robustness of the proposed algorithm, with the maximum error-to-range ratio remaining below 0.2% of the range. Full article

Consolidated bioprocessing (CBP) is difficult to monitor because enzyme production, lignocellulose degradation, sugar release, and fermentation occur simultaneously under sparse measurement, feedstock variability, and plant–model mismatch conditions. This study proposes a computational sensor-set design framework for digital-twin-assisted CBP monitoring. A five-state virtual plant, consisting of active biomass, cellulolytic enzyme activity, residual insoluble substrate, soluble sugar, and ethanol, was used to evaluate all 16 ethanol-mandatory measurement packages formed from ethanol, sugar, biomass, enzyme, and residual-substrate proxy channels. Candidate sensor sets were assessed using finite-difference output sensitivities, Fisher-information-based state-observability and parameter-identifiability analyses, eigenvalue and parameter-correlation diagnostics, and paired Monte Carlo unscented Kalman filter soft-sensing reconstruction. Within the tested five-state virtual-plant benchmark and with the specified excitation schedule, noise assumptions, burden indices, and scoring objective, ethanol-only sensing provided the weakest support for state-aware CBP digital-twin reconstruction. At a $6\mathrm{h}$ sampling interval, the state-observability log-pseudodeterminant increased from 4.18 with ethanol-only sensing to 8.56 after adding soluble sugar and to 16.42 with full-proxy monitoring. The ethanol–sugar–biomass–substrate package also gave strong reduced state-observability performance, with log-pseudodeterminants of 15.12, 13.76, and 12.51 at 6, 12, and $24\mathrm{h}$, respectively. Biomass and enzyme proxies contributed strongly to parameter learning, and the ethanol–sugar–biomass–enzyme package gave the strongest active parameter-identifiability performance, with log-pseudodeterminants of 10.82, 9.06, and 6.67 at 6, 12, and $24\mathrm{h}$, respectively. In the paired soft-sensing analysis, full-proxy monitoring reduced the mean latent-state RMSE from 1.1899 to 0.3756, followed by ethanol–biomass–enzyme–substrate with 0.3843 and ethanol–sugar–biomass–substrate with 0.4121. The primary aggregate ranking identified ethanol–sugar–biomass–substrate as the best overall package, with a sensor-value score of 0.8432 and a burden index of 7.0, followed by full-proxy monitoring with a score of 0.8173 and a burden index of 10.0. Robustness tests showed that ethanol–sugar–biomass–substrate remained top-ranked under uniform noise scaling, full UKF missingness, delay and bias stress test conditions, most scoring-weight scenarios, and all tested sensor-specific burden workflows. Full-proxy monitoring remained a close competitor under independent sensor-specific noise variation conditions and became top-ranked for some alternative operating trajectories. The proposed framework provides a simulation-based method for prioritizing informative measurement packages before implementing CBP digital twins in laboratory and pilot-plant settings. Full article

Accurate state-of-charge (SOC) estimation is essential for the safe and efficient operation of lithium-ion batteries. However, the weak voltage observability of lithium iron phosphate (LFP) batteries within the voltage plateau region limits the accuracy of conventional voltage-based methods. To address this issue, an electro–mechanical information fusion framework for SOC estimation is proposed. Fiber Bragg grating (FBG) sensors were employed to simultaneously measure the surface strain and temperature of prismatic LFP batteries. Experimental results showed that the strain signal exhibited a stronger correlation with SOC than the voltage signal, with an average absolute correlation coefficient of 0.92. A Thevenin equivalent circuit model combined with an adaptive forgetting factor recursive least squares (AFFRLS) algorithm was established for online voltage modeling, while a Mamba-based strain model was developed to capture the nonlinear temporal relationship between multidimensional sensing data and battery strain. The two models were further integrated with adaptive unscented Kalman filters (AUKFs) and fused through a dual-layer adaptive weighting strategy. Experimental results under the five operating conditions considered in this study demonstrated that the proposed method achieved average RMSE and MAE values of 0.98% and 0.80%, respectively, outperforming standalone voltage- and strain-based methods. Full article

In Australia’s National Electricity Market (NEM), community-scale battery energy storage systems (BESS) operate under five-minute price volatility and frequent negative pricing. However, unmeasured internal states and degradation processes constrain the effectiveness of rule-based and simplified optimisation methods for real-time arbitrage. To address these challenges, this study proposes a data-driven digital-twin framework for real-time management of a 1 MW/4 MWh grid-connected community BESS. The framework integrates a control-oriented single-particle model (SPM), an Unscented Kalman Filter (UKF)-based estimation layer for state-of-charge (SOC), state-of-health (SOH) and internal-state estimation and a degradation-aware nonlinear model predictive control (NMPC) strategy. Within this architecture, the SPM provides an interpretable electrochemical representation, the estimation layer reconstructs internal states from measurable signals, and the NMPC performs five-minute rolling arbitrage subject to voltage, power, and SOC constraints while accounting for ageing-related costs and ramp penalties. Simulation case studies based on high-volatility daily price profiles from four NEM regions indicate that the proposed framework can coordinate arbitrage-oriented dispatch, constraint-aware operation, and degradation-related cost consideration under the tested conditions. These results suggest the potential of the SPM–UKF–NMPC digital-twin architecture for supporting real-time community-scale BESS management, while further validation under forecast uncertainty and hardware or field conditions remains necessary. Full article

The Global Navigation Satellite System (GNSS) provides meter-level positioning in open environments, but its accuracy degrades severely in dense urban areas due to signal blockage and multipath effects. To address this problem, this paper proposes a hierarchical collaborative fusion positioning method based on GNSS, 5G, and the Inertial Navigation System (INS) with cross-source observation quality assessment. The proposed method integrates dual-domain error suppression, adaptive-shrinkage Unscented Kalman Filter (UKF) estimation, and observation-quality-aware adaptive weighting to mitigate systematic bias, random gross errors, and observation degradation. Unlike conventional fixed-weight or single-source-quality fusion schemes, the proposed method jointly combines gross-error detection, residual-driven covariance shrinkage, and adaptive weight regulation in a unified framework. Experiments were conducted in open outdoor, semi-occluded outdoor, and fully occluded indoor scenarios. The proposed method achieved a horizontal RMSE of 1.61 m in the semi-occluded outdoor environment. Compared with the the long short-term memory (LSTM)-aided UKF baseline, the positioning RMSE was reduced by 32.4%, and the positioning interruption rate was reduced by 49.5%. Full article

Accurate vehicle state estimation is a critical prerequisite for electric vehicle motion control, yet its performance is highly sensitive to deviations in inertial parameters. Variations in vehicle mass and moment of inertia caused by changing loads can lead to model mismatch, thereby degrading the accuracy and robustness of state estimation. To this end, this paper proposes a hierarchical collaborative estimation framework that integrates the Maximum Correntropy Adaptive Unscented Kalman Filter (MCAUKF) with a Physics-Informed Neural Network (PINN) for inertial parameter identification and key state estimation in electric vehicles. The upper layer employs MCAUKF for robust online identification of unknown inertial parameters, such as vehicle mass and moment of inertia. The lower layer develops a PINN-based state estimator that incorporates physical constraints by embedding the coupled dynamic residuals of longitudinal, lateral, and roll motions into the supervised learning process, thereby enabling high-precision real-time estimation of key dynamic states, including yaw angle, longitudinal velocity, and roll angle. Simulation results demonstrate that the proposed method can effectively achieve coordinated estimation of inertial parameters and key states under varying load conditions and complex maneuvering scenarios, significantly improving overall estimation accuracy and robustness. Full article

Long propagation delays, severe multipaths, and narrow bandwidths make feedback-based link adaptation impractical in UWA channels at kilometer ranges, so we replace the feedback step with a prediction step. The transmitter runs a two-layer coded OFDM link in which Polar codes handle bit errors, and Raptor fountain codes handle packet erasures, with the Raptor overhead (OH) as the only real-time knob. The OH is picked from a lookup table indexed by three quantities the receiver can estimate online: SNR, RMS delay spread, and Doppler frequency. Two CSI predictors feed that table: Temporal Multiple Sparse Bayesian Learning (TMSBL), which exploits delay-domain sparsity, and the Square-Root Unscented Kalman Filter (SRUKF), which tracks per-subcarrier variations. We evaluate the system in five channel environments (AWGN, Rayleigh, K-distribution, Bellhop ray-tracing, and synthetic proxies parameterized from the KAM11 and WATERMARK sea-trial statistics). Across the nine Bellhop scenarios, the adaptive link’s throughput gain over a fixed-OH ($\mathrm{OH}=1.5$) baseline at SNR $=4$ dB spans roughly $-4\%$ to $+30\%$, with the largest benefit in the marginal short-range cases (shallow 500 m, $+30\%$) where the fixed baseline is most over-provisioned and near-parity elsewhere. The scheme’s principal benefit is collapse prevention, tracking the Oracle within the safety margin and avoiding the throughput collapse the fixed baseline suffers at low SNRs. This effect is specific to the physically structured Bellhop channels; in the homogeneous Rayleigh and K-distribution channels, both schemes enter deep outage at very low SNRs, so it is not a universal guarantee. A 1000-trial high-resolution Rayleigh campaign sharpens the head-to-head between predictors: at SNR $=4$ dB, SRUKF + OH reaches PER $0.048$ (95% Wilson CI $[0.036, 0.063]$) and TMSBL + OH reaches $0.071$ ($[0.057, 0.089]$), and at SNR $=12$ dB, their throughputs ($0.748$ and $0.746$) are statistically indistinguishable from each other (95% Wilson halfwidth $\sim \pm$0.014) and lie close to the Oracle’s $0.768$ (within $\sim 0.02$). The two predictors therefore occupy overlapping operating regions once the safety margin is matched, and a sparsity-dependent tendency (TMSBL in sparse multipath, SRUKF in dense multipath) appears only in physically structured channels and only at the $n=100$ screening level, where it is not statistically resolved and would benefit from higher-trial confirmation. A finite-blocklength check confirms that CA-SCL-decoded Polar codes at $N=128$ stay within $0.5$ dB of the Polyanskiy normal approximation, which makes Polar a sensible inner code at UWA block lengths. Full article

This paper presents the design and implementation of a field-programmable gate array (FPGA)-based System-on-Programmable-Chip (SoPC) architecture for sensorless speed control of permanent magnet synchronous motor (PMSM) drives. To enable real-time execution of the computationally intensive estimation stage, a parallelized Unscented Kalman Filter (UKF) is proposed for the joint estimation of rotor speed, position, and load torque. Unlike traditional sequential processor-based UKF implementations, the proposed parallel architecture simplifies the iterative process and significantly reduces computational latency and hardware resource utilization while preserving high estimation fidelity. This transformation reduces the number of sequential dependency stages within one estimation cycle and enables simultaneous execution of matrix operations using dedicated FPGA resources, thereby decreasing effective iteration latency. The complete control system comprises current regulators, a coordinate transformation module, a proportional–integral (PI) speed controller, and auxiliary functional blocks—all fully integrated within a single SoPC. The UKF estimator and control components are described using a hardware description language (HDL), enabling efficient hardware-level parallelism and real-time execution. The proposed system is validated through co-simulation and experimental verification on a Xilinx ZCU102 platform driving an inverter-fed PMSM. The results confirm correct real-time operation of the proposed architecture and demonstrate its feasibility for FPGA-based sensorless motor drive implementation. A detailed quantitative comparison with a fully sequential FPGA-based UKF implementation is identified as future work to further substantiate the reported latency reduction. Full article