<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Robust learning control for autonomous vehicle with network delays and disturbances</title>
      <link>https://trid.trb.org/View/2596542</link>
      <description><![CDATA[AbstractThis paper deals with a robust learning nonlinear model predictive control (RL-NMPC) scheme under time-varying delays and disturbances. It is well known that the in-vehicle network has considerable advantages over the traditional point-to-point communication. However, on the other hand, these technologies would also induce the probability of time-varying delays, which would be a hazard in the active safety of over-actuated autonomous vehicles (AVs). To enjoy the advantages and deal with in-vehicle network delays and external disturbances, a robust learning nonlinear model predictive control (RL-NMPC) scheme is proposed. First, the machine learning (Support Vector Machine called SVM) method is adopted to train delayed measurement signals and disturbances. Then, according to the predictions of the SVM and corrupted sensory signals, the Unscented Kalman filter (UKF) is applied to acquire accurate predictions of the vehicle motion states. Furthermore, the NMPC scheme is used to generate real-time control signals by solving an open-loop optimization problem. The main purpose of the addressed problem is to design a robust learning controller to ensure that the AVs can track the desirable path and run smoothly suffering network delays and disturbances. Finally, simulations with a full-vehicle model are carried out to show the effectiveness of our proposed control scheme.]]></description>
      <pubDate>Wed, 06 May 2026 15:21:26 GMT</pubDate>
      <guid>https://trid.trb.org/View/2596542</guid>
    </item>
    <item>
      <title>A framework for a digital twin for a Kamsarmax bulk carrier's maneuvering to avoid grounding consists of deep learning ship dynamic models and model predictive controller</title>
      <link>https://trid.trb.org/View/2697763</link>
      <description><![CDATA[The prediction and control of bulk carrier movements in waterways are essential for safe navigation and the development of autonomous surface vessels. This paper introduces a deep learning–based digital twin framework for a Kamsarmax bulk carrier, integrating surrogate models within a model predictive controller (MPC) to determine control inputs for path-following and grounding avoidance. A framework was introduced and ship motions and positions were predicted using autoregressive deep learning models, validated through mean absolute error, probability density comparisons, and Kullback–Leibler divergence (KL). The result from the auto regressive model showed good accuracy with predicting surge and sway speeds, yet it underestimated the proportional change of heading. The MAE from the surge and sway speeds were 0.175 and 1.052 respectively while the heading produced a MAE of 0.751. A trust-region optimization algorithm was applied to minimize trajectory error within the MPC. The results from the MPC showed that the model was able to follow the desired path; however, greater attention is needed to the sampling frequency in which the data is collected as higher sampling frequency will improve the accuracy of the deep learning dynamic models in predicting surge speed and in capturing the proportional change of the desired heading resulting from adjustments to the rudder angle. The reason for that is the motions are nonlinear and the current sampling frequency leads to the loss of important information. Finally, the developed model can be used within a real-time system onboard ship or within a simulation environment.]]></description>
      <pubDate>Thu, 30 Apr 2026 16:39:45 GMT</pubDate>
      <guid>https://trid.trb.org/View/2697763</guid>
    </item>
    <item>
      <title>Energy-Efficient Wheel Torque Distribution for Heavy Electric Vehicles With Adaptive Model Predictive Control and Control Allocation</title>
      <link>https://trid.trb.org/View/2658695</link>
      <description><![CDATA[This paper proposes an energy efficient hierarchical wheel torque controller for a 4 × 4 heavy electric vehicle equipped with multiple electric drivetrains. The controller consists of two main components: a global force reference generator and a control allocator. The global force reference generator computes motion requests based on steering wheel angle and longitudinal acceleration inputs, while adhering to actuator and tire force constraints. For this purpose, a linear time-varying model predictive controller (LTV-MPC) is employed to minimize the squared errors in yaw rate and longitudinal acceleration over a short prediction horizon. Concurrently, the controller dynamically identifies safe operating limits based on current driving conditions. These limits are then used to adjust the state cost weights dynamically, thereby improving the effectiveness of the MPC cost function. The control allocator (CA) subsequently distributes the force demands from the global reference generator among the electric machines and friction brakes. This allocation process minimizes instantaneous power losses while respecting actuator and tire force constraints. To further enhance energy efficiency, the method leverages the heterogeneous nature of the electric machines by minimizing not only operational power losses but also idle losses (power losses at zero torque), ensuring safe vehicle operation. The proposed strategy is evaluated using a high-fidelity vehicle model under various driving scenarios, including low-friction surfaces and near-handling-limit conditions. Simulation results demonstrate that dynamically varying state cost weights in conjunction with safe operating limits significantly improves vehicle performance, enhances energy efficiency, and reduces driver effort.]]></description>
      <pubDate>Wed, 29 Apr 2026 16:50:11 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658695</guid>
    </item>
    <item>
      <title>Distributed adaptive signal control optimization and stability analysis based on nonlinear small-gain theorem</title>
      <link>https://trid.trb.org/View/2659384</link>
      <description><![CDATA[Network-wide signal control optimization is of practical importance to shorten and stabilize travel time, improve productivity, enhance energy consumption efficiency, mitigate congestion, and reduce vehicle emissions. In this study, a deep learning–empowered distributed control strategy is developed to adaptively optimize network-wide traffic signal control coordination. To simplify the problem formulation and enhance its applicability, the entire traffic system is decomposed into multiple areas, and multilayer perceptron concepts are used to formulate traffic control system operations in each area. The distributed deep learning, velocity-based model predictive control (MPC) strategy is designed to optimize traffic signal coordination. Furthermore, a gain-scheduling control model is developed to linearize each learned nonlinear system around its most recent operating status, and then a distributed MPC controller is applied to the linearized systems. Simulation results demonstrate that the proposed control strategy can effectively reduce travel time by 15.1% compared with fixed-time control plans and by 8.0% compared with a decentralized control plan. This study is the first research effort to integrate the deep learning framework and multiagent MPC to optimize traffic control coordination. Moreover, a sufficient condition is theoretically formulated for the bounded-input, bounded-output stability of the closed-loop, large-scale traffic system based on the nonlinear small-gain theorem.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:06 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659384</guid>
    </item>
    <item>
      <title>Event-triggered robust MPC for underactuated ships path following with roll constraint</title>
      <link>https://trid.trb.org/View/2586738</link>
      <description><![CDATA[This study proposes a novel path following control strategy for underactuated ships to address roll instability, actuator wear, and path following inaccuracy. The method integrates adaptive line-of-sight (ALOS) guidance, a finite-time extended state observer (FTESO), and event-triggered robust model predictive control (ET-RMPC). ALOS dynamically adjusts the acceptable radius at waypoints, enhancing maneuverability. The FTESO estimates ship motion states and unknown disturbances, while a radial basis function neural network (RBFNN) compensates the effect of sideslip angle. The ET-RMPC controller incorporates input and roll constraints, along with infinite time domain performance indices, into an optimization framework reformulated as a convex problem using linear matrix inequalities. An event-triggered mechanism is introduced to mitigate actuator wear. Compared with conventional RMPC, the proposed method has a 45.45 % reduction in roll and a 45.57 % reduction in communication efficiency. Lyapunov stability and homogeneity theories guarantee closed-loop stable and finite-time convergence. Simulation results demonstrate the proposed method’s effectiveness and robustness.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2586738</guid>
    </item>
    <item>
      <title>Research on effective trajectory planning, tracking, and reconstruction for USV formation in complex environments</title>
      <link>https://trid.trb.org/View/2631537</link>
      <description><![CDATA[Unmanned surface vehicle (USV) formation holds significant promise for enhancing marine operations through coordinated actions. However, realizing their potential necessitates overcoming key challenges in motion planning and control. This study explores an effective framework for USV formation trajectory planning, deformation, and tracking through Formation Prediction Trajectory Approach (FPTA) and Model Predictive Control (MPC) algorithms. The FPTA integrates heading angle optimization to obtain smooth and safe trajectories of both fixed-structure and leader-follower formations while accounting for USV dynamic constraints. A model predictive controller is designed to accurately track the planned trajectory, thereby facilitating constraint handling and formation control. To address formation reconstruction issues, the lead and lag tracking strategies are proposed to efficiently adjust configurations while avoiding collisions. Simulations compare our integrated planning and control strategies against conventional methods and examine the navigation of fixed-structure and leader-follower formations in the presence of environmental interference and control uncertainty.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2631537</guid>
    </item>
    <item>
      <title>Hybrid-driven meta-learning for wind field prediction in unmanned sailboat applications</title>
      <link>https://trid.trb.org/View/2634976</link>
      <description><![CDATA[As an observation platform highly sensitive to variations in marine wind fields, unmanned sailboats require accurate wind field modeling to ensure navigational safety and performance. However, existing modeling approaches exhibit limited adaptability across different sea regions and voyage conditions. To address this issue, this paper proposes a hybrid-driven meta-learning method, which models three-dimensional wind speed data and indirectly infers wind direction. The approach integrates multiple base model architectures and input strategies, and introduces a Continuous Interval Prediction Weighted Performance Index (CIPWPI) to quantify the impact of wind direction on unmanned sailboat navigation. Validation based on real voyage data from a new sea area shows that the proposed method achieves a root mean square error (RMSE) of 0.0735 in wind speed prediction, representing a reduction of approximately 29.5 % compared to the best baseline model. In addition, the mean absolute error (MAE), correlation coefficient (CC), and coefficient of determination (R2) all outperform the baseline models significantly, demonstrating superior prediction accuracy and stability. For wind direction prediction, significant improvements are observed across multiple error thresholds (1°, 3°, 5°, 10°), with an overall performance gain of approximately 9.1 %. Notably, improvements under small-scale error thresholds (1° and 3°) are particularly significant, reaching 22.3 % and 14.3 %, respectively. These results confirm the method's robustness and generalization capability, offering more reliable wind field prediction support for unmanned sailboat applications.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2634976</guid>
    </item>
    <item>
      <title>Acceleration based control strategies applied to ROV station keeping under wave disturbance</title>
      <link>https://trid.trb.org/View/2638226</link>
      <description><![CDATA[When Remotely Operated Vehicles (ROVs) operate in shallow depths, wave disturbances introduce challenges to station-keeping control loops. To address this issue, this work proposes a novel control scheme for ROV dynamic positioning in the dive plane. The approach combines an Augmented Wave Filter (AWF) with acceleration-based control strategies: (i) Acceleration Feedback (AFB) and (ii) Acceleration Feedforward (AFF). The proposed controllers are compared to standard Proportional-Integral-Derivative (PID) control and Adaptive Model Predictive Control (A-MPC). The novel controller replicates the preemptive action of the A-MPC against wave disturbances while utilizing a simpler mathematical structure. Two variations of the AWF were tested: one based on the Extended Kalman Filter (EKF) and the other using the Unscented Kalman Filter (UKF). Simulations were conducted under various scenarios, including still water and sea states with significant wave heights of 1 and 2 m. Controller robustness to model parameter variations was assessed through Monte Carlo simulations. Results indicate that AFF and A-MPC outperformed the other strategies in terms of positioning error, motion oscillations, and drift damping with greater parametric robustness. Additionally, the wave filters performed well, removing wave-induced depth measurement noise and enabling offset-free control. The EKF converged faster in the case of mismatched initial conditions.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2638226</guid>
    </item>
    <item>
      <title>Risk-aware model predictive control framework for collision avoidance in unmanned surface vehicles</title>
      <link>https://trid.trb.org/View/2638204</link>
      <description><![CDATA[Unmanned Surface Vehicles (USVs) are critical for maritime missions. Collision avoidance emerges as the core challenge for USV safety in complex maritime environments, requiring simultaneous handling of dynamic obstacles, uncertain sea environment, and COLREGs-compliant maneuvers. Existing collision avoidance methods primarily focus on evaluating immediate risk without predicting how risks develop over time, which limits their effectiveness in handling complex, dynamic scenarios. To this end, this work proposes a risk-aware model predictive control framework that works in three steps: precise real-time risk assessment, predictive risk assessment, and objective dynamic optimization. First, we introduce a new collision risk parameter, which enhances the traditional Collision Risk Index by integrating the direction of relative displacement and the direction of the relative velocity. Next, we employ a ship dynamic model, augmented by environmental disturbance, to predict the future state of the USV and collision risk over a prediction horizon. Finally, a model predictive control framework is employed to integrate real-time and predicted risk information into an objective optimization function, which computes the optimal control forces. Experiments show that our approach reduces computation time, maintains high decision stability, and improves path efficiency and collision avoidance in different scenarios.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2638204</guid>
    </item>
    <item>
      <title>Nonparametric online modeling for differential-driven unmanned surface vehicle based on fuzzy C-means and PD-like event-triggered strategy</title>
      <link>https://trid.trb.org/View/2638163</link>
      <description><![CDATA[The modeling reliability of differential-driven unmanned surface vessel (USV) is greatly affected by the limited on-board computational resources, highly coupled nonlinear models, and internal and external variable factors. For balancing computational resources and prediction accuracy, this paper proposes an novel nonparametric online modeling method for dynamic model of differential-drive USV based on fuzzy C-means (FCM) and PD-like event-triggered (PDLET) strategy. Firstly, an improved sand cat swarm optimization (ISCSO) is proposed to optimize the hyperparameters of least squares-support vector machine (LS-SVM). Secondly, the FCM soft clustering algorithm is adopted to cluster the collected offline data, and then the optimized LS-SVM is adopted to establish offline models for different clusters. Thirdly, the velocity and acceleration errors are simultaneously considered to construct the PDLET strategy, which determines model update moments and better corrects the model. Further, the training data of each cluster is filtered by the designed data selection window to reduce prediction errors in model switching process. The offline experimental results show that the hyperparameters optimized by ISCSO have better predictive accuracy compared with the existing sand cat swarm optimization (SCSO), particle swarm optimization (PSO), and cuckoo search (CS). The online experimental results show that the proposed online method has better prediction accuracy and could significantly reduce modeling time and the number of model updates.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2638163</guid>
    </item>
    <item>
      <title>Adaptive obstacle avoidance algorithm for wave gliders in dynamic marine environments based on improved DAPF with multi-model prediction</title>
      <link>https://trid.trb.org/View/2607010</link>
      <description><![CDATA[Current obstacle avoidance algorithms for wave gliders (WGs) often neglect inherent steering constraints and employ fixed-parameter artificial potential field (APF), which limits adaptability. Additionally, existing algorithms typically assume that the speeds of obstacles are comparable to that of the WG, which is inconsistent with the real marine environment. To address these limitations, this paper proposes a fusion obstacle avoidance algorithm combining an improved dynamic prediction (IDP) collision model with a dynamic APF (DAPF), specifically designed for scenarios involving a single dynamic obstacle (DO). A multi-model hybrid prediction approach based on interactive multiple model (IMM) is used by the IDP for DO prediction, enabling robust adaptation to DO motion states. The DAPF introduces a speed-adaptive repulsion gain coefficient and yaw attraction field constraints through a dynamic elliptical repulsion field mechanism. Compared with improved APF and environmental improved APF (EAPF), simulation results show that IDP-DAPF can increase the minimum obstacle avoidance distance for high-speed obstacles by 36.2 % and reduce the navigation efficiency index by 40.01 %. Sea trials further validate the effectiveness of the proposed algorithm in real marine environments.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:19 GMT</pubDate>
      <guid>https://trid.trb.org/View/2607010</guid>
    </item>
    <item>
      <title>Application of Machine Learning for Geometric Alignment in Automotive Assembly</title>
      <link>https://trid.trb.org/View/2693544</link>
      <description><![CDATA[This paper investigates the application of Machine Learning (ML) techniques for the optimization of geometric alignment processes in industrial assembly, with a focus on automotive headlamp adjustment. The research objective is to evaluate the potential of data-driven models to enhance process accuracy, reduce variability, and minimize manual intervention. A diverse set of regression-based ML methods—including linear, kernel-based, tree-based, ensemble, and interpretable piecewise linear regression—is systematically compared. Models are trained and validated on real-world production data using 10-fold cross-validation and separate test sets. The results show that several ML models achieve high predictive performance, particularly for one of the three alignment targets, while prediction accuracy is more challenging for the other targets due to greater geometric variability. The study highlights both the benefits and limitations of applying ML in data-constrained assembly contexts. Beyond the specific automotive use case, the findings demonstrate the broader potential of adaptive, data-driven modeling to support intelligent process control in smart manufacturing environments.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:14 GMT</pubDate>
      <guid>https://trid.trb.org/View/2693544</guid>
    </item>
    <item>
      <title>A nonlinear model predictive control method for ship path following in waves</title>
      <link>https://trid.trb.org/View/2632865</link>
      <description><![CDATA[With the development of intelligent ships, the problem of ship path following has attracted much attention in recent years. Ship motion is inevitably affected by an external environment and thus the design of the ship control system in currents or waves, etc. has always been a challenging work. In the present work, a Nonlinear Model Predictive Control (NMPC) method is proposed for ship path following in waves. The NMPC controller can obtain the desired rudder angle of nonlinear ship system by solving an open-loop optimisation problem. The NMPC controller integrated with a Line-of-Sight (LOS) guidance is proposed and applied for a container ship which is controlled to follow two desired paths in both calm water and waves. The simulation results demonstrate the more effectiveness and greater anti-interference of the proposed NMPC method, compared with that the results by using a traditional Proportional-Derivative (PD) control method.]]></description>
      <pubDate>Mon, 27 Apr 2026 14:59:00 GMT</pubDate>
      <guid>https://trid.trb.org/View/2632865</guid>
    </item>
    <item>
      <title>Multi-timescale Optimization for Reversible Converter in DC Traction Power System Based on Model Predictive Control</title>
      <link>https://trid.trb.org/View/2632985</link>
      <description><![CDATA[In urban rail flexible traction power supply system (FTPSS), conventional energy-saving strategies for reversible converter (RC) predominantly rely on offline optimization with fixed parameters. However, inherent uncertainties in train operations, such as timetable deviations and stochastic load fluctuations, result in energy consumption volatility, rendering traditional approaches suboptimal. To address this, we propose a multi-timescale model predictive control (MPC) framework that integrates day-ahead scheduling and intraday rolling optimization. Second, we propose a novel data processing method for neural network training in the intraday to construct a neural network-based load prediction model, which is used as the model prediction control (MPC) input for rolling optimization. Validated on Qingdao Metro Line 11 datasets, the prediction model achieves a correlation coefficient (R2) value of 95.2%, and the mean squared error (MSE) is 0.078, outperforming conventional prediction methods. By integrating MPC-based rolling optimization with day-ahead scheduling, the proposed strategy improves the energy-saving rate by 2.00% over traditional offline optimization methods. Demonstrating robustness against timetable perturbations and load uncertainties.]]></description>
      <pubDate>Mon, 27 Apr 2026 14:58:59 GMT</pubDate>
      <guid>https://trid.trb.org/View/2632985</guid>
    </item>
    <item>
      <title>Ship propulsion modelling and optimisation in stern quartering seas</title>
      <link>https://trid.trb.org/View/2690477</link>
      <description><![CDATA[This paper investigates propulsion dynamics and performance optimisation for vessels operating in following seas using a 6-DoF time-domain simulation framework that combines a unified seakeeping and manoeuvring model with propulsion control. The work focuses on the behaviour of the propulsion system and the impact of wave disturbances, and studies propulsion controllers and the effect of thrust oscillations. Five control strategies are implemented, and shown that the combined shaft-speed/power scheme yields favourable engine operating points, while Model Predictive Control (MPC) attains the highest mean propeller efficiency with constrained power variability. For long-time-scale optimisation, a model-free Extremum Seeking Control (ESC) method is implemented to adjust the propeller pitch in real time. The results provide proof of concept that ESC converges to near-optimal operating conditions for propulsion efficiency and vessel speed within practical time scales. A novel approach is proposed to maximise the vessel speed while keeping the power constant, and we study how this approach is favourable relative to alternatives. Hydrodynamic analysis further shows that surge dynamics behave approximately as a linear low-pass filter at the investigated operating conditions, which explains the ESC convergence behaviour and links vessel dynamics to controller performance.]]></description>
      <pubDate>Mon, 27 Apr 2026 14:58:14 GMT</pubDate>
      <guid>https://trid.trb.org/View/2690477</guid>
    </item>
  </channel>
</rss>