Browse Topic: Machine learning
Space membrane structures, prized for their lightness, affordability, and ability to fold and unfold, hold immense promise for missions like deep space exploration, high-definition Earth observation, and advanced satellite communications. However, these structures are typically large-size, ultra-lightweight, highly flexible, and have very low damping. This leads to complex dynamic behavior featuring significant geometric nonlinearity, strong coupling between rigid and flexible motions, and uncertain parameters. Consequently, creating accurate yet simple mathematical models for designing controllers is exceptionally difficult, and traditional control methods that rely on such models often fall short. To tackle this problem, this study focuses on a membrane antenna spacecraft equipped with cable actuators, aiming to suppress its vibrations during attitude maneuvers. Using the Deep Deterministic Policy Gradient (DDPG) algorithm, an intelligent vibration controller based on the Actor-Critic architecture is designed. This approach is data-driven, meaning it learns effective control strategies directly through interaction with the system’s environment, bypassing the need for a precise pre-defined mathematical model. Simulation results show that the controller demonstrated superior vibration suppression and strong adaptability across several key scenarios.
The rapid evolution of satellite communication systems enables direct network access for terrestrial users via portable devices, but also introduces critical security challenges in access control. In particular, identity hijacking attacks allow malicious users to exploit legitimate access channels, leading to inefficient occupation of scarce satellite resources. Existing approaches mainly depend on trust evaluation or heuristic-based decision strategies, which are inadequate for adapting to dynamic environments. To address this issue, this paper proposes a trust-aware deep reinforcement learning (DRL) framework for satellite access control, which jointly considers user behavior evaluation and decision optimization. Specifically, a Bayesian trust evaluation model is first developed to quantify the credibility of user behavior by incorporating historical interaction records as prior information. Unlike conventional static trust schemes, an adaptive mechanism is introduced to enhance sensitivity to behavioral dynamics. Building upon the trust assessment, the satellite access process is formulated as a Markov Decision Process (MDP), where system states integrate channel conditions, user trust levels, and service demands. A Deep Q-Network (DQN) is then employed to learn optimal admission and scheduling policies that maximize long-term system utility under resource constraints. Simulation results demonstrate that the proposed method significantly outperforms conventional greedy and threshold-based strategies. In particular, it effectively mitigates the impact of malicious behaviors and improves system utility by over 80%, while achieving more efficient resource utilization in dynamic environments.
This paper details the successful scaling demonstration of a comprehensive supply chain screening process for commercial off-the-shelf (COTS) motherboard subassemblies used in tactical servers for naval applications. Our approach leverages Power Fingerprinting (PFP) technology, which uses unintended analog emissions and machine learning to provide independent, non-destructive, and scalable integrity assessment of microelectronics. The primary goal of the effort was to demonstrate the effectiveness and scalability of the PFP screening process without disrupting or delaying the manufacturing workflow. The screening successfully detected hardware and firmware modifications and identified two cases of abnormal behavior: unusual BIOS power reset and elevated CPU sensor readings on two motherboard subassemblies. Following our quality control forensic analysis, we determined the root cause of these anomalies and their potential impact on the host platform.
Unmanned Aerial Systems (UAS) pose a growing threat on the modern battlefield, demanding rapid detection and characterization capabilities for the warfighter. Existing single-model solutions are inadequate for Counter-UAS (C-UAS), as they struggle across varying ranges and cannot provide detailed contextual information beyond bounding boxes. We present ZEUS (Zero-shot Explainable Universal Segmentation), a multi-model detection and recognition system that integrates several machine learning approaches. ZEUS employs a high-performance UAS detector trained on synthetic, internally collected, and open-source datasets, with real-time capability demonstrated on edge hardware across both electro-optical and infrared modalities. For classification, ZEUS uses a zero-shot approach: detected UAS are segmented and compared against a library of 3D reference models rendered at various poses, enabling identification of new UAS types without retraining. This methodology additionally provides UAS pose and range estimates critical for threat assessment and engagement decisions.
Artificial Intelligence (AI) and Machine Learning (ML) are increasingly transforming Computer-Aided Engineering (CAE) workflows by enabling faster design iterations and reducing computational costs. This paper presents the application of Ansys SimAI and Ansys GeomAI in modelling an automotive side impact scenario using high-fidelity data from LS-DYNA simulations. Two AI models are trained on datasets with systematically varied parameters: one encompassing pole impact position and door beam configurations, and another focusing on rocker panel reinforcements. Both models exhibit strong predictive performance, reliably capturing deformation patterns and force-time histories for previously unseen configurations. The datasets are subsequently merged to train a comprehensive surrogate model capable of simultaneously representing variations in pole position, door beam geometry, and rocker reinforcement design, demonstrating robust generalization across a multidimensional design space. To address the emerging bottleneck of geometry creation, GeomAI’s geometry exploration functionality is employed to generate new rocker reinforcement geometries from existing ones, which are then rapidly validated using the pre-trained surrogate model. The results confirm that LS-DYNA simulations can be leveraged effectively to build AI models that dramatically reduce design exploration time. With SimAI and GeomAI in the loop, CAE workflows can evolve from simulation-driven design toward AI-augmented autonomous engineering, where geometry generation, simulation, validation, and optimization converge into an intelligent closed loop.
Drum brake systems are becoming increasingly important in electric vehicles (EV) and purpose-built vehicles due to cost competitiveness and EURO-7 particulate emission regulations. Despite this trend, drum brake friction behavior remains incompletely characterized due to its dependence on multiple coupled variables: temperature history, braking conditions, and component interactions. To address this gap, this study presents a method for developing a time-series friction torque prediction model using the Mixed-effects Random Forest (MERF) machine learning framework. Time-series data collected from sensors during drum brake dynamometer tests were analyzed to identify the key variables that govern the friction torque. Significant inputs were selected through Exploratory Data Analysis (EDA), considering test-to-test variability and potential mixed effects, and were then used to train and tune the MERF model. Model performance was evaluated by comparing predicted friction torque with measured torque, and prediction error was quantified by using Mean Absolute Error (MAE) to check whether predicted model is reliable. The proposed prediction model demonstrates a high level of agreement with experimental measurements, confirming that the MERF approach can effectively capture the non-linear and transient characteristics of drum brake friction torque from time-series sensor signals. These results indicate that friction torque estimation is feasible using only sensor signals already available from conventional test instrumentation, without additional dedicated sensors. This capability is expected to support broader applications, including brake performance prediction for vehicles equipped with drum brakes and enhanced simulation of drum brake thermal performance across operating conditions.
It is hardly a new trend for on road, vehicle intensive tuning and testing of chassis control features such as Anti-Lock Brakes, Traction Control, and Electronic Stability Control to move away from vehicle testing and towards non-vehicle test platforms such as Hardware-In the Loop (HIL) simulations and even further into pure math-based simulations. However, a significant acceleration of these activities has been occurring recently in the automotive industry, reducing or eliminating calibration time on vehicles and amplifying the demand for highly representative, non-vehicle test platforms to validate and even calibrate chassis controls features. In current state of the art HIL simulation, the input (brake pressure) to output (brake torque) of each wheel brake in a vehicle’s brake system is modeled relatively simplistically, including at most pressure and brake temperature sensitivities, usually in lookup table form. Each brake corner contains over 20 different friction interfaces, which in turn can cause hysteretic behavior (a difference in the output for a given input, depending on whether the brake is applying or releasing against the hysteretic friction). This hysteresis is neglected in most state of the art HIL simulations. Past studies by General Motors have shown that the importance of brake corner hysteresis in vehicle level, customer facing performance of chassis controls features can range from inconsequential to significant. With the crescendo-ing demand for high quality non-vehicle based methods for assessing chassis controls function, the effect of hysteresis is no longer academic. The present study starts with HIL based simulations, establishing the effect of brake corner hysteresis on one of the most visible chassis controls behaviors. An inertia dynamometer-based test was developed to exercises the subject brake corners through apply and release cycles, thus enabling any hysteretic behavior to be observed and characterized. Machine Learning models were trained with these data to represent brake corner hysteretic behavior and then deployed into an HIL simulation rig. The impact of these models – representing brake corner hysteretic behavior – was characterized for straight line stopping distance on low, medium, and high coefficient road surfaces.
Unmanned Underwater Vehicles (UUVs) operate in complex and uncertain environments, which require a suitable controller. While traditional PID controllers are widely used, they often have slow response speed and inadequate disturbance rejection, particularly under complex and uncertain conditions. To overcome these shortcomings, this paper introduces the DDPG-DLPID, an adaptive motion controller, including a Deep Deterministic Policy Gradient (DDPG) reinforcement learning that can acquire the parameters of PID controllers. In this paper, we design two loops: the inner loop handles velocity regulation, and the outer loop handles position and attitude. By using DDPG, the system can efficiently adjust the PID parameters of both loops in real time, allowing it to effectively adapt to environmental changes and achieve optimized requirements. To evaluate the controller, we design the following scenarios, including straight-line and complex path-following tasks. Compared with single-loop PID and dual-loop PID controllers, the proposed DDPGDLPID approach achieves faster response and higher tracking accuracy, while substantially reducing tracking errors under interference conditions. Physical experiments under three conditions-straight-line voyage, attitude maintaining, and depth control-were further carried out to validate the strategy’s real-world applicability. Experimental data confirm that DDPG-DLPID has better performance when compared with both traditional PID and dual-loop PID controllers across all test scenarios.
Steady advancement is observed in global research on eco-friendly and sustainable transportation. Rapid technological evolution of hybrid electric vehicles (HEVs) is documented. Lower overall noise output and more compact structures are achieved in HEV engines relative to conventional internal combustion engines. The perceptibility of harmonic impulsive sounds is significantly enhanced by these design characteristics. A close correlation is observed between these acoustic phenomena and negative human auditory perceptions. These events are treated as a core focus for HEV noise, vibration, and harshness optimization. Accurate quantification of harmonic impulsive sounds is not achieved by conventional objective indicators. A favorable balance between reliability and accuracy is not established by existing subjective prediction models. Practical engineering applications of these methods are severely restricted. A novel objective quantification method for harmonic impulsive sounds is proposed in this study. The method is established based on time–frequency masking theory and tonal strength. Bench tests in a semi-anechoic chamber and subjective evaluation experiments with standardized rating scales are performed for data collection. Collected sound signals are decomposed through an integrated approach of wavelet transform and variational mode decomposition. Targeted feature extraction is completed for harmonic impulsive sounds. A quantitative index incorporating human auditory temporal and frequency masking effects is developed. The proposed index exhibits a significantly stronger correlation with subjective evaluation results than traditional objective metrics, confirming its superior ability to reflect actual perceived sound quality. An interval prediction model for sound quality evaluation is established based on support vector machines and kernel density estimation. Traditional objective metrics and the proposed index are introduced as key input parameters. Effective and reliable prediction of HEV engine noise subjective satisfaction is achieved by the model.
A machine-learning strategy has generated a new class of ultra-high strength and ductility steel for 3D printing that costs less, resists rust, and requires only a fraction of the usual processing time.
With the development of controlled nuclear fusion technology, the tokamak device, as the most promising magnetic confinement fusion reactor for advanced engineering applications, requires remote maintenance of its internal components, which has become a key factor affecting both operational efficiency and safety. As a critical component directly exposed to high-temperature plasma, the divertor target plate needs to be periodically replaced and carefully maintained to ensure stable and reliable reactor operation. However, this region is subject to extreme conditions, including high temperature, high vacuum, and intense radiation, making conventional manual maintenance infeasible. This necessitates the development of intelligent and automated teleoperation systems. To address the automated assembly and disassembly requirements of divertor target plates, this study designs an integrated target plate actuator comprising key functional units: a positioning module, a screwing module, a quick-change module, and a passive compliance structure. The actuator achieves rapid and precise alignment with target plate holes, accommodates bolts of different specifications, and exhibits excellent impact resistance. Furthermore, stiffness and mechanical analyses, supported by finite element simulations, verify the actuator’s safety and reliability under high loads and impact forces. To further enhance operational performance, a segmented disassembly and assembly control strategy based on reinforcement learning is proposed, enabling the actuator to adaptively handle torque variations and ensure precise and stable bolt operations. The results demonstrate that the proposed actuator and control strategy significantly improve the accuracy, stability, and efficiency of target plate operations under complex working conditions, providing a reliable solution for automated divertor maintenance in tokamak devices.
The inconsistency in bearing data distributions under diverse conditions often affects the representations of the faulty data and leads to indistinct decision boundaries and even negative transfer resulted from overlapping class distributions, greatly limiting the accuracy of the diagnosis model. To cope with the challenge, a pseudo-label-guided dual-supervised alignment (PDSA) method is developed for bearing fault diagnosis across diverse operating scenarios in this paper. To address the fixed alignment strategy issue, an adaptive distribution alignment layer is incorporated to ResNet18 to achieve dynamic data distribution alignment under varying condition, To enhance classification performances, a dual-supervised mechanism, comprising shallow-layer supervised contrastive learning is introduced through target domain pseudo-labels in target domain and deep-layer regularization class consistency. Experiments on two publicly available bearing datasets demonstrated this model realizes refined class-level alignment, strengthens fault states representation, and shows notable superiority in both accuracy and robustness.
Items per page:
50
1 – 50 of 1612