Search Results (227)

Agentic AI systems, possessing capabilities for autonomous planning and action, show great potential across diverse domains. However, their practical deployment is hindered by challenges in aligning their behavior with varied human values, complex safety requirements, and specific compliance needs. Existing alignment methodologies often falter when faced with the complex task of providing personalized context without inducing confabulation or operational inefficiencies. This paper introduces a novel solution: a ‘superego’ agent, designed as a personalized oversight mechanism for agentic AI. This system dynamically steers AI planning by referencing user-selected ‘Creed Constitutions’—encapsulating diverse rule sets—with adjustable adherence levels to fit non-negotiable values. A real-time compliance enforcer validates plans against these constitutions and a universal ethical floor before execution. We present a functional system, including a demonstration interface with a prototypical constitution-sharing portal, and successful integration with third-party models via the Model Context Protocol (MCP). Comprehensive benchmark evaluations (HarmBench, AgentHarm) demonstrate that our Superego agent dramatically reduces harmful outputs—achieving up to a 98.3% harm score reduction and near-perfect refusal rates (e.g., 100% with Claude Sonnet 4 on AgentHarm’s harmful set) for leading LLMs like Gemini 2.5 Flash and GPT-4o. This approach substantially simplifies personalized AI alignment, rendering agentic systems more reliably attuned to individual and cultural contexts, while also enabling substantial safety improvements. Full article

Millimeter-wave frequencies are crucial for meeting the high-capacity, low-latency demands of 5G communication systems, thereby driving the need for compact, high-gain antenna arrays capable of efficient beamforming. This paper presents the design, simulation, fabrication, and experimental validation of a compact, high-efficiency 1 × 6 linear series-fed microstrip patch antenna array for 5G millimeter-wave communication operating at 28 GHz. The proposed antenna is fabricated on a low-loss Rogers RO3003 substrate and incorporates an integrated symmetric two-way microstrip power divider to ensure balanced feeding and phase uniformity across elements. The antenna achieves a simulated peak gain of 11.5 dBi and a broad simulated impedance bandwidth of 30.21%, with measured results confirming strong impedance matching and a return loss better than −20 dB. The far-field radiation patterns demonstrate a narrow, highly directive beam in the E-plane, and the H-plane results reveal beam tilting behavior, validating the antenna’s capability for passive beam steering through feedline geometry and element spacing (~0.5λ). Surface current distribution analysis confirms uniform excitation and efficient radiation, further validating the design’s stability. The fabricated prototype shows excellent agreement with the simulation, with minor discrepancies attributed to fabrication tolerances. These results establish the proposed antenna as a promising candidate for applications requiring compact, high-gain, and beam-steerable solutions, such as 5G mm-wave wireless communication systems, point-to-point wireless backhaul, and automotive radar sensing. Full article

Driver models, which provide mathematical or computational representations of human driving behavior, are crucial for intelligent driving systems by enabling stable and repeatable operations. However, existing models typically employ fixed weighting parameters to simulate preview delay, failing to reflect individual driver differences and real-time dynamic behaviors. This paper proposes a Brain-Memory Driver Model (BMDM) that emulates human brain memory mechanisms to dynamically adjust preview weights by integrating global path curvature, real-time vehicle speed, and steering torque. This emulation involves a three-stage process: capturing data in an Instantaneous Memory (IM) region, filtering data via a forgetting mechanism in a Short-Time Memory (STM) region to reduce scale, and retaining data based on correlation strength in a Long-Time Memory (LTM) region for persistent mining. By deploying a trained behavioral memory database, the model dynamically calibrates preview weights based on the driver’s state and real-time curvature variations under different road conditions. This enables the model to more accurately simulate authentic preview characteristics and improves its adaptability. Simulation results from an automated steering case study demonstrate that the improved model exhibits control performance closer to the real driving process, reproducing authentic steering behavior within the human–vehicle–road closed-loop system from an intelligent biomimetic perspective. Full article

The paper presents the innovative approach to a high-fidelity motorcycle riding simulator based on VR (Virtual Reality)-visualization, equipped with a Gough-Stewart 6-DOF (Degrees of Freedom) motion platform. Such a solution integrates a real-time tension sensor system as a source for highly realistic motion cueing control as well as the servomotor integrated into the steering system. Tension forces are measured at four points on the mock-up chassis, allowing a comprehensive analysis of rider interaction during various maneuvers. The simulator is developed to simulate realistic riding scenarios with immersive motion and visual feedback, enhanced with the simulation of external influences—headwind. This paper presents results of a validation study—pilot experiments conducted to evaluate selected riding scenarios and validate the innovative simulator setup, focusing on force distribution and system responsiveness to support further research in motorcycle HMI (Human–Machine Interaction), rider behavior, and training. Full article

In order to mitigate human-machine conflicts and optimize shared control strategy in advance, it is essential for the shared control system to understand and predict driver behavior. This paper proposes a method for predicting driver steering intention with a CNN-GRU hybrid machine learning model. The convolutional neural network (CNN) layer extracts features from the stochastic driver behavior, which is input to the gated-recurrent-unit (GRU) layer. And the driver’s steering intention is forecasted based on the GRU model. Our study was conducted using a driving simulator to observe the lateral control behaviors of 18 participants in four different driving circumstances. Finally, the efficiency of the suggested prediction approach was evaluated employing long-short-term-memory, GRU, CNN, Transformer, and back propagation networks. Experimental results demonstrated that the proposed CNN-GRU model performs significantly better than baseline models. Compared with the GRU network, the CNN-GRU network reduced the RMSE, MAE, and MAPE of the driver’s input torque by 33.22%, 32.33%, and 35.86%, respectively. The proposed prediction method also possesses adaptability to different driver behaviors. Full article

Fescue toxicosis (FT) is a mycotoxin-related disease caused by the ingestion of tall fescue, naturally infected with the ergot alkaloid (EA)-producing endophyte Epichloë coenophiala. Some grazing on endophyte-free (E−) or non-toxic (NT), commercial endophyte-infected pastures takes place in the US as well. Earlier, we found that grazing on toxic fescue with low levels of EAs during fall affects thermoregulation, behavior, and weight gain. Building on these findings, the current study aimed to investigate how the presence of low EA-producing E+ or NT endophytes can influence animal metabolome, microbiome, and, ultimately, overall animal health. Eighteen Angus steers were placed on NT, E+, and E− fescue pastures for 28 days. Urine, rumen fluid (RF), rumen solid (RS), and feces were collected pre-exposure, and on days 2, 7, 14, 21, and 28. An untargeted high-resolution metabolomics approach was used to analyze urine and RF, while 16S rRNA-based next-generation sequencing (NGS) was used to examine RF, RS, feces, and fescue plant microbiomes. While alpha- or beta-microbiota diversity across all analyzed matrices were unaffected, there were specific effects of E+ on the relative abundance of some taxa (i.e., Prevotellaceae). Additionally, E+ grazing impacted aromatic amino acid metabolism in the urine and the metabolism of lipids in both the RF and urine. In both matrices, trace amine-related metabolic features differed markedly between E+ and the other groups. Compared to the endophyte-free group, endophyte presence, whether novel or toxic, influenced amino acid and carbohydrate metabolism, as well as unsaturated fatty acid biosynthesis. These findings suggest that low-EA-producing and non-toxic endophytes in fescue have more prominent effects on the metabolome than the microbiome, and this metabolome perturbation might be associated with decreased performance and reported physiological signs of FT. Full article

The advancement of autonomous vehicles has shifted from modular pipeline architectures to end-to-end frameworks, enabling direct learning of control policies from sensory inputs. While frame-based RGB cameras are commonly utilized, they face challenges in dynamic environments, such as motion blur and varying illumination. Alternatively, event-based cameras, with their high temporal resolution and wide dynamic range, offer a promising solution. However, existing end-to-end models for event camera inputs are primarily constructed using traditional convolutional networks and time-sequence models (e.g., Recurrent Neural Networks, RNNs), which suffer from large parameter counts and excessive redundant computations. To address this gap, we propose LiS-Net, a novel framework that incorporates brain-inspired neural networks to construct the overall architecture, applying it to the task of end-to-end steering prediction. The core of LiS-Net is a liquid neural network, which is designed to simulate the behavior of C. elegans neurons for modeling purposes. By leveraging the strengths of event cameras and brain-inspired computation, LiS-Net achieves superior accuracy, smoothness, and efficiency. Specifically, LiS-Net outperforms existing models with the lowest RMSE and MAE, indicating better accuracy, while also maintaining the fewest number of neurons and achieving competitive FLOPs results, showcasing its computational efficiency. Experiments on the simulated EventScape dataset demonstrate its robustness, while validation on our self-collected dataset showcases its generalization capability. We also release the collected dataset comprising synchronized event cameras, RGB cameras, and GPS and CAN data. LiS-Net lays the foundation for scalable and efficient autonomous driving solutions by integrating bio-inspired sensors with brain-inspired computation. Full article

This paper examines the dynamics of quantum steering and fidelity in a two-photon system subjected to dephasing interactions, examining their behavior in Markovian and non-Markovian environments. We consider the case of identical and distinct dephasing rates with experimental parameter values to ensure that the analysis reflects realistic conditions, enhancing its relevance to practical quantum systems. Quantum steering, the ability to remotely influence a quantum state, and fidelity, a measure of initial-state preservation, are investigated for time evolution, initial-state configuration, dephasing parameters, and system characteristics. We model each photon as independently interacting with its environment and derive the time-evolved reduced-density matrix for the bipartite system, focusing on how environmental effects shape the system’s behavior. By integrating experimentally feasible parameter values, this work establishes a practical framework for tuning quantum steering and fidelity, providing valuable insights for applications in quantum information processing, such as secure communication and state preservation. Full article

High-performance control valves are essential components in power plants. High-parameter control valves are specialized valves for controlling high-pressure, high-flow, high-temperature, and highly corrosive media. Control valve performance is critical for the stable operation of power plants. The multi-stage counter-flow passage is a common structure in pressure-reducing control valves, effectively mitigating cavitation and erosion on the valve walls. However, in practice, vibration issues in multi-stage passage valves are particularly pronounced. This study employs FSI (fluid–structure interaction) to simulate the vibration characteristics of multi-stage passages. Flow field data for the multi-stage passage are obtained through FLUENT software. A time-frequency analysis of the lift coefficient in the multi-stage passage flow field was performed. The vibration characteristics of the valve core’s inlet and outlet surfaces were studied using Transient Structural software. The results show that when high-pressure fluid passes through the valve core’s passage, it undergoes buffering, steering, and rotating motions, leading to a gradual pressure drop and generating resistance and lift. These phenomena are primarily caused by vortex shedding in the flow field, with the dominant frequency observed to be approximately 5400 Hz. Additionally, as the valve core progresses through the P1 phase at the inlet and the P2 phase at the outlet, the vibration intensity gradually decreases, reaching a minimum in the sixth phase, before increasing and peaking in the final stage. Analysis of the flow field characteristics within the valve core passage reveals the significant impact of vortex shedding on the valve core’s vibration and lift. Phase analysis of the valve core’s vibration intensity further clarifies its behavioral changes at different operational stages. These findings help optimize the design of multi-stage buffering valve cores, improving their performance and stability. Full article

Virtual reality (VR) continues to evolve, offering immersive experiences across various domains, especially in virtual training scenarios. The aim of this study is to present the development of a VR simulator and to examine its realism, usability, and acceptance by luge experts after an experiment with a VR simulation. We present a novel photogrammetry sensing to VR pipeline for the sport of luge designed with the goal to be as close to the real luge experience as possible, potentially enabling users to learn critical techniques safely prior to real-world trials. Key features of our application include realistic terrain created with photogrammetry and responsive sled dynamics. A consultation of experts from the Austrian Luge Federation led to several design improvements to the VR environment, especially based on user experience aspects such as lifelike feedback and interface responsiveness. Furthermore, user interaction was optimized to enable precise steering and maneuvering. Moreover, two learning modes were developed to accommodate user experience levels (novice and expert). The results indicated a good level of realism of the VR luge simulator. Participants reported scene, audience behavior, and sound realism scores that ranged from 3/5 to 4/5. Our findings indicated adequate usability (system usability score: 72.7, SD = 13.9). Moderate scores were observed for the acceptance of VRodel. In conclusion, our virtual luge application offers a promising avenue for exploring the potential of VR technology in delivering authentic outdoor recreation experiences that could increase safety in the sport of luge. By integrating advanced sensing, simulations, and interactive features, we aim to push the boundaries of realism in virtual lugeing and pave the way for future advancements in immersive entertainment and simulation applications. Full article

This study presents the first investigation into the problem of autonomous vehicle (AV) merging into existing platoons, proposing a multi-agent deep reinforcement learning (MA-DRL)-based cooperative control framework. The developed MA-DRL architecture enables coordinated learning among multiple autonomous agents to address the multi-objective coordination challenge through synchronized control of platoon longitudinal acceleration, AV steering and acceleration. To enhance training efficiency, we develop a dual-layer multi-agent maximum Q-value proximal policy optimization (MAMQPPO) method, which extends the multi-agent PPO algorithm (a policy gradient method ensuring stable policy updates) by incorporating maximum Q-value action selection for platoon gap control and discrete command generation. This method simplifies the training process by using maximum Q-value action policy optimization to learn platoon gap selection and discrete action commands. Furthermore, a partially decoupled reward function (PD-Reward) is designed to properly guide the behavioral actions of both AVs and platoons while accelerating network convergence. Comprehensive highway simulation experiments show the proposed method reduces merging time by 37.69% (12.4 s vs. 19.9 s) and energy consumption by 58% (3.56 kWh vs. 8.47 kWh) compared to existing methods (the quintic polynomial-based + PID (Proportional–Integral–Differential)). Full article

Background: A rat robot can be constructed by electrically stimulating specific brain regions to control rat locomotion and behavior. The rat robot makes full use of the rat’s motor function and energy supply and has significant advantages in motor flexibility, environmental adaptability, and covertness. It can be widely used in disaster search and rescue, terrain survey, anti-terrorism, and explosion-proof tasks. However, the motor control of existing rat robots mainly relies on the virtual whisker touch produced by the electrical stimulation of the barrel area of the somatosensory cortex and the virtual reward generated by the electrical stimulation of the medial forebrain bundle. The methods requires substantial experimental training to encourage the animals to match the virtual sensation with the motor behavior. However, the conditioned reflexes acquired by the animals will gradually disappear after a period of time at the end of the experiments, which will lead to a decrease in the stability of the motor control system. Methods: In this study, we developed a new method to gain control of inclined movement in rats by the electrical stimulation of the ventral tegmental area (VTA) of the midbrain and motor control of steering in rats by the electrical stimulation of nigrostriatal (NS) pathway. Results: The results showed that the electrical stimulation of the rat VTA could induce stable inclined movement in rats and that the neuromodulatory effect significantly correlated with the electrical stimulation parameters. In addition, the electrical stimulation of the NS pathway was able to directly and stably induce the steering movements of the head and trunk to the contralateral side of the stimulated side of the rat. Conclusions: These findings are of great importance for the motor control of rat robots, especially in the field environment with many slopes. In addition, the rat robot constructed based on this method does not need pre-training while ensuring reliability, which greatly improves the preparation efficiency and has certain practical application value. Full article

Autonomous vehicles are increasingly becoming a part of our daily lives, with active chassis control systems playing a pivotal role and drawing significant attention from both academia and industry. Current research on vehicle-to-vehicle overtaking behavior predominantly focuses on low-to-moderate speeds, with insufficient studies addressing high-speed lane-changing maneuvers. Under high-speed conditions, the variability and complexity of road environments significantly increase tracking errors, posing challenges for control algorithms that perform well at lower speeds but may suffer from reduced accuracy or instability at higher speeds. A hybrid control strategy based on vehicle dynamics for high-speed overtaking path tracking is developed to ensure vehicle stability and maneuverability. By integrating Model Predictive Control (MPC) with Four-Wheel Steering (4WS) controllers and employing a two-degree-of-freedom ideal model as the path-tracking response model, we have achieved effective control and path tracking for autonomous vehicles equipped with four-wheel steering. The effectiveness of the proposed control strategy was validated on the Carsim–Simulink integrated simulation platform. Experimental results demonstrate that this strategy offers higher path-tracking accuracy than single-controller approaches under high-speed conditions while also meeting vehicle stability requirements. The model provides robust support for enhancing the path-tracking performance of autonomous four-wheel steering vehicles at medium-to-high speeds, thereby advancing the reliability and safety of autonomous driving technology in practical applications. Full article

The Pacific Subtropical High (PSH) predominantly develops during the boreal summer (June–August) over the Northwest Pacific (NWP) basin, with August accounting for the highest tropical storm (TS) frequency (46.9%). This study examines the critical influence of the PSH’s position on TS trajectories and the consequent exposure of affected countries, utilizing four decades (1977–2016) of August TS data from the NWP. A total of 55 TSs, unaffected by other environmental factors, were analyzed. The PSH’s observed position during each TS’s turning point was delineated using a geopotential height of 500 hPa, while track sinuosity was quantified using a validated sinuosity index (SI). Three distinct TS paths were identified: an eastward PSH position leads to highly sinuous tracks, directing TSs toward Japan; a westward PSH position results in straighter tracks, steering TSs toward the South China Sea (SCS) below Taiwan; and a mid-position guides TSs toward Taiwan. These findings underscore the PSH’s pivotal role in modulating TS behavior and provide valuable insights for disaster risk management agencies to mitigate TS impacts in the NWP basin, the world’s most active TS region, responsible for one-third of global tropical cyclones. Full article

Terrestrial and satellite communications, tactical data links, positioning, navigation, and timing (PNT), as well as distributed sensing will continue to require precise timing and the ability to synchronize and disseminate time effectively. However, the supply of space-qualified clocks that meet Global Navigation Satellite Systems (GNSS)-level performance standards is limited. As the awareness of potential disruptions to GNSS due to adversarial actions grows, the current reliance on GNSS-level timing appears costly and outdated. This is especially relevant given the benefits of developing robust and stable time scale references in orbit, especially as various alternatives to GNSS are being explored. The onboard realization of clock ensembles is particularly promising for applications such as those providing the on-demand dissemination of a reference time scale for navigation services via a proliferated Low-Earth Orbit (pLEO) constellation. This article investigates potential inter-satellite network architectures for coordinating time and frequency across pLEO platforms. These architectures dynamically allocate radio resources for clock data transport based on the requirements for pLEO time scale formations. Additionally, this work proposes a model-based control system for wireless networked timekeeping systems. It envisions the optimal placement of critical information concerning the implicit ensemble mean (IEM) estimation across a multi-platform clock ensemble, which can offer better stability than relying on any single ensemble member. This approach aims to reduce data traffic flexibly. By making the IEM estimation sensor more intelligent and running it on the anchor platform while also optimizing the steering of remote frequency standards on participating platforms, the networked control system can better predict the future behavior of local reference clocks paired with low-noise oscillators. This system would then send precise IEM estimation information at critical moments to ensure a common pLEO time scale is realized across all participating platforms. Clock steering is essential for establishing these time scales, and the effectiveness of the realization depends on the selected control intervals and steering techniques. To enhance performance reliability beyond what the existing Linear Quadratic Gaussian (LQG) control technique can provide, the minimal-cost-variance (MCV) control theory is proposed for clock steering operations. The steering process enabled by the MCV control technique significantly impacts the overall performance reliability of the time scale, which is generated by the onboard ensemble of compact, lightweight, and low-power clocks. This is achieved by minimizing the variance of the chi-squared random performance of LQG control while maintaining a constraint on its mean. Full article