搜索结果：Drive

Understanding and adhering to soft constraints is essential for safe and socially compliant autonomous driving. However, such constraints are often implicit, context-dependent, and difficult to specify explicitly. In this work, we present DRIVE, a novel framework for Dynamic Rule Inference and Verified Evaluation that models and evaluates human-like driving constraints from expert demonstrations. DRIVE leverages exponential-family likelihood modeling to estimate the feasibility of state transitions, constructing a probabilistic representation of soft behavioral rules that vary across driving contexts. These learned rule distributions are then embedded into a convex optimization-based planning module, enabling the generation of trajectories that are not only dynamically feasible but also compliant with inferred human preferences. Unlike prior approaches that rely on fixed constraint forms or purely reward-based modeling, DRIVE offers a unified framework that tightly couples rule inference with trajectory-level decision-making. It supports both data-driven constraint generalization and principled feasibility verification. We validate DRIVE on large-scale naturalistic driving datasets

Vision-Language-Action (VLA) models have advanced autonomous driving, but existing benchmarks still lack scenario diversity, reliable action-level annotation, and evaluation protocols aligned with human preferences. To address these limitations, we introduce DriveAction, the first action-driven benchmark specifically designed for VLA models, comprising 16,185 QA pairs generated from 2,610 driving scenarios. DriveAction leverages real-world driving data proactively collected by drivers of autonomous vehicles to ensure broad and representative scenario coverage, offers high-level discrete action labels collected directly from drivers' actual driving operations, and implements an action-rooted tree-structured evaluation framework that explicitly links vision, language, and action tasks, supporting both comprehensive and task-specific assessment. Our experiments demonstrate that state-of-the-art vision-language models (VLMs) require both vision and language guidance for accurate action prediction: on average, accuracy drops by 3.3% without vision input, by 4.1% without language input, and by 8.0% without either. Our evaluation supports precise identification of model bottlenecks with r

Open-Vocabulary Object Navigation (OVON) requires an embodied agent to locate a language-specified target in unknown environments. Many zero-shot methods rely on frontier-candidate reasoning under incomplete observations, while topology-aware methods reduce candidate redundancy but may still introduce panoramic inspection overhead and repeated reconsideration. We present DRIVE-Nav, a structured framework that organizes exploration around persistent directions rather than raw frontiers. By inspecting encountered directions more completely and restricting subsequent decisions to still-relevant directions within a forward 240-degree view range, DRIVE-Nav reduces redundant revisits and improves path efficiency. The framework extracts and tracks directional candidates from weighted Fast Marching Method (FMM) paths, maintains representative views for semantic inspection, and combines vision-language-guided prompt enrichment with cross-frame verification to improve grounding reliability. Experiments on HM3D-OVON, HM3Dv1, HM3Dv2, and MP3D demonstrate strong overall performance and consistent efficiency gains. On HM3D-OVON, DRIVE-Nav achieves 50.2% SR and 32.6% SPL, improving the previous b

We present ReinDriveGen, a framework that enables full controllability over dynamic driving scenes, allowing users to freely edit actor trajectories to simulate safety-critical corner cases such as front-vehicle collisions, drifting cars, vehicles spinning out of control, pedestrians jaywalking, and cyclists cutting across lanes. Our approach constructs a dynamic 3D point cloud scene from multi-frame LiDAR data, introduces a vehicle completion module to reconstruct full 360° geometry from partial observations, and renders the edited scene into 2D condition images that guide a video diffusion model to synthesize realistic driving videos. Since such edited scenarios inevitably fall outside the training distribution, we further propose an RL-based post-training strategy with a pairwise preference model and a pairwise reward mechanism, enabling robust quality improvement under out-of-distribution conditions without ground-truth supervision. Extensive experiments demonstrate that ReinDriveGen outperforms existing approaches on edited driving scenarios and achieves state-of-the-art results on novel ego viewpoint synthesis.

Multiphoton blockade provides an efficient way to achieve entangled photon sources and leads to wide applications in modern quantum technologies. Here, we propose a scheme to realize multiphoton blockade by a multi-tone drive. Specifically, we demonstrate two-photon and three-photon blockades in a single-mode optical Kerr resonator using a two-tone and a three-tone drive, respectively. In comparison with the single-tone drive, except for the blockade of the $(n+1)$th photon excitation due to large detuning, the key mechanism in this scheme is the sequently resonant excitations of all the $m$-photon states ($m\leq n$) by the $n$-tone drive, which lead to the enhancement of photon generation and the demonstration of multiphoton blockade in the weak driving regime. Moreover, the photon distribution within the system can be adjusted on demand by tuning the relative amplitudes of the driving fields for different frequencies. The scheme can be extended to other bosonic systems and be applied to demonstrate other multiphoton physical effects.

The Alcubierre warp drive is an exotic solution in general relativity. It allows for superluminal travel at the cost of enormous amounts of matter with negative mass density. For this reason, the Alcubierre warp drive has been widely considered unphysical. In this study, we develop a model of a general warp drive spacetime in classical relativity that encloses all existing warp drive definitions and allows for new metrics without the most serious issues present in the Alcubierre solution. We present the first general model for subluminal positive-energy, spherically symmetric warp drives; construct superluminal warp-drive solutions which satisfy quantum inequalities; provide optimizations for the Alcubierre metric that decrease the negative energy requirements by two orders of magnitude; and introduce a warp drive spacetime in which space capacity and the rate of time can be chosen in a controlled manner. Conceptually, we demonstrate that any warp drive, including the Alcubierre drive, is a shell of regular or exotic material moving inertially with a certain velocity. Therefore, any warp drive requires propulsion. We show that a class of subluminal, spherically symmetric warp drive

We propose a correspondence between the Morris--Thorne wormhole metric and a warp drive metric, which generalizes an earlier result by H. Ellis regarding the Schwarzschild black hole metric and makes it possible to embed a warp drive in a wormhole background. We demonstrate that in order to do that, one needs to also generalize the Natario--Alcubierre definition of warp drive and introduce nonzero intrinsic curvature. However, we also find out that in order to be traversable by a warp drive, the wormhole should have a horizon: in other words, humanly traversable wormholes cannot be traversed by a warp drive, and vice versa. We also discuss possible loopholes in this "no-go" theorem

It is commonly accepted that superluminal travel may be used to facilitate time travel. This is a purely special-relativistic argument, using the fact that for observers in two frames of reference, separated by a spacelike interval, the non-causal (spacelike) future of one observer includes part of the causal past of the other. In this paper we provide a concrete realization of this argument in a curved general-relativistic spacetime, using warp drives as the means of faster-than-light travel. By generalizing the usual warp drive metric to allow for a non-unit lapse function, we allow the warp drive to switch between reference frames in a purely geometric way. With an additional modification allowing the warp drive to have compact support, this permits us to glue two warp drives together to construct a closed timelike geodesic, such that a test particle following the geodesics of the two warp drives travels back to its own past. This provides a precise mathematical model for the connection between faster-than-light travel and time travel in general relativity, and the first such model to be explicitly formulated using two warp drives. We also give a detailed discussion of weak ener

High-performance propulsion for mission-critical applications demands unprecedented reliability and real-time fault resilience. Conventional diagnostic methods (signal-based analysis and standard ML models) are essential for stator/rotor fault detection but suffer from high latency and poor generalization across variable speeds. This paper proposes a 1-D Convolutional Neural Network (CNN) framework for real-time fault classification in the HPDM-350 interior permanent magnet synchronous motor (IPMSM). The proposed architecture extracts discriminative features directly from high-frequency current and speed signals, enabling sub-millisecond inference on embedded controllers. Compared to state-of-the-art long short term memory (LSTM) and classical ML approaches, the 1-D CNN achieves a superior weighted F1-score of 0.9834. Validated through high-fidelity magnetic-domain MATLAB/Simscape models, the method demonstrates robust performance across a +-2700 RPM envelope, providing a lightweight solution for mission-critical electric propulsion systems.

The field of warp research has been dominated by analytical methods to investigate potential solutions. However, these approaches often favor simple metric forms that facilitate analysis but ultimately limit the range of exploration of novel solutions. So far the proposed solutions have been unphysical, requiring energy condition violations and large energy requirements. To overcome the analytical limitations in warp research, we introduce Warp Factory: a numerical toolkit designed for modeling warp drive spacetimes. By leveraging numerical analysis, Warp Factory enables the examination of general warp drive geometries by evaluating the Einstein field equations and computing energy conditions. Furthermore, this comprehensive toolkit provides the determination of metric scalars and insightful visualizations in both 2D and 3D, offering a deeper understanding of metrics and their corresponding stress-energy tensors. The paper delves into the methodology employed by Warp Factory in evaluating the physicality of warp drive spacetimes and highlights its application in assessing commonly modeled warp drive metrics. By leveraging the capabilities of Warp Factory, we aim to further warp dri

We propose a mechanism to engineer an $n$-photon blockade in a nonlinear cavity with an $n$-photon parametric drive $λ(\hat{a}^{†n}+\hat{a}^n)$. When an $n$-photon-excitation resonance condition is satisfied, the presence of n photons in the cavity suppresses the absorption of the subsequent photons. To confirm the validity of this proposal, we study the n-photon blockade in an atom-cavity system, a Kerr-nonlinear resonator, and two-coupled Kerr nonlinear resonators. Our results demonstrate that $n$-photon bunching and $(n+1)$-photon antibunching can be simultaneously obtained in these systems. This effect is due both to the anharmonic energy ladder and to the nature of the $n$-photon drive. To show the importance of the drive, we compare the results of the $n$-photon drive with a coherent (one-photon) drive, proving the enhancement of antibunching in the parametric-drive case. This proposal is general and can be applied to realize the $n$-photon blockade in other nonlinear systems.

Warp drives are exotic solutions of general relativity that offer novel means of transportation. In this study, we present a solution for a constant-velocity subluminal warp drive that satisfies all of the energy conditions. The solution involves combining a stable matter shell with a shift vector distribution that closely matches well-known warp drive solutions such as the Alcubierre metric. We generate the spacetime metric numerically, evaluate the energy conditions, and confirm that the shift vector distribution cannot be reduced to a coordinate transformation. This study demonstrates that classic warp drive spacetimes can be made to satisfy the energy conditions by adding a regular matter shell with a positive ADM mass.

Recent advancements in Visual Language Models (VLMs) have made them crucial for visual question answering (VQA) in autonomous driving, enabling natural human-vehicle interactions. However, existing methods often struggle in dynamic driving environments, as they usually focus on static images or videos and rely on downsampling to manage computational costs. This results in the loss of critical details and the difficulty in effectively integrating spatial and temporal information, undermining fine-grained perception and temporal coherence essential for effective decision-making. To tackle these challenges, we introduce LaVida Drive, a novel and efficient VQA framework for autonomous driving. LaVida Drive seamlessly integrates temporal data while maintaining high-resolution inputs for detailed visual perception. It optimizes spatial processing by retaining high-resolution data for intricate details and using lower-resolution inputs for temporal analysis to focus on motion-related features, thereby boosting computational efficiency. The core of LaVida Drive consists of two modules: the \textit{Query-aware Token Selection} module and the \textit{Spatial-Temporal Token Recovery and Enhan

Non-ideal position estimation results in degraded performance of synchronous motor drive systems due to reduction of the average capability of the drive as well as torque harmonics of different orders. The signature and extent of the performance degradation is further dependent, quite significantly, on the current control architecture, i.e., feedforward or feedback control, employed. This paper presents a comprehensive analysis of non-idealities or errors in position estimation and their effects on the control performance of synchronous motor drives. Analytical models capturing the error in various signals caused by position sensing errors in the drive system for different control architectures are presented and are validated with simulation and experimental results on a prototype permanent magnet synchronous motor drive.

This study underscores the vital importance of intelligent driving functions in enhancing road safety and driving comfort. Central to our research is the challenge of obtaining sufficient test data for evaluating these functions, especially in high-risk, safety-critical driving scenarios. Such scenarios often suffer from a dearth of available data, primarily due to their inherent complexity and the risks involved. Addressing this gap, our research introduces a novel methodology designed to create a wide array of diverse and realistic safety-critical driving scenarios. This approach significantly broadens the testing spectrum for driver assistance systems and autonomous vehicle functions. We particularly focus on the follow-up drive scenario due to its high relevance in practical applications. Here, vehicle movements are intricately modeled using kinematic equations, incorporating factors like driver reaction times. We vary parameters to generate a spectrum of plausible driving scenarios. The utilization of the Difference Space Stopping (DSS) metric is a pivotal element in our research. This metric plays a crucial role in the safety evaluation of follow-up drives, facilitating a mor

We study the generalizations of the original Alcubierre warp drive metric to the case of curved spacetime background. We find that the presence of a horizon is essential when one moves from spherical coordinates to Cartesian coordinates in order to avoid additional singularities. For the specific case of Schwarzschild black hole, the horizon would be effectively absent for the observers inside the warp bubble, implying that warp drives may provide a safe route to cross horizons. Moreover, we discover that the black hole's gravitational field can decrease the amount of negative energy required to sustain a warp drive, which may be instrumental for creating microscopic warp drives in lab experiments. A BEC model is also introduced to propose possible test in the Analogue Gravity framework.

A process for using curvature invariants is applied to evaluate the metrics for the Alcubierre and the Natario warp drives at a constant velocity.Curvature invariants are independent of coordinate bases, so plotting these invariants will be free of coordinate mapping distortions. As a consequence, they provide a novel perspective into complex spacetimes such as warp drives. Warp drives are the theoretical solutions to Einstein's field equations that allow the possibility for faster-than-light (FTL) travel. While their mathematics is well established, the visualisation of such spacetimes is unexplored. This paper uses the methods of computing and plotting the warp drive curvature invariants to reveal these spacetimes. The warp drive parameters of velocity, skin depth and radius are varied individually and then plotted to see each parameter's unique effect on the surrounding curvature. For each warp drive, this research shows a safe harbor and how the shape function forms the warp bubble. The curvature plots for the constant velocity Natario warp drive do not contain a wake or a constant curvature indicating that these are unique features of the accelerating Natario warp drive.