arXiv subjects

Zhidong Wang

Publications and source records attributed to Zhidong Wang.

5 recordsLinked to original sources

arXiv2026

Ego-Dynamics-Augmented World Model for Autonomous Driving with Zero-Shot Cross-Embodiment Adaptation

End-to-end autonomous driving requires generalization ability across platforms with dissimilar physical characteristics. The chassis defines the physical embodiment of each platform, and real-world fleets span sub-tonne microcars to bus-class vehicles. Consequently, the driving stack must either be retrained per platform or adapt to the underlying chassis dynamics online. World model (WM)-based reinforcement learning offers a sample-efficient path toward end-to-end autonomous driving on egocentric bird's-eye-view (BEV) representations, but its effectiveness hinges on how faithfully the WM captures the ego vehicle's dynamics. This work identifies a structural bottleneck in BEV-based WMs: observation transitions entangle ego-motion with scene dynamics, consuming modeling capacity at the cost of imagination accuracy. This burden is embodiment-dependent: dissimilar chassis produce different observation warps under the same control input. The proposed DynaDreamer addresses this bottleneck by conditioning the WM's latent distributions on a physics-informed ego-dynamics context derived from a lateral dynamics model with a neural tire force formulation. This context is extracted online via a neural-ODE encoder-decoder that simultaneously identifies the underlying chassis parameters. Information-theoretic analysis confirms that this conditioning removes the ego-motion terms from both the WM's transition entropy and its prior-posterior KL divergence. The identified physical parameterization enables zero-shot cross-embodiment adaptation across a dynamically diverse fleet without per-platform retraining. Simulation results show 28% and 43% improvements in driving task success rates over the strongest baseline in urban and highway scenarios, and the advantage over the base Transformer WM reaches up to 73% when extrapolating to unseen chassis.

cs.RO↗

arXiv2026

AutoMoT: A Unified Vision-Language-Action Model with Asynchronous Mixture-of-Transformers for End-to-End Autonomous Driving

Integrating vision-language models (VLMs) into end-to-end (E2E) autonomous driving (AD) systems has shown promise in improving scene understanding. However, existing integration strategies suffer from several limitations: they either struggle to resolve distribution misalignment between reasoning and action spaces, underexploit the general reasoning capabilities of pretrained VLMs, or incur substantial inference latency during action policy generation, which degrades driving performance. To address these challenges, we propose AutoMoT in this work, an end-to-end AD framework that unifies reasoning and action generation within a single vision-language-action (VLA) model. Our approach leverages a mixture-of-transformer (MoT) architecture with joint attention sharing, which preserves the general reasoning capabilities of pre-trained VLMs while enabling efficient fast-slow inference through asynchronous execution at different task frequencies. Extensive experiments on multiple benchmarks, under both open- and closed-loop settings, demonstrate that AutoMoT achieves competitive performance compared to state-of-the-art methods. We further investigate the functional boundary of pre-trained VLMs in AD, examining when AD-tailored fine-tuning is necessary. Our results show that pre-trained VLMs can achieve competitive multi-task scene understanding performance through semantic prompting alone, while fine-tuning remains essential for action-level tasks such as decision-making and trajectory planning. We refer to https://automot-website.github.io/ for the demonstration videos and qualitative results.

cs.CV↗

arXiv2025

Strong convergence and Mittag-Leffler stability of stochastic theta method for time-changed stochastic differential equations

We propose the first $α$-parameterized framework for solving time-changed stochastic differential equations (TCSDEs), explicitly linking convergence rates to the driving parameter of the underlying stochastic processes. Theoretically, we derive exact moment estimates and exponential moment estimates of inverse $α$-stable subordinator $E$ using Mittag-Leffler functions. The stochastic theta (ST) method is investigated for a class of SDEs driven by a time-changed Brownian motion, whose coefficients are time-space-dependent and satisfy the local Lipschitz condition. We prove that the convergence order dynamically responds to the stability index $α$ of stable subordinator $D$, filling a gap in traditional methods that treat these factors independently. We also introduce the notion of Mittag-Leffler stability for TCSDEs, and investigate the criterion of Mittag-Leffler stability for both the exact and numerical solutions. Finally, some numerical simulations are presented to illustrate the theoretical results.

math.PR↗

arXiv2024

A clustering adaptive Gaussian process regression method: response patterns based real-time prediction for nonlinear solid mechanics problems

Numerical simulation is powerful to study nonlinear solid mechanics problems. However, mesh-based or particle-based numerical methods suffer from the common shortcoming of being time-consuming, particularly for complex problems with real-time analysis requirements. This study presents a clustering adaptive Gaussian process regression (CAG) method aiming for real-time prediction for nonlinear structural responses in solid mechanics. It is a data-driven machine learning method featuring a small sample size, high accuracy, and high efficiency, leveraging nonlinear structural response patterns. Similar to the traditional Gaussian process regression (GPR) method, it operates in offline and online stages. In the offline stage, an adaptive sample generation technique is introduced to cluster datasets into distinct patterns for demand-driven sample allocation. This ensures comprehensive coverage of the critical samples for the solution space of interest. In the online stage, following the divide-and-conquer strategy, a pre-prediction classification categorizes problems into predefined patterns sequentially predicted by the trained multi-pattern Gaussian process regressor. In addition, dimension reduction and restoration techniques are employed in the proposed method to enhance its efficiency. A set of problems involving material, geometric, and boundary condition nonlinearities is presented to demonstrate the CAG method's abilities. The proposed method can offer predictions within a second and attain high precision with only about 20 samples within the context of this study, outperforming the traditional GPR using uniformly distributed samples for error reductions ranging from 1 to 3 orders of magnitude. The CAG method is expected to offer a powerful tool for real-time prediction of nonlinear solid mechanical problems and shed light on the complex nonlinear structural response pattern.

stat.ML↗

arXiv2019

Quantum Efficiency Enhancement by Mie Resonance from GaAs Photocathodes Structured with Surface Nanopillar Arrays

A new type of negative electron affinity (NEA) photocathode based on GaAs nanopillar array (NPA) Mie-type resonators was demonstrated for the first time. For visible wavelengths, the Mie resonances in GaAs NPA reduced light reflectivity to less than 6 percent compared to a typical value great then 35 percent. Other benefits of NPA resonators include an enhanced density of optical states due to increased light concentration and increased electron emission area. These features resulted in improved photoemission performance at the resonance wavelength demonstrating maximum quantum efficiency 3.5 times greater than a GaAs wafer photocathode without the NPA structure. This optically dark photocathode (sub-percentage light reflectance over visible wavelengths) but possessing electrically high-brightness (enhanced electron emission) provides new opportunities for practical applications such as large-scale electron accelerators, high-resolution night-vision imaging and low energy electron microscopy.

physics.optics↗