To avoid unmanned ground vehicles being obstructed by deformed terrain in off-road, effective vehicle mobility analysis is required. However, the computational complexity of existing mobility analysis methods, such as discrete element analysis, poses significant challenges when applied to large-scale terrains. To address this problem, we propose an efficient and high-fidelity vehicle mobiliy prediction method for a large-scale terrain. Initially, precise terrain models are constructed employing Gaussian sampling, thereby serving as optimal inputs for the mobility simulation. Subsequently, we introduce a co-simulation method based on a multi-body dynamics model and discrete element analysis to obtain high-fidelity vehicle mobility data on sampled terrains. Following that, the mobility data is utilized to train a PSO-kriging neural network, enabling accurate predictions of the global mobility map. Through rigorous simulation experiments, the proposed method demonstrates its remarkable effectiveness.

Parameter calibration is critical for self-localization based on dead reckoning in the control of intelligent vehicles such as autonomous driving. Most traditional calibration methods for robotics control based on dead reckoning rely on data collection with specially designed paths. For the calibration of parameters in the control of intelligent vehicles, the design of such paths is considered impossible due to the complexity of road conditions. To solve this problem, an optimization-based dead reckoning calibration scheme is introduced in this research using the differential global positioning system to obtain the actual positions of the intelligent vehicle. In this scheme, the difference between the positions obtained through dead reckoning and the positions obtained through the differential global positioning system is selected as the optimization objective function to be minimized. An adaptive quantum-inspired evolutionary algorithm is developed to improve the quality and efficiency of optimization. Experiments with an intelligent vehicle were also conducted to demonstrate the effectiveness of the developed calibration scheme. In addition, the newly introduced adaptive quantum-inspired evolutionary algorithm is compared with the classic genetic algorithm and the classic quantum-inspired evolutionary algorithm using eight benchmark test functions considering computation quality and efficiency.

Aiming at the influence of fewer feature points and dynamic obstacles on location and mapping in off-road environments, we propose a dual-constraint LiDAR-based Simultaneous Localization and Mapping (SLAM) scheme. By abstracting LiDAR registration into two constraints, namely, in-window constraints and out-of-window constraints, the in-window constraints allow our scheme to compromise between accuracy and real-time performance, and out-of-window constraints can exploit optimized variables to provide richer constraint information. The advantages of incremental SLAM map construction can be used to design a variety of map forms. Although the variables outside the window are no longer involved in the optimization, we can use the two-dimensional probability grid map to provide binary semantic information and dynamic weights for the constraints outside the window to enhance the registration accuracy. Finally, we conducted experiments in off-road environment and compared with the mainstream LiDAR SLAM scheme, which can prove that our scheme has more advantages in accuracy.

Generative pre-training via neural rendering has become a cornerstone for scaling 3D perception in autonomous driving. However, prevalent approaches relying on implicit Neural Radiance Fields (NeRFs) face two fundamental limitations: the shape-radiance ambiguity inherent in vision-centric optimization and the prohibitive computational overhead of volumetric ray marching. To address these challenges, we propose AES-Gaussian, a novel multi-modal pre-training framework grounded in the efficient 3D Gaussian Splatting (3DGS) representation. Diverging from symmetric fusion paradigms, our core innovation is an Asymmetric Encoder architecture that couples a deep semantic vision backbone with a lightweight, physics-aware LiDAR branch. In this framework, LiDAR data serve not merely for semantic extraction, but as sparse physical anchors. By employing a novel Explicit Feature Synergy mechanism, we directly inject raw LiDAR intensity and depth priors into the Gaussian decoding process, thereby rigidly constraining scene geometry in open-world environments. Extensive empirical validation on the nuScenes dataset demonstrates the superiority of our approach. AES-Gaussian achieves state-of-the-art transfer performance, yielding a substantial 7.0% improvement in NDS for 3D Object Detection and a 4.8% mIoU gain in 3D semantic occupancy prediction compared to baselines. Notably, our method reduces geometric reconstruction error by over 50% while significantly improving training and inference efficiency, attributed to the streamlined asymmetric design and rapid Gaussian rasterization. Ultimately, by enhancing both perception accuracy and system efficiency, this work contributes to the development of safer and more reliable autonomous driving systems.

Deep learning-based perception algorithms in autonomous driving are vulnerable to adversarial examples. Existing adversarial attacks methods often struggle to reconcile the trade-off between visual fidelity and high adversarial aggressiveness, either generating conspicuous noise patterns that lack visual naturalness or requiring computationally expensive iterative optimization during inference. To address these limitations, we propose DSAdv, a novel framework for high-fidelity adversarial example generation via improved LoRA fine-tuning of Latent Diffusion Models (LDMs). We introduce a decoupled Dual-Stream LoRA architecture consisting of an Adversarial Stream to inject blinding perturbations and an Integrity Keeper Stream to maintain semantic realism. These streams are dynamically fused via an Adaptive Input Gating mechanism, effectively resolving the dynamic coordination problem between visual fidelity and high adversariality. Furthermore, our method guides the LoRA fine-tuning of the pre-trained LDM by backpropagating gradients from a target detector, which significantly shortens the inference process. Extensive experiments on the NuScenes, KITTI, and the Science Island datasets validate that DSAdv generates naturalistic adversarial examples with higher transferability and visual quality compared to state-of-the-art methods like AdvDiff and PGD. The proposed framework enables the efficient generation of naturalistic and stealthy adversarial scenarios, providing a rigorous tool for evaluating and enhancing the robustness of autonomous driving systems.