To address the complex operating conditions and diverse fault types of electric mining truck drive motors, a fault diagnosis method integrating the Caterpillar Fungus Optimizer (CFO) with Support Vector Machine (SVM) is proposed. A parameter optimization model is constructed to jointly optimize the slack variable, penalty factor, and kernel function parameter of SVM, thereby improving fault recognition accuracy and model stability. Fault sample data are collected under typical operating conditions for identification experiments, and uniformly distributed random disturbances with different amplitudes are introduced to evaluate model robustness. Comparative analysis with several commonly used optimization methods demonstrates that the proposed approach exhibits strong adaptability in terms of fault classification accuracy, anti-interference capability, and model training stability. The average recognition accuracy reaches 98.67%, while the accuracy remains above 94.7% under disturbance conditions.

To address challenges in autonomous driving scenarios, such as uneven distribution of target scales, significant background interference, structural distortion of small objects, and high computational complexity, this paper proposes an improved algorithm, ELD-YOLO, based on YOLOv11n. The C3k2-EMSC module is introduced to enhance multi-scale feature extraction, capturing richer contextual information in complex traffic scenarios. The LSDECD module is designed to strengthen feature representation through shared convolutions and group normalization, reducing redundant responses. The DySample module is introduced to enhance small object and edge feature representation through dynamic sampling. Results demonstrate that on the KITTI and BDD100K datasets, the parameter quantity of ELD-YOLO is reduced by 16% compared with YOLOv11n, while achieving mAP@0.5s of 89.7% and 52.7%, respectively. This approach significantly improves detection accuracy and robustness while reducing model complexity.

To address the challenges of complex surface textures, blurred boundaries, and significant illumination variations in unstructured road scenarios, this paper proposes a freespace detection method using only red-green-blue (RGB) images and integrating boundary enhancement and geometry-depth prior attention. Based on the SegFormer framework, the proposed method extracts explicit boundary information through auxiliary boundary supervision and introduces boundary cues into the semantic feature space via a boundary feature fusion module, thereby enhancing the representation of soft boundaries and irregular contours. Meanwhile, a geometry-depth prior attention module is designed to incorporate geometric continuity priors and vertical pseudo-depth priors, enabling structured feature reweighting and improving prediction stability in complex terrain. Experimental results show that the proposed method achieves an IoU of 90.8% and an F1-score of 95.1% on the Off-Road Freespace Detection (ORFD) dataset, outperforming the compared monocular RGB methods on major segmentation metrics and demonstrating good cross-dataset generalization capability. These results indicate that the proposed method achieves a favorable balance among segmentation accuracy, boundary quality, and inference efficiency.

To address the challenge of balancing accuracy and real-time performance when autonomous vehicles detect traffic diversion devices in roadworks scenarios, a detection algorithm that improves upon YOLOv8 is proposed. This paper utiliseS Make Sense for data annotation and combined this with data augmentation techniques to construct a dedicated dataset. The backbone network with EfficientNetV2-S to enhance feature extraction is replaced and an inverted Residual Moving Block (iRMB) module before the Spatial Pyramid Pooling Fast (SPPF) module to focus on key targets is edded. Morever, the Distance-based Intersection-over-Union (DIou) loss function is adopted to optimise the model. Experimental results validate that the improved model achieves a precision of 90.7%, with a recall increase of 4.8 percentage points and an average precision (mAP@0.5 and mAP@0.5∶0.95) that has risen by 4.3 percentage points. Upon deployment to an Android mobile device, the frame rate remains stable at over 40 frames per second.

In the process of testing intelligent vehicle camera sensor functionality in foggy environments, real-world fog test scenarios face high acquisition costs, poor repeatability, and uncontrollable visibility, while virtual simulation methods lack sufficient realism. To address these issues, this paper proposes a simulation test method specifically for camera sensors in fog environments. First, atmospheric light in different scene regions is estimated via semantic segmentation; next, combining target visibility and fog microphysical parameters, the atmospheric scattering coefficient is computed using a scattering model, and depth information is corrected with the camera intrinsic and extrinsic parameters to produce an accurate transmission map; finally, a real, controllable fog test platform is constructed, and the proposed method is validated through image quality evaluation (PSNR/SSIM) and YOLO-V5-based object recognition tests. Experimental results demonstrate that this method generates high-fidelity foggy images across a 25~100 m visibility range and outperforms existing simulation approaches in both image quality and functional testing, significantly reducing data-collection costs while ensuring repeatable, controllable, and high-quality camera sensor validation.

To address the high dependency of AUTOSAR toolchains on commercial operating systems and insufficient FreeRTOS adaptation, a modular RTE code auto-generation tool is proposed. The tool integrates components such as an information extractor, parser, processing template, and code generator to automatically produce RTE code compliant with AUTOSAR standards in FreeRTOS environments. Test results show that the generated code meets industrial requirements in structural compliance, type safety, and real-time performance. Development time per module is reduced from 8.2 man-hours to 0.5 man-hours. Cross-platform adaptation efficiency increases 12.7 times. This research provides an efficient and reliable solution for embedded software development in intelligent vehicles.

To accurately evaluate the decision precision of Automatic Emergency Braking (AEB) system intervention timing under varying load conditions in highway scenarios, this study screens 255 commercial vehicle rear-end collision cases from the China In-Depth Accident Study (CIDAS) database and extracts typical scenarios using hierarchical clustering. The clustering variables encompass traffic environment (weather, road type, lighting, and adhesion conditions), subject vehicle type, pre-crash status, and collision characteristics (load state, motion state, impact location, and initial speed discretized into 4 intervals: low, medium, high, and extremely high), as well as target vehicle type and motion state. 4 highly representative test scenarios are obtained, in which the load state is innovatively incorporated as a key clustering variable into the accident data structure analysis. Combined with vehicle dynamics principles, the influence of load on braking distance and AEB response threshold is discussed, enabling the extracted scenarios to simultaneously reflect statistical patterns in accident data and practical braking characteristics of commercial vehicles, thereby addressing the deficiency of existing studies in considering load factors. The constructed scenarios satisfy the requirements of UN R131 regulation and are further extended to beyond-regulation speeds and offset-overlap conditions, significantly enhancing adaptability to the complex traffic environment on Chinese highways. This paper provides a scientific basis for commercial vehicle AEB development, performance optimization, and standard formulation and revision.

To describe the coupling effect of multiple nonlinear factors on the torsional vibration of pure electric vehicles, a full-vehicle torsional vibration model with time-varying meshing force is established. 3 nonlinear factors exist in gear meshing. They are time-varying meshing stiffness, gear backlash and transmission error. This work analyzes their physical characteristics and their influence on meshing force. A time-varying meshing force equation is formulated. The permanent magnet synchronous motor model couples with the transmission torsional vibration model containing time-varying meshing force. A complete electromechanical coupling simulation model for the whole vehicle is built. Numerical calculation obtains the torsional vibration responses of the vehicle. The coupling influence of multiple nonlinear factors on dynamic characteristics of each vehicle component is systematically analyzed. The results show that under the coupling of multiple nonlinear factors, the peak value of time-varying meshing force rises by 17.46%. The frequency-domain response of the differential increases by 30.6%. The dynamic response of vehicle body angular acceleration changes from predictable to unpredictable.

Aming at the challenges that trajectory planning and tracking control of autonomous vehicles struggle to respond to dynamic obstacles in real-time under urban road conditions and the insufficient robustness of complex traffic environments, this paper proposes a trajectory optimization method that integrates hierarchical planning and dynamic replanning. It first generates a coarse path using fast linear programming, then refines the path using an improved artificial potential field method and numerical optimization. A trajectory replanning module is designed to predict dynamic obstacle trends and adjust the path in real-time. Lateral control uses Model Predictive Control (MPC) based on the Frenet coordinate system, and longitudinal control uses a dual PID controller. Simulation results show that, in the straight-road static obstacle avoidance scenario, the proposed method starts obstacle avoidance approximately 45 m before reaching the obstacle. In the overtaking scenario, its speed fluctuation is about 5 km/h, lower than 17.5 km/h for GD-APF and 11 km/h for D-RRT. The lateral acceleration is mainly maintained within −1 to 1 m/s². Moreover, the maximum lateral tracking error is 0.84 m, and the longitudinal speed error is controlled within 1 m/s. These results indicate that the proposed method achieves good comprehensive performance in path smoothness, dynamic obstacle avoidance, and trajectory tracking accuracy.

To address the issues of strong search blindness and poor path feasibility in the traditional Bi-RRT* algorithm, this paper proposes an improved approach named GC-APF-Bi-RRT* (Bi-RRT* based on Guided Connection and Artificial Potential Field). The algorithm first establishes a collision model for vehicles considering four-wheel steering vehicles. It then overcomes the problems of sampling difficulties and reciprocating paths by extracting environmental guide points and employing a search-based connection strategy constrained by direction consistency. Furthermore, an improved APF integrated with a goal-distance factor is introduced to guide the tree growth direction. Finally, greedy pruning and B-spline smoothing techniques are applied to optimize the trajectory. Simulation results demonstrate that in complex environments, compared to traditional RRT, RRT*, RRT-Connect, Bi-RRT*, and Goal-Biased RRT*, the proposed algorithm reduces path length by 19.45%, 9.94%, 13.85%, 4.10%, and 3.61%, respectively, and shortens search time by 62.57%, 65.15%, 0.97%, 66.63%, and 71.28%, respectively.