Accurate 3-D rail-track detection is vital to the perception of the railway environments for autonomous trains. However, existing methods based on monocular image cannot capture 3-D spatial features, facing challenges in detecting 3-D rail-track in turnouts and distant scenarios. This study introduces 3DRailNet, a long-range 3-D rail-track detection network using multifocal cameras. 3DRailNet consists of two modules: disparity-based feature extraction (DFE) module and long-short rail-track detection (LSRD) module. Specifically, the DFE module utilizes multifocal images to generate a disparity image and depth image, acquiring 3-D depth features to enhance the spatial information of rail-track. Based on the 3-D depth features, the LSRD module designs a detection head for long and short focal cameras to predict the 3-D position of rail-track. Experimental results demonstrate that the mean F1 score (mF1) of our proposed 3DRailNet is 83.2%, establishing it as the state-of-the-art method in this field. All these results indicate that 3DRailNet has the potential to be readily applicable in 3-D rail-track detection in railway environments.

This study proposes a risk-tolerant dispatch approach for a fleet of autonomous mining trucks in an open-pit mine, leveraging early signs to reduce the impact of potential failures that may or may not occur later. Unlike traditional methods that ignore these early signs or wait until a failure has actually happened, our approach proactively plans for both possible outcomes without relying on probabilities. We propose a Y-shaped solution structure composed of a shared trunk that covers the period before it becomes clear if the failure will occur and two separate branches that address the final scenarios. We formulate the dispatch problem as a mixed-integer linear program and solve it via Gurobi. To facilitate the solution process with Gurobi, an evolutionary algorithm is adopted to explore the solution space for a good initial guess. A high-performance discrete-event simulator is embedded in the cost function evaluation module of the evolutionary algorithm for quickly selecting qualified solution candidates. By integrating both the failure and non-failure scenarios into one unified plan, we avoid extreme risk-taking or undue conservatism, ensuring stable operational performance. Simulations and field trials at a real open-pit mine confirm that this risk-tolerant approach effectively manages failure risks when early signs are available.

Path tracking control is essential for ensuring the safe and efficient operation of autonomous trucks, but traditional methods often struggle with nonlinear vehicle dynamics. While deep reinforcement learning (DRL) approaches are model-free, they may lack the stability and interpretability required for reliable deployment. This study presents a hybrid control framework that combines Pure Pursuit (PP) with Proximal Policy Optimization (PPO) to enhance tracking accuracy and robustness. PP provides baseline stability and interpretability, while PPO refines control actions by optimizing policy gradients, ensuring better adaptability to nonlinear dynamics and complex driving conditions. An adaptive look-ahead mechanism, responsive to speed and curvature, dynamically adjusts preview distances using PPO-generated coefficients, facilitating early corrections during high-speed turns and enabling greater precision on sharp curves. A fusion training method, leveraging high-reward initialization and a decreasing learning rate, supports efficient exploration and stable convergence. The approach was validated in a high-fidelity simulation environment using PreScan, Simulink, and ROS, along with real-world experiments on a proportionally scaled intelligent vehicle chassis, demonstrating notable improvements in path tracking accuracy and robustness across varied path profiles.

Addressing the crucial challenge of autonomous navigation for unmanned ground vehicles (UGVs), this paper presents a dual-layer path planning method integrating Probabilistic Roadmap (PRM) and Proximal Policy Optimization (PPO). Combining global guidance with local optimization, this approach effectively mitigates the shortcomings of traditional path planning methods such as blindness and local optimality, thus enhancing the efficiency and feasibility of path planning. Specifically, we propose a PRM-RL dual-layer path planning framework that employs the PRM algorithm to generate sub-goals for guiding reinforcement learning exploration, thereby improving training efficiency. Simultaneously, we utilize the PPO algorithm to optimize paths, considering vehicle kinematics and introducing soft constraints to ensure smoother paths adaptable to diverse application scenarios. The superiority and practicality of our method are validated through ablation experiments and comparative experiments, offering a reliable path planning solution for autonomous navigation of UGVs.

As intelligent and autonomous driving technologies advance, the secure and productive functioning of mining trucks during autonomous operations is critically dependent on precise object detection within open-pit mining environments. Due to the sparse nature of point cloud data for small objects such as stones, cones, and equipment parts, they are susceptible to generating false negatives and false positives. Although camera images are rich in semantic content, the absence of precise depth cues renders them vulnerable to interference from background noise. Currently, the technology of fusing LiDAR point cloud and camera image data for 3D object detection is gaining popularity in this field. In this paper, we introduce a query feature generation strategy based on multi-scale image features in the cross-attention-based feature fusion module to generate high-quality fusion features for small objects. Additionally, within the traditional non-maximum suppression strategy, we introduce size factors and shape factors to alleviate the sensitivity of small objects to bounding box offsets. Notably, our network model has achieved an mAP of 88.0 and an NDS of 77.9 on the mining dataset, significantly enhancing the precision of detecting small objects over previous approaches.

Autonomous transportation systems, which have reconstructed open-pit mining operations, depend on accurate and real-time obstacle detection in challenging environments. Existing methods often exhibit limitations in accuracy due to their reliance on traditional image-based techniques, leading to errors and incomplete results. To address these issues, we developed a vision-based 3D object detection algorithm designed for open-pit mining environments. Our approach leverages the BEVDepth model, enabling accurate 3D object recognition using monocular camera input. Moreover, camera parameters are integrated to enhance resilience and flexibility across diverse mining environments. Last, our algorithm was implemented on a newly customized open-pit mining dataset. The experimental results verify the algorithm efficiency by attaining a high mean Average Precision score (m$A$P) of 67% while offering real-time performance with inference times as short as 25ms per frame. It enables accurate and efficient obstacle detection in autonomous mining vehicles and contributes to the development of safer and more productive mining operations.

The fusion of camera and LiDAR features in the Bird's Eye View (BEV) perspective has become a prevalent solution for 3D detection in autonomous driving due to its simplicity and efficiency. However, in challenging environments like mining areas with dusty roads, existing BEV -based fusion networks struggle due to dust occlusion and misidentification of dust as obstacles. To address this, we propose a robust fusion framework that integrates fine-grained depth supervision and channel-wise attention mechanisms. Combined with a temporal multi-frame mechanism, our framework effectively mitigates issues caused by dust occlusion and misidentification. We validated our method's detection accuracy on a self-constructed mining road dataset, achieving 89.6% mAP, surpassing BEVFusion's 88.3% mAP. Tests on sensor occlusion and failure further demonstrated its robustness in adverse conditions typical of unstructured road scenarios.

The development of unmanned transportation technology has improved operational efficiency and safety in open-pit mining areas. However, there remain significant challenges to be addressed. One pressing issue is the need for mining trucks to pass through the area where the cable is laid on the ground. Accidentally crushing or damaging these cables would lead to significant risk to the mining area operation. However, due to the complexity of the mining environment and the characteristics of cables being thin and curved, the existing methods such as edge segmentation are difficult to meet the requirements for cable segmentation. This paper introduces a cable segmentation method based on Mask2Former and carries out comprehensive experiments to examine the effectiveness of the method. Experimental results show that the proposed method achieves IoU of 66.04% and PA of 80.08%, which can meet the requirements of practical applications.

Real-time and full-sample vehicle origin-destination (OD) information is essential for traffic management and control in urban road network. However, the low coverage of automatic vehicle identification (AVI) detection devices leads to difficulty in estimating OD. As an emerging traffic data, the trajectories of connected vehicles (CVs) can effectively provide information on their origin and destination. To this end, this paper presents a framework of an autoencoder network utilizing feature transfer to estimate urban dynamic OD based on the characteristics of two data sources. Specifically, a generative adversarial network is introduced to learn high-dimensional feature that is domain-invariant in two data domains. In addition, a pre-training fine-tuning approach is proposed to transfer knowledge pretrained from CV data to the limited AVI observation for OD imputation. Finally, the model was subjected to a real-world road network test. The results showed that for all OD flows the relative error was 11.23 vehicles/30 minutes, which outperformed baseline models, including popular neural networks and existing estimation models for multi-source data fusion. Furthermore, the model’s robustness to external factors, such as observation conditions and data quality, was examined. The results demonstrated that the model consistently delivers satisfactory estimation performance across a diverse range of conditions.

Rail track segmentation is key to environmental perception of autonomous train. However, due to the complexity of railway track environment, critical issues such as the detection of rail tracks with different curvatures remain to be overcome. In this study, a novel architecture called FarNet is proposed for long-range railway track point cloud segmentation. The proposed FarNet is mainly divided into three parts, i.e., spherical projection, attention-aggregation network and results refinement. Specifically, spherical projection converts the LiDAR point cloud into a pseudo range image, and attention-aggregation network enables railway track detection using the pseudo range image. Furthermore, in the attention-aggregation network two components, i.e., spatial attention module and information aggregation module, are proposed to enhance the capability of rail track segmentation. Last, the results refinement helps further filter out the noise points after segmentation. Experimental results show that the proposed FarNet achieved 98.0% mean intersection-over-union (MIoU) and 98.9% mean pixel accuracy (MPA) for rail track segmentation.

Accurate train positioning is crucial for train safety. In this paper, we propose a train positioning method which fuses vison and millimeter-wave radar data. The proposed method contains two parts: loop closure detection (LCD) and radar-based odometry. The loop closure detection part fuses the convolutional neural network (CNN) features and the line features to achieve accurate key location detection. The radar-based odometry part proposes a train speed measurement algorithm using millimeter-wave radar, and combines the results of loop closure detection to further realize train positioning. Experiments conducted on the Hong Kong metro Tsuen Wan line show that our proposed loop closure detection can achieve an efficient key location detection with 98.57% precision and 99.37% recall; the speed detection method fulfills the ETCS requirements; and the relative error of the proposed train positioning method is 0.45%. Besides, the proposed method has been applied on the Hong Kong Metro TSUEN WAN line.