The optimization of open-pit metal mining transportation systems is pivotal for enhancing economic viability, operational efficiency, and environmental stewardship. While the traditional Truck-Shovel (TS) system remains prevalent, the In-Pit Crushing and Conveying (IPCC) system presents a transformative alternative with significant potential. This paper provides a comprehensive and systematic review of the advancements in both TS and IPCC systems, critically evaluating their performance across economic, safety, social, and environmental dimensions. Given the promise of IPCC technology, the review dedicates particular focus to its core optimization challenges: (i) determining the Ultimate Pit Limit (UPL) integrated with conveyor layout, (ii) optimizing the transition depth and crusher location, and (iii) synchronizing production scheduling with crusher relocation strategies. We synthesize a wide body of literature to analyze the strengths and limitations of prevalent optimization methodologies, from classical algorithms to modern stochastic and integrated frameworks. Furthermore, the paper identifies and discusses persistent challenges in transportation process evaluation and system-wide IPCC optimization. Finally, we propose a forward-looking research agenda aimed at developing intelligent, sustainable, and integrated transportation systems for the deep open-pit mines of the future. This work aims to serve as a foundational reference for mine planners considering IPCC implementation and to chart a course for future academic inquiry in open-pit mine transportation optimization.

Accurate spatial measurement of blasting areas, typically acquired via unmanned aerial vehicle aerial surveys, is a foundational prerequisite for unmanned open-pit mine operations. Driven by such high-fidelity geometric data, intelligent explosive vehicles still face critical challenges in parking efficiency, route planning, and avoiding redundant travel in irregular explosive zones. This paper proposes Thiessen Edge Shift (TES), a parking and route planning strategy based on the Thiessen (Voronoi) skeleton. TES extracts a Voronoi (Thiessen) skeleton for each blasting area and treats edges and vertices as candidate stopping locations. Parking selection is formulated as a topology-aware greedy set-cover problem that favors compact coverage and suppresses dead-end branches. The selected parking points are embedded into a weighted graph, which is subsequently simplified through graph straightening and used for kinematically constrained route optimization via augmented Dijkstra search and 2-opt refinement. The proposed method is evaluated on eight representative blasting areas from a real open-pit mine under six coverage radii ranging from 5 m to 30 m. The results show that TES consistently reduces parking point count and route length compared with a manual heuristic baseline, while maintaining favorable geometric characteristics and practical execution efficiency. In addition, curvature analysis and closed-loop Pure Pursuit simulations indicate that the generated routes are trackable and suitable for further smoothing-based refinement. Runtime evaluation further confirms that TES remains computationally practical for onboard deployment under typical operational settings. Overall, TES provides a geometrically interpretable and computationally efficient framework for joint parking-location selection and route planning in high-density, irregular blasting scenarios.

In the fields of 3D geological modeling and geoscience big data analysis, efficiently representing volumetric data characterized by massive scale, non-uniform spatial distribution, and complex geological features remains a core bottleneck constraining the performance of geophysical inversion, geostatistics, and spatial data analysis. To address the challenge of representing large-scale geological volumetric data, this paper proposes and implements a novel hierarchical sparse voxel data structure and its implementation framework—VoxOct. The core innovation of the VoxOct framework lies in its integration of the efficient indexing capability of sparse voxel grid with the flexibility of adaptive spatial partitioning offered by octrees, resulting in a hybrid sparse voxel octree (SVO) structure. VoxOct is a hybrid data structure based on octree spatial subdivision and N‑tree hierarchical compression, designed to achieve the representation, storage, and computation of large‑scale sparse geological volumetric data through the adaptive spatial division. In terms of design principles, VoxOct follows a topology‑attribute‑separated construction approach, which can drastically reduce storage space for invalid or redundant data through pruning optimization strategies. At the algorithmic level, VoxOct designs a set of efficient construction, dynamic traversal, and rapid update algorithms for large‑scale geological volumes. The framework implements an efficient sparse storage scheme and memory management strategy, supports native access to explicit octree nodes, and optimizes fast traversal and dynamic refinement algorithms for massive grids. Through an efficient sparse octree indexing mechanism, the framework enables high‑performance processing of billion‑scale voxel models on general‑purpose computing devices and facilitates dynamic updates to the octree grid model. Experimental results demonstrate that VoxOct reduces runtime memory access overhead compared with pointer-based octrees and requires less storage space than out-of-core octrees, which strikes a balance between the two for large and sparse geological volume data with high-resolution attribute distributions. Currently, the core algorithms of the framework have been integrated into commercial geological application software.

Dust significantly degrades LiDAR perception in industrial environments, particularly in mining sites. However, collecting real-world dust data is expensive and difficult to control, which severely constrains the development and validation of perception models. To overcome this limitation, we propose a lightweight dust generation method built on the open-source CARLA simulation platform. Without modifying the underlying engine or LiDAR sensor models, the proposed approach efficiently produces industrial-scene point clouds in which dust effects are explicitly perceivable. The method is based on a parameterized dust blueprint and is further integrated with a reflection-intensity post-processing strategy guided by real mining LiDAR data, enabling the synthesized point clouds to better match real observations in both spatial distribution and intensity statistics. Extensive experiments across multiple datasets and perception models demonstrate that the generated simulation data achieve stable cross-domain generalization on real mining scenarios as well as external public datasets, while effectively supporting the modeling of dust interference and engineering machinery structures. Overall, the proposed method offers an efficient and user-friendly data generation solution for LiDAR perception research in complex industrial environments.

Teleoperated open-pit excavation requires fast and reliable 3D scene modeling under lightweight sensor configurations. To this end, this paper proposes a monocular camera–LiDAR fusion-based fast 3D Gaussian reconstruction method tailored for teleoperated open-pit excavation. The proposed approach uses only two sensors, a monocular camera and LiDAR, and integrates SPNet, a depth completion network, to improve the geometric completeness of the reconstructed scene. It further introduces a stride-aware initialization strategy that leverages the depth–stride correlation to jointly construct the initial Gaussian set and estimate the initial scales. During optimization, scale and color regularization are applied to prevent uncontrolled growth of Gaussians. Experiments in a Carla-simulated open-pit excavation scenario show that, under high-resolution input of 1920 × 1080, the proposed method achieves a stable 3D model update rate of approximately 2.5 Hz. The reconstruction quality under training viewpoints reaches PSNR 30.5388, SSIM 0.9161, and LPIPS 0.1333. Compared with 4DTAM and MonoGS, the proposed method achieves better overall reconstruction quality. It also maintains a much higher update rate than 4DTAM and a comparable update rate to MonoGS. Ablation studies further verify the critical contribution of the depth completion module and the stride-aware initialization strategy to the overall reconstruction performance. In addition, preliminary validation on field data further demonstrates the applicability of the proposed method under real-world open-pit excavation-loading conditions. The proposed method generates stable and usable 3D models of rock-pile working face under a lightweight sensor configuration, providing a reliable geometric basis for remote situational awareness and excavation assistance.

To address challenges in automated charging operations of bulk explosive trucks in open-pit mines—specifically difficulties in borehole identification, positioning inaccuracies, and low operational efficiency—this study proposes an intelligent collaborative charging system integrating three modular components: (1) an explosive transport vehicle (with onboard terminal, explosive compartment, and mobility system enabling optimal routing and quantitative dispensing), (2) a charging robot (equipped with borehole detection, loading mechanisms, and mobility system for optimized search path planning and precision positioning), and (3) interconnection systems (coupling devices and interfaces facilitating auxiliary explosive transfer). This approach resolves three critical limitations of conventional systems: (i) mechanical arm-based borehole detection difficulties, (ii) blast hole positioning inaccuracies, and (iii) complex transport routing. The experimental results demonstrate that the intelligent cooperative charging method for open-pit mines achieves an 18% improvement in operational efficiency through intelligent collaboration among its modular components, while simultaneously realizing automated and intelligent charging operations. This advancement has significant implications for promoting intelligent development in open-pit mining operations.

Intelligent mining technology, as the core driving force for the digital transformation of the mining industry, integrates cyber-physical systems, artificial intelligence, and industrial internet technologies to establish a “cloud–edge–end” collaborative system. In this paper, the development trajectory of intelligent mining technology has been systematically reviewed, which has gone through four stages: stand-alone automation, integrated automation and informatization, digital and intelligent initial, and comprehensive intelligence. And the current development status of “cloud–edge–end” technologies has been reviewed: (i) The end layer achieves environmental state monitoring and precise control through a multi-source sensing network and intelligent equipment. (ii) The edge layer leverages 5G and edge computing to accomplish real-time data processing, 3D dynamic modeling, and safety early warning. (iii) The cloud layer realizes digital planning and intelligent decision-making, based on the industrial Internet platform. The three-layer collaboration forms a “perception–analysis–decision–execution” closed loop. Currently, there are still many challenges in the development of the technology, including the lack of a standardization system, the bottleneck of multi-source heterogeneous data fusion, the lack of a cross-process coordination of the equipment, and the shortage of interdisciplinary talents. Accordingly, this paper focuses on future development trends from four aspects, providing systematic solutions for a safe, efficient, and sustainable mining operation. Technological evolution will accelerate the formation of an intelligent ecosystem characterized by “standard-driven, data-empowered, equipment-autonomous, and human–machine collaboration”.

As mining operations extend to greater depths, the risk of deformation in high-stress tunnels increases significantly, posing a substantial threat. This study introduces a novel framework known as “robust mobility deformation detection” (RM2D), designed for real-time tunnel deformation detection. RM2D employs mobile LiDAR scanner to capture real-time point cloud data within the tunnel. This data is then voxelized and analyzed using covariance matrices to create a voxel-based multi-distribution representation of the rugged tunnel surface. Leveraging this representation, we assess deformations and scrutinize results through machine learning models to swiftly pinpoint tunnel deformation locations. Extensive experimental validation confirms the framework’s capacity to successfully detect deformations, including floor heave, side rib spalling, and roof fall, with remarkable accuracy. For deformation levels at 0.15 m, RM2D was able to successfully detect deformations with an area greater than 2 m2. For deformation areas of (3 ± 0.5) m2, RM2D successfully detected deformations of levels at (0.05 ± 0.01) m, and its detection capability meets the standard criteria for mining tunnel deformation detection. When compared to two conventional methods, RM2D demonstrates its real-time deformation detection capability in complex environments and on rough surfaces with precision, all at speeds below 10 km/h. Furthermore, we evaluated the predictive performance using multiple evaluation metrics and provided insights into the decision mechanism of the machine learning employed in our research, thereby offering valuable information for practical engineering applications in tunnel deformation detection.

As the demand for multi-scenario semantic updates in intelligent open-pit mining grows, existing semantic segmentation techniques encounter substantial challenges, particularly in handling visually indistinguishable, continuous geometric engineering objects. These approaches frequently fall short of meeting the accuracy standards required for practical applications. To tackle this challenge, the present study introduces a novel semantic segmentation method, CAD-to-PC matching and label propagation (CPMLP). The method constructs unified graph representations from 2D computer-aided design (CAD) drawings and 3D point clouds (PCs), extracts subgraph features using a graph-based feature encoding module, performs cross-graph alignment through contrastive learning, and propagates semantic labels from CAD to PC nodes. Experimental results demonstrate that CPMLP achieves high segmentation accuracy, with an average 3D mean Intersection over Union of approximately 0.7 across different acquisition routes, and robust generalization within complex mining environments. The proposed method provides a feasible solution for high-precision semantic segmentation in intelligent mining scenarios.

Accurate three-dimensional geological models are critical for intelligent coal mining. We present a 2.5-dimensional coal seam modeling method leveraging thin-plate spline interpolation, integrating fault network geometries and kinematic constraints. This approach targets laterally continuous, vertically stratified units (e.g., coal seams). The theoretical foundation draws an analogy from digital elevation surface modeling techniques. It uses a two-dimensional interpolation method to reconstruct the geological interface by performing spatial interpolation of three-dimensional vertex coordinates (including elevation attributes) projected onto two-dimensional parametric grids. Each geological layer (including the coal seam) is obtained through two-dimensional interpolation of the corresponding layer's borehole data. The key steps include constraint grid construction, stratigraphic sequencing, thin-plate spline interpolation, kinematic adjustment, and volumetric model generation. Compared with conventional three-dimensional modeling approaches, our method directly interpolates grid vertices, avoiding isosurface extraction to reduce computational complexity and enhance efficiency. The method explicitly incorporates kinematic parameters that govern fault displacement, which are capable of processing diverse fault types, while supporting stratigraphic modeling with complex fault networks. Pre-fault restoration during interpolation, combined with tectonic evolution forward simulation, ensures structurally valid stratigraphic models.

To address the problem of incomplete perception of limited viewpoints of ore blocks in future remote and intelligent shoveling-dominated mining scenarios, a method of using new view generation technology to predict ore blocks with limited view based on a latent diffusion model is proposed. Initially, an ore block image-pose dataset is created. Then, based on prior knowledge, the latent diffusion model undergoes transfer learning to develop an intelligent ore block shape prediction model (IOBSPM) for rock blocks. During training, structural similarity loss is innovatively introduced to constrain the prediction results and solve the issue of discontinuity in generated images. Finally, neural surface reconstruction is performed using the generated multi-view images of rock blocks to obtain a 3D model. Experimental results show that the prediction model, trained on the rock block dataset, produces better morphological and detail generation compared to the original model, with single-view generation time within 5 s. The average PSNR, SSIM, and LPIPS values reach 23.02 dB, 0.754, and 0.268, respectively. The generated views also demonstrate good performance in 3D reconstruction, highlighting significant implications for future research on remote and autonomous shoveling.

The simplification of complex networks is a research field closely related to graph theory in discrete mathematics. The existing methods are typically limited to simplifying the series sub-networks, parallel sub-networks, diagonal sub-networks, and nested simple sub-networks. From the current perspective, there are no available methods that can handle complex sub-networks and nested complex sub-networks. In this paper, we innovatively propose an efficient and automatic equivalence simplification method for arbitrary complex ventilation networks. The method enables, for the first time, the maximum possible equivalence simplification of nested simple sub-networks and nested complex sub-networks. In order to avoid the NP-hard problem caused by the searching of simplifiable sub-networks, it is necessary to analyze the intrinsic topology relationship between simplifiable sub-networks and spanning sub-graphs to optimize the searching process. One of our main contributions is that we present an efficient searching method for arbitrarily nested reducible sub-networks based on the bidirectional traversal process of a directed tree. The method optimizes the searching process for simplifiable node pairs by combining the characteristics of a directed tree with the judgment rules of simplifiable sub-networks. Moreover, by deriving the formula of an equivalent air resistance calculation for complex sub-networks, another one of our main contributions is that we present an equivalent calculation and simplification method for arbitrarily complex sub-networks based on the principle of energy conservation. The basic idea of the method is to calculate the equivalent air resistance using the ventilation network resolution of the constructed virtual sub-networks. We realize the simplification method of arbitrarily complex mine ventilation networks, and we validate the reliability of the simplification method by comparing the air distribution results using the network solution method before and after simplification. It can be determined that, with appropriate modifications to meet specific requirements, the proposed method can also be applicable to equivalent simplification instances of other types of complex networks. Based on the results analysis of several real-world mine ventilation network examples, the effectiveness of the proposed method is further verified, which can satisfactorily meet the requirements for simplifying complex networks.

A backfill system in underground mines supports the walls and roofs of mined-out areas and improves the structural integrity of mines. However, there has been a significant gap in the visualization and monitoring of the backfill progress. To better observe the process of the paste backfill material filling the tunnels, a LiDAR-based backfill monitoring system is proposed. As long as the rising top surface of the backfill material enters the LiDAR range, the proposed system can compute the plane coefficient of this surface. The intersection boundary of the tunnel and the backfill material can be obtained by substituting the plane coefficient into the space where the initial tunnel is located. A surface point generation and slurry point determination algorithm are proposed to obtain the point cloud of the backfill body based on the intersection boundary. After Poisson surface reconstruction and volume computation, the point cloud model is reconstructed into a 3D mesh, and the backfill progress is digitized as the ratio of the backfill body volume to the initial tunnel volume. The volumes of the meshes are compared with the results computed by two other algorithms; the error is less than 1%. The time to compute a set of data increases with the amount of data, ranging from 8 to 20 s, which is sufficient to update a set of data with a tiny increase in progress. As the digitized results update, the visualization progress is transmitted to the mining control center, allowing unexpected problems inside the tunnel to be monitored and addressed based on the messages provided by the proposed system.

When a fire occurs underground, pre-existing emergency escape routes may become ineffective. Such scenarios necessitate the real-time planning of escape routes that consider evolving conditions and prioritize safety. This paper proposes a multi-objective real-time search method for emergency escape routes using dynamic programming to address these challenges. Firstly, this paper discusses how the evacuation of an underground mine fire caused by external factors depends on the fire monitoring system, personnel positioning system, and emergency evacuation system, as well as the modeling and solution methods of evacuation route planning. Then, based on fire smoke, CO, and temperature sensor data from the mine ventilation network structure and monitoring system, the possible smoke spread in the event of an underground fire was calculated. The objectives are to optimize both the shortest equivalent length of the path and the shortest time to walk through the smoke flow while ensuring that temporary escape time does not exceed the rated protection time of the self-rescuer. A mathematical model for emergency escape route planning is established under these conditions. A labeling algorithm based on dynamic programming is employed to find the Pareto optimal solution set of emergency evacuation routes that meet emergency requirements. Finally, two path evaluation indicators, namely “escape target priority” and “personnel temporary escape time”, are introduced to re-rank the solutions in the Pareto optimal set, thereby obtaining disaster evacuation routes with different priorities. Example verification shows that the algorithm can quickly solve the disaster evacuation routes that meet the actual disaster evacuation needs in complex networks. Example verification shows that the algorithm can quickly solve the emergency escape routes in complex networks that meet actual emergency escape needs.

The dust emitted from open-pit mines poses a constant menace to both the health of the workers and the environment. However, the traditional manual decision-making and scheduled watering methods are constrained by delayed dust control and excessive water resource consumption, making it challenging to fulfill the requirements of green mining and environmental protection in mining operations. To address the issue of unmanned dust monitoring, an intelligent dust recognition technology based on deep learning was proposed. This technology introduces an innovative approach for tracking and recognizing dust-emitting areas based on a vehicle tracking algorithm. A cascaded deep learning architecture YRCNet (YOLOv5 Tracking with ResNet-50 Classification Cascade Network) was developed, which consists of two main parts: vehicle dynamic tracking and the concentration classification. The YOLOv5 is employed for real-time tracking and capturing images of the rear of vehicles, which enhances the interference resilience of dust image recognition by implementing initial screening of target presence areas. Following this, an optimized ResNet-50 network is used for the classification and recognition module. It employs spatial pyramid pooling layers integrated with deep convolutional blocks, to enhance dust feature extraction and enable precise classification of dust concentrations into high, medium, and low levels. To train this deep learning model, simulated dust images of different concentration levels are generated using simulation techniques and combined with field-collected mining data to create a well-balanced open-pit mining road dust mixed dataset that meets the requirements for model training. The results show that YRCNet achieves a recognition accuracy of 94.25% on mixed dust dataset, representing a 13.58% improvement over the previous network. The model exhibits generalization across various open-pit mining road scenarios, providing an effective perceptual means for intelligent dust reduction. Additionally, it contributes to the monitoring and enhancement of environmental quality in mining areas, ensuring the health and safety of the workforce.

Mine ventilation is crucial for ensuring safe production in mines, as it is integral to the entire underground mining process. This study addresses the issues of high energy consumption, regulation difficulties, and unreasonable regulation schemes in mine ventilation systems. To this end, we construct an optimization model for mine ventilation network regulation using mixed-integer nonlinear programming (MINLP), focusing on objectives such as minimizing energy consumption, optimal regulation locations and modes, and minimizing the number of regulators. We analyze the construction methods of the mathematical optimization model for both selected and unselected fans. To handle high-order terms in the MINLP model, we propose a variable discretization strategy that introduces 0-1 binary variables to discretize fan branches’ air quantity and frequency regulation ratios. This transformation converts high-order terms in the constraints of fan frequency regulation into quadratic terms, making the model suitable for solvers based on globally accurate algorithms. Example analysis demonstrate that the proposed method can find the optimal solution in all cases, confirming its effectiveness. Finally, we apply the optimization method of ventilation network regulation based on MINLP to a coal mine ventilation network. The results indicate that the power of the main fan after frequency regulation is 71.84 kW, achieving a significant energy savings rate of 65.60% compared to before optimization power levels. Notably, ventilation network can be regulated without adding new regulators, thereby reducing management and maintenance costs. This optimization method provides a solid foundation for the implementation of intelligent ventilation systems.

Real-time and accurate bucket pose estimation plays a vital role in improving the intelligence level of mining excavators, as the bucket is a crucial component of the excavator. Existing methods for bucket pose estimation are realized by installing multiple non-visual sensors. However, these sensors suffer from cumulative errors caused by loose connections and short service lives caused by strong vibrations. In this paper, we propose a method for bucket pose estimation based on deep neural network and registration to solve the large registration error problem caused by occlusion. Specifically, we optimize the Point Transformer network for bucket point cloud semantic segmentation, significantly improving the segmentation accuracy. We employ point cloud preprocessing and continuous frame registration to reduce the registration distance and accelerate the Fast Iterative Closest Point algorithm, enabling real-time pose estimation. By achieving precise semantic segmentation and faster registration, we effectively address the problem of intermittent pose estimation caused by occlusion. We collected our own dataset for training and testing, and the experimental results are compared with other relevant studies, validating the accuracy and effectiveness of the proposed method.

Open-pit mines are becoming more intelligent and unmanned. One of the core technologies for autonomous driving is high-definition map technology, while refined mine maps are also needed to support scenarios such as vehicle–shovel collaboration, loading and unloading, production scheduling, and traffic safety control. To better serve the construction of intelligent mines and address the autonomous driving needs in open-pit mining scenarios, a universal high-definition map data model oriented toward data production and exchange, Open-Pit Map (OP map for short), has been designed. Based on various working scenarios of open-pit mines, eight practical basic elements, eleven location elements, and eight traffic elements in OP map were proposed. A map visualization tool, MapViewer, was developed based on OP map, and the effectiveness and practicality of the data model in the practical application of an intelligent shipping system in a Chinese mine were validated.