Next Article in Journal
Influence of Mining Layout on Efficiency of NRE Drill Rig
Next Article in Special Issue
Optimization of the Design of Underground Hydrogen Storage in Salt Caverns in Southern Ontario, Canada
Previous Article in Journal
Social Acceptability of Critical and Strategic Minerals (CSMs) Development: A Systematic Review with a Particular Focus on Quebec, Canada
Previous Article in Special Issue
Evaluation of Socioeconomic Dynamics and Their Impact on Life Expectancy in Coal Mining Communities in Colombia
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Envisioning Human–Machine Relationship Towards Mining of the Future: An Overview

by
Peter Kolapo
1,*,
Nafiu Olanrewaju Ogunsola
2,3,
Kayode Komolafe
4 and
Dare Daniel Omole
5
1
Geotechnical Engineer, Universal Engineering Service, Lexington, KY 40505, USA
2
Department of Mining Engineering and Mine Surveying, University of Johannesburg, Johannesburg 2006, South Africa
3
Sibanye-Stillwater Centre for Sustainable Mining, University of Johannesburg, Johannesburg 2006, South Africa
4
Department of Mining Engineering, College of Engineering, Colorado School of Mines, Golden, CO 80401, USA
5
Midgate Global Resources Ltd., Lagos 100223, Nigeria
*
Author to whom correspondence should be addressed.
Submission received: 10 December 2024 / Revised: 31 December 2024 / Accepted: 2 January 2025 / Published: 6 January 2025
(This article belongs to the Special Issue Envisioning the Future of Mining, 2nd Edition)

Abstract

:
Automation is increasingly gaining attention as the global industry moves toward intelligent, unmanned approaches to perform hazardous tasks. Although the integration of autonomous technologies has revolutionized various industries for decades, the mining sector has only recently started to harness the potential of autonomous technology. Lately, the mining industry has been transforming by implementing automated systems to shape the future of mining and minimize human involvement in the process. Automated systems such as robotics, artificial intelligence (AI), the Industrial Internet of Things (IIOT), and data analytics have contributed immensely towards ensuring improved productivity and safety and promoting sustainable mineral industry. Despite the substantial benefits and promising potential of automation in the mining sector, its adoption faces challenges due to concerns about human–machine interaction. This paper extensively reviews the current trends, attempts, and trials in converting traditional mining machines to automated systems with no or less human involvement. It also delves into the application of AI in mining operations from the exploration phase to the processing stage. To advance the knowledge base in this domain, the study describes the method used to develop the human–machine interface (HMI) that controls and monitors the activity of a six-degrees-of-freedom robotic arm, a roof bolter machine, and the status of the automated machine. The notable findings in this study draw attention to the critical roles of humans in automated mining operations. This study shows that human operators are still relevant and must control, operate, and maintain these innovative technologies in mining operations. Thus, establishing an effective interaction between human operators and machines can promote the acceptability and implementation of autonomous technologies in mineral extraction processes.

1. Introduction

In recent years, the mining industry has witnessed an increase in the introduction of automated technologies to enhance mine worker’s safety and increase productivity. Due to the demanding operational conditions in underground mines, many of the developed systems are proffering technical solutions to inherent mining issues in health and safety, productivity, and sustainability. Moreover, the rapid rise in global demand for critical and rare earth minerals for industrial growth and various technological advancements necessitates implementing innovative and digital systems. The implementation is promising in solving several current issues in the mining sector. In most cases, automated technologies mimic human intelligence to perform human functions in the mining value chain. Notably, the introduction brought about a change in workflow between humans and machines showing how they collaboratively work together to achieve a common goal set by humans. These interactions open up new possibilities and synergies that can significantly enhance our ability to carry out specific tasks.
Mining is regarded as one of the most hazardous occupations due to the number of accidents and injuries recorded in underground mining operations where personnel are exposed to an unsupported roof, dust inhalation, high temperature, poisonous gases, and noise levels above the recommended limits [1]. Besides this, the harsh mining conditions in underground mines with long working hours are the leading cause of fatigue among miners, which can result in accidents and injuries. To address this, miners should move away from hazardous working conditions and allow autonomous technology that can be remotely controlled using HMI devices, rather than directly operating the equipment. An HMI device connects humans and machines, allowing smooth communication between the operator and the system. The introduction of autonomous systems in mineral extraction processes enables machines to perform human tasks. Although these tasks are limited, humans still control the action of automated technologies. Human operators manage and control the automated machines using the HMI device to override the actions of automated systems. The main purpose of the HMI is to help operators control and guide the behavior of automated systems, giving them an interface to manage these systems directly. The HMI controls the automated machines using the digital input and output (I/O) signals to regulate the force that is exerted by the actuator to flip switches, activate and engage gears and joysticks, turn the steering wheel, and prevent the need for pressing and releasing pedals. The I/O module is a key component of autonomous technologies that enables seamless connectivity and control of various parts, such as sensors and actuators.
This study aims to review automated technologies in the mining industry and explore the potential ergonomic impact of these innovative technologies. No study in the literature succinctly discusses the human–machine relationship in the context of mining while combining automation, digitization, and artificial intelligence. This study will be a go-to reference for mining engineering practitioners, stakeholders, industries, and governments to implement autonomous technologies for sustainable growth and development in the mining industry. The remaining sections of this study are organized as follows: Section 2 explains the global trends of implementing autonomous and robotic technology in mining operations and various attempts and trials of converting traditional mining machines to autonomous equipment are illustrated. Section 3 discusses applications of artificial intelligence in rock engineering. Section 4 explains the techniques used to enhance human–machine collaboration utilizing a case study of automating a traditional roof bolter machine. The design of a user-friendly interface for an automated roof bolting process is extensively discussed in this section. Section 5 analyzes the benefits and challenges facing the implementation of autonomous technology in both surface and underground mining processes. The paper ends with a short conclusion and outlook in Section 6.

2. Trends of Automation and Implementation of Digital Systems in Mineral Industry

Automation and digitization in mining operations are mutually supportive trends. The application of autonomous systems, ubiquitous sensing technologies, and artificial intelligence are gaining attention, becoming popular, and increasingly being deployed in various engineering and industrial fields to carry out tasks. These technologies are widely used in industries such as manufacturing, agriculture, healthcare, traffic management, and intelligent transportation systems, enabling operations, such as welding, material handling, spray coating, assembly, and surface finishing. In today’s technological market, innovative autonomous technologies have emerged, presenting advanced capabilities that enhance various aspects of industrial processes. According to Ralston et al. [2], these new technologies are primarily designed to collaborate with human operators and perform hazardous tasks to alleviate the workload and ensure safety. Due to the monotonous nature of mining operations, the collaborative robot (cobot) is the transformation needed to perform a repetitive and hazardous task, thereby improving the safety of miners. Autonomous machines are often equipped with cameras, sensors, and robust vision systems to detect the surroundings and perform unstructured tasks. Javaid et al. [3] stated that the integration of these sensors enables the machines to share space with humans and guarantee their safety.
The implementation of autonomous machines within the mining value chain is expected to grow as mining companies are intensifying their efforts to achieve zero harm in their operations. In recent years, the global mineral industry has witnessed the deployment of various autonomous technologies to perform mineral extraction tasks such as dozing, excavation, drilling and blasting, and transporting fragmented rock. Existing autonomous mining technologies have shown substantial improvements, contributing to higher productivity and lower operational costs. According to Paredes and Fleming-Muñoz [4], the notable improvement in mining productivity resulting from implementing adaptive automated systems motivates the mineral industry to deploy more autonomous systems in their operations.
In the last decade, mining operations around the globe, especially in the United States, have undergone substantial transformations, with increasing autonomy driven by the integration of digital systems designed to enhance miner safety and optimize productivity. Some of the equipment garnering significant attention in the extractive industry includes autonomous drilling technology, smart sensor systems, long-distance haulage trucks, loaders, explosive charging and rock fragment systems, 3D LiDAR systems, drones, and automated monitoring technologies. Lopes et al. [5] acknowledged that the increasing adoption of autonomous technologies in mining operations indicates a shift where mining companies are moving from manual labor to fully autonomous operations. A noticeable innovation in the mining industry in recent years is the incorporation of smart sensing devices and monitoring systems in traditional mining machines to work autonomously. This technique converts conventional mine machines to automated mining equipment. Integrated sensors in autonomous systems provide real-time data acquisition, enabling sound decisions regarding positioning, recognizing surrounding objects, autonomous steering control, and equipment performance. Examples of these sensors include ultrasonic sensors for distance measurement, LiDAR for autonomous equipment mapping and navigation, and radar for determining the location, angle, and velocity of objects. Sam-Aggrey [6] acknowledged that the main function of integrated smart sensors in autonomous machines is to send information about the engine’s functioning and systems, and issue alerts when maintenance is needed. It is important to state that a robust communication protocol is essential for the effective operation of smart sensors in underground mines and for ensuring reliable underground Wi-Fi connectivity. Rowduru et al. [7] stated that optimal sensor performance and accurate results are supported by favorable environmental conditions. However, most underground mining environments are characterized by extreme temperatures and challenging working conditions. Additionally, changes in weather patterns and climate change can significantly influence the functionality, response time, accuracy residual life, and reliability of autonomous equipment. Thus, conducting regular preventive maintenance, testing, diagnostics, and technical analyses is crucial to maintaining the reliability and optimal performance of the installed digital systems.
It is imperative to state that introducing autonomous machines into the mining value chain does not eliminate human involvement in mining operations; rather, it reallocates humans to new roles that may require the development of additional skills. Therefore, adopting autonomous equipment is expected to increase the demand for a highly skilled labor force in the mining industry. The autonomous system will require a workforce with the appropriate background and technical expertise, including skills in computing and coding, to operate the automated technologies. Unfortunately, the implementation of autonomous machines in underground mining operations remains limited in comparison to other engineering fields [8]. However, there have been several attempts by researchers and mining companies to automate various mining equipment by the integration of smart sensors and digital systems into the components. For instance, there have been growing efforts to automate loading, hauling, and dumping (LHD) vehicles driven by rapid advancements in the research and development of underground mining machinery. In the study of Larson et al. [9], they proposed a fully automated navigation system for underground LHD vehicles, utilizing a fuzzy behavior-based approach for navigation. A notable success was recorded by Larsson et al. [10] to automate LHD machines through the installation of smart sensors and control systems for autonomous navigation of the LHD vehicle. Similarly, Gu et al. [11] proposed an optimal trajectory planning method for LHD vehicles during turning maneuvers to minimize driving time.
Recently, there have been studies on how to automate conveyor belts to move mined material from the working face to a processing plant. In the works of Rocha et al. [12] and Szrek et al. [13], they developed an autonomous conveyor belt monitoring system using robotic technology to provide real-time status updates, detecting potential failures before the belt breaks. This robot is equipped with an RGB camera to visualize the conveyor path and identify hotspots using infrared thermography. The video feed is wirelessly transmitted via a 5.8 GHz frequency channel to a remote-control panel. During belt failures, the robot’s GPS module enables precise failure identification. In underground mines, where GPS signals are unavailable, an ultra-wideband (UWB) transceiver-based localization system is employed for accurate positioning.
In advanced integrated sensors used in mining machinery, global navigation satellite system (GNSS) plays a crucial role in automating mining operations. One of the common applications of GNSS is the self-positioning of mining equipment and providing guidance for navigation in surface mining environments. Additionally, it reduces accidents caused by low visibility and blind spots around trucks by providing 3D situational awareness of the truck’s surroundings, using GNSS data. Over the years, researchers have studied GNSS integration in mining equipment for machine tracking and management. For Instance, Gaber et al. [14] installed GPS on driverless autonomous haulage system (AHS) for autonomous control and navigation, while Chaowasakoo et al. [15] utilized GPS to manage the truck fleet at the Adaro coal mine in Indonesia. However, the lack of satellite signals in underground mining environments presents a challenge in generating accurate road maps. To address this, researchers developed the simultaneous localization and mapping (SLAM) algorithm that provides machine location and position on a map. In the study of Thrun et al. [16] and Nüchter [17], SLAM algorithms were used to successfully map an abandoned mine. Similarly, Lösch et al. [18] combined SLAM with other technologies such as the inertial measurement unit (IMU), an autonomous robot, a lighting system, two RGB cameras, a laser scanner, and a SLAM algorithm to navigate underground mining operations.
Another evolving autonomous technology in the mineral industry is the usage of robotic arms and collaborative robots to perform repetitive and dangerous tasks such as explosive charging, handling toxic chemicals during mineral processing, and installing roof bolts in underground mines. The robot’s precision, strength, and endurance in task execution help minimize the physical and cognitive demands on humans. These attributes enable operators to focus their decision-making skills on value-added tasks. Recently, Asea Brown Boveri (ABB) partnered with Boliden and LKAB mines to develop an automated explosive charging robot that combines machine vision and a robotic arm. The technology is designed to identify boreholes in the rockface and then prime them with explosive charges [19]. Similarly, CSIRO in Australia developed an automated vehicle, the robotic explosive charging system (RECS), for charging blast holes [20]. Instead of the traditional approach of pumping explosives into a drill hole; the technology performs the entire sequence of charging the hole with explosives.
Notable progress has been made in automating the drilling process in mining operations. For instance, in 2018, Rio Tinto deployed twenty autonomous drill systems (ADS) capable of remote operation at the Pilbara mines in Western Australia. The main goal is to eliminate human involvement by enabling autonomous drilling fleets to perform drilling tasks independently. These fleets are equipped with advanced technologies, including GNSS and machine learning algorithms to perform autonomous drilling processes, as shown in Figure 1. The automated drill rig utilizes satellite technology to collect and transmit data from onboard sensors, enabling the monitoring and measurement of drilling metrics. This reliable satellite connectivity also facilitates precise positioning and seamless remote operations. The development of autonomous drilling technology improves drilling accuracy for optimized blasts, enhances fragmentation, increases penetration rate, and reduces CO2 emissions through its integrated battery–electric drivelines.
Notable advancements have been achieved in the automation technologies for longwall systems in underground mining operations. For instance, the horizon control, which ensures the shearer’s cutting gradient aligns with the coal seam gradient, is a critical technique in automating longwall mining. The work of Peng et al. [21] introduced a gamma-ray sensor for measuring coal seam thickness in longwall automation to determine the shearer’s exploitation gradient. Ralston et al. [2] implemented inertial navigation technology in longwall machines to enable precise positioning and guidance within underground environments. Similarly, Billingsley & Brett [22] used an inertial navigation system (INS) to determine the shearer path in three-dimensions. The application of some innovative technologies in mining operations as discussed is presented in Table 1.
Conclusively, the adoption and integration of automated technologies across the mineral industry value chain have experienced exponential growth, driven by the significant advantages these systems have delivered since their inception. The increase in the rate of integrating these technologies indicates that mining companies are gradually moving from manual labor to autonomous machinery as solutions to improve safety, increase productivity, and reduce mining costs. These benefits cut across all mining phases including prospecting, exploration, development, excavation, and reclamation.

3. Application of Artificial Intelligence in Mining Operations

This section discusses the application of artificial intelligence (AI) technologies in different areas of mining operations. AI is a field within computer science that focuses on developing models capable of performing tasks that typically require human intelligence. It is often used interchangeably with machine learning (ML) or soft computing (SC) techniques. However, ML or SC is a subset of AI and is grouped into four groups: supervised, semi-supervised, unsupervised, and reinforced learning [27,28]. ML or SC is a data-driven technique and comprises various algorithms such as artificial neural networks (ANN), support vector machines (SVM), gene expression programming (GEP), etc., used in developing predictive models.
The mining industry is undergoing a significant transformation driven by advancements in AI. The application of AI in mining operations offers transformative potential, improving efficiency, safety, and sustainability. The various applications of AI and ML technologies in mining engineering are discussed in the following subsections. The discussion is under three headings: mineral prospecting and exploration, mineral excavation, and mineral processing.

3.1. Mineral Prospecting and Exploration

Mineral prospecting and exploration are the first stages of mining, which involve searching for and qualitative appraisal or valuation of minerals and ore-bearing rocks. ML techniques have recently been employed in mineral prospecting and exploration to replace the often laborious and time-consuming conventional methods. Employing ML techniques in mineral prospecting and exploration ensures a faster and easier means of searching and quantifying mineral deposits. AI techniques for prospecting and exploration were studied to delineate various minerals and ores [29,30,31,32,33,34,35,36,37]. An IBM Watson AI supercomputer was used to interpret the exploration data obtained from borehole drilling in a gold mineralized zone in the Red Lake mine in Canada to delineate new gold deposits [38]. Zhang and Zhou [30] developed a mineral potential map for prospecting and exploring gold deposits in the western Junggar area of Xinjiang Province, China, using weights-of-evidence (WofE) and fuzzy logic models and compared the results. Their results showed that the data and expert knowledge-driven fuzzy logic model displayed a strong and superior correlation between areas of high probabilities and known gold deposits compared to the weights-of-evidence model.
The next operation after mineral prospecting and exploration includes mine cost estimation and budgeting, mine design, and scheduling of mining activities. These operations are usually laborious and time-consuming when done manually or conventionally. Introducing AI and ML in these domains can help develop and implement time-efficient mine plans and designs, optimize production efficiency, and ensure compliance with safety and environmental standards. As an alternative to the conventional resource and reserve estimation methods (geometric and geostatistical), ML techniques have been applied to estimate grades of ores with high prediction accuracy [39,40,41,42,43,44,45,46]. Afeni et al. [42] employed geostatistics (ordinary kriging) and artificial neural network (ANN) techniques to estimate the reserve of Itakpe iron ore in Nigeria using exploration data from borehole drilling. Their results showed the suitability of ANN for the grade estimation of the Itakpe iron ore. For mine planning and scheduling, ML techniques have been used to accurately estimate mine operation expenses and to operate mine design [47,48,49,50,51]. Some applications of ML models in mineral prospecting and exploration including mine cost estimation and budgeting, mine design, and scheduling are listed in Table 2.

3.2. Mineral Excavation

Mineral excavation involves extracting valuable minerals and ores from the Earth’s crust. The term encompasses a variety of operations or phases, including the exploration, extraction, processing, and management of these resources. However, this review categorized mineral excavation as rock or geotechnical testing before mining, drilling and blasting, mine safety and environmental pollution, and reclamation and closure.
The first stage of mineral excavation is geomechanical testing of the rocks or geomaterials. This stage is important because the rock’s geomechanical properties determine the selection of mining methods and equipment, the type of explosives for blasting, and the evaluation of mine slope safety and stability [53]. AI and ML technologies have been successfully employed to estimate rock properties such as uniaxial compressive strength, fracture toughness, tensile strength, modulus of elasticity, the factor of safety of slopes, etc. [54,55,56,57,58,59,60]. Ogunsola et al. [55] employed ANN and metaheuristic algorithm-optimized ANN models to estimate the Mode-I fracture toughness of andesite and sandstone. The proposed optimum model achieved a high prediction accuracy with a coefficient of correlation (R) of 0.985, mean square error (MSE) of 0.003, and a 20-index of 0.967. In another study, Mahmoodzadeh et al. [60] developed a predictive model involving Gaussian process regression (GPR), support vector regression (SVR), decision trees (DT), and long-short-term memory (LSTM) to estimate the shear strength parameters such as cohesion (C) and the friction angle (φ) of rock. The results of their study showed that the LSTM model has the highest accuracy with a coefficient of determination (R2), root (MSE), and mean absolute percentage error of 0.984, 1.295, and 0.009, respectively, for C, and 0.854, 1.857, and 1.430, respectively, for φ.
Drilling and blasting are important operations in the extraction of minerals. It involves drilling holes in the rock to accommodate explosives for the fragmentation of the rock. Achieving fragmentation is the goal of blasting rocks; however, a percentage of the explosive energy causes adverse environmental effects such as blast-induced ground vibration, air overpressure, back-break, and noxious gases [61,62]. ML techniques have been used to predict positive and negative mining outcomes with remarkable results [63,64,65,66,67,68]. Amiri et al. [67] developed prediction models to predict and minimize blast-induced vibration and air overpressure in a mine in Iran using ML methods of ANN, K-nearest neighbors (KNN), and an empirical model. Their study demonstrated the superiority of ML models over the conventional empirical model. In another study, Amoako et al. [68] predicted rock fragmentation from different mines using ANN and support vector regression (SVR) and compared their results with the conventional Kuznetsov model. Their results showed that the two ML techniques achieved higher performance when compared with the Kuznetsov model.
Mine reclamation and closure are critical phases in the lifecycle of a mining operation, focusing on restoring the land and mitigating environmental impacts after the extraction of minerals is complete. These processes are essential to ensure that the mining site does not pose long-term ecological hazards and can be used for other purposes in the future. Samadi et al. [69] proposed a model to evaluate post-closure strategies for open pit mine reclamation using fuzzy cognitive mapping and evidential reasoning. Maxwell et al. [70] proposed a coupled ML, satellite imagery, and light detection and ranging (LiDAR) model to classify mining and mine reclamation. ML techniques of SVR, random forests (RF), and boosted classification and regression trees were employed to classify the satellite imagery and LiDAR data obtained from the mine. Their results showed the suitability of ML techniques to classify mining and post-mining land reclamation. Some applications of ML models in mineral excavation are listed in Table 3.

3.3. Mineral Processing

Mineral processing is an essential stage in the mining value chain and relates to the various physical and chemical methods to achieve this goal. It involves separating and concentrating valuable minerals from raw ore containing waste materials after extraction to produce a high-grade mineral for industrial use. Various ML techniques have been employed with high prediction accuracy [73,74,75,76,77]. Hosseini et al. [77] demonstrated the suitability of ANN in predicting the iron, phosphor, sulfur, and iron oxide content of the final concentrate from an iron plant. The input parameters for their developed model include iron, phosphor, sulfur, and iron run-of-mine (ROM) percentages. Their results showed high prediction accuracy with an R2 of 0.98, and can be used to accurately determine the effects of changes in feed and concentrate grade of iron ore plants in dry and wet magnetic processes. In a related study, Jahedsaravani et al. [78] proposed ML models of ANN and adaptive neuro-fuzzy inference system (ANFIS) to estimate the batch floatation of copper sulfide ore. The input parameters for the models were the process variables (i.e., gas flow rate, slurry solids%, frother/collector dosages, and pH) and the metallurgical parameters (i.e., copper/mass/water recoveries and concentrate grade). Their results showed that ML techniques are more efficient tools for modeling complicated processes like flotation, which are essential for developing model-based control systems. Some applications of ML models in mineral processing are listed in Table 4.

4. Enhancing Human–Machine Collaboration: A Scenario of Roof Bolting Machine Automation

Advancements in technology have led to significant breakthroughs over the years in mitigating machine-related accidents in mining operations. The global mining industry is witnessing a trend of continuous adoption of digital technologies in their operations due to the demanding operational conditions in underground mining operations. Many of the implemented systems are proffering technical solutions to inherent mining issues in mine health and safety, productivity, and sustainability. Despite the significant impact of autonomous technologies, zero harm has not been achieved in mining operations. This is a major concern for mining companies and equipment manufacturers as they seek to achieve zero harm in mining operations. Therefore, enhancing collaboration necessitates the development of an efficient human–machine interface. The introduction of automated machines enables humans to delegate specific tasks to them; however, this does not imply a full transfer of operational responsibilities; humans still control the machines. Human–machine collaboration is an essential part of autonomous systems that provide the control through which the user operates the machine. The primary function of the human–machine device is to make the functions of the machine self-evident to the operator. The HMI presents operators with a control interface for automated systems. The primary function of the HMI is to monitor and control the behavior of automated systems. A well-designed HMI presents the user with all the buttons and images of the task. The device manages the system without direct manual interaction, such as flicking switches, engaging gears, using joysticks, turning the steering wheel, or pressing pedals. Kolapo [79] acknowledged that the HMI provides real-time, reliable information about the machine’s behavior through a graphical user interface (GUI), allowing the operator to assess its status, monitor performance, and adjust operational parameters accurately. The effectiveness of the HMI can influence the acceptability of the entire system; in many applications, it can impact the product’s overall success or failure.
The design and development of an effective and efficient HMI device allow the operator to manage the activity of an automated roof bolting machine safely from a remote location. Kolapo et al. [80] described the rockbolt as a plain bar installed in a rock mass to prevent loosened key blocks from moving from their positions by bonding them together. Traditionally, the roof bolting installation involves performing tasks that can pose risk to the safety of operators. This includes working under unsupported roofs, loose bolts, and heavy spinning metal, all of which can lead to fatal accidents or injuries. In most cases, the accident records during roof bolting installation can be attributed to the challenging conditions in underground mining operations. These conditions include elevated temperatures and humidity, toxic and combustible atmospheres, high noise levels from equipment, and the generation of hazardous dust particles. Apart from challenging working conditions, operators face increased fatigue due to long exposure to hot temperatures, limited visibility, long work hours, early morning awakenings, and extended commute times resulting from the remote location of mine sites. Therefore, it is necessary to remove operators from these hazardous conditions and enable smart technologies to handle the dangerous and repetitive tasks, while human operators oversee the entire bolting process. Approaching converting a manually operated roof bolting machine into an autonomous system involves integrating a robotic arm into the machine, enabling it to replicate and enhance human tasks during the roof bolting process. The goal integrating a robotic arm is to move humans away from the roof bolting process and allow the robot to perform human functions such as moving, lifting, grasping, and positioning of consumables for the roof bolting process.
To enable the robot to autonomously perform the roof bolting process, executable task plans and motion paths were developed. As detailed in the work of Zhang et al. [81], task and motion planning techniques were integrated to execute functions such as installing drill steels through a sequence of robot-executable motion paths. Algorithms were developed to generate task and motion plans for the ABB robot, enabling it to autonomously handle drill steels, bolts, and simulated pumpable resins. The objective was to achieve a seamless collaboration between the ABB robot, the roof bolter module, and other integrated systems to fully automate roof bolting. The cobot system selected for this study is ABB robot IRB 1600. This robotic arm is made up of two main parts, namely the manipulator and the controller. The body of the robotic arm consists of links, joints, and other structures with a net weight of 250 kg (551.156 lbs) as shown in Figure 2.
It is important to state that the implementation of robotic arms will not completely eliminate human involvement but will instead redefine their roles. That is, the tasks that were done manually by human operators are now achieved by robotic systems with little or no human interventions.
The method used for integrating robotic system into an existing roof bolting machine, resulting in the development of an autonomous roof bolter machine was extensively discussed by Kolapo et al. [82]. The work highlights some of the factors to be considered when automating underground machines. However, the study states the importance of establishing a robust communication path to facilitate seamless interaction between different hydraulic, mechanical, and electrical components. In the context of underground mining equipment, the HMI is designed to facilitate communication and interaction between human operators and the automated machines when working in subsurface structures. The HMI device enables humans to interrupt or override the action of the technology remotely or in a safe place. The operator monitors and controls the bolting process through the HMI device. In the context of the automated roof bolting process, the HMI device is linked to the programmable logic controller (PLC) through an iQAN program enabling the operator to override the robot’s action in the event of unexpected and unsafe activities.
The design of a human–machine device depends on the functions and tasks to be performed by the machine. The HMI system is designed with an operator-friendly interface to enhance acceptance of the technology by the users. In general, the typical features of HMI should include touchscreen operation, button or keypad features, pages with navigation for more functions, and logs for alarms and events as shown in Figure 3. Due to the working conditions of underground mining operations, it is essential to consider other factors such as operating temperature and humidity, vibration and shock rating, dust, and a combustible atmosphere.
Contrarily, a poorly designed HMI can lead to system malfunctions, operational downtime, and become a major cause of accidents and fatalities in autonomous machine work environments. According to Gruhn [83], poorly designed HMIs can hinder operators instead of assisting them, as the device fails to provide the necessary information for the operator to effectively work with the automated machine. In fact, the design of the human–machine device can impact the overall success or failure of the automated machine. Therefore, manufacturers of automated machines should emphasize the design of HMI, the adoption of emerging technologies, and the evolving skill requirements for operators and maintenance teams, shifting focus away from traditional priorities like manual tasks and environmental ergonomics.

4.1. Factors to Be Considered for an Effective HMI Design

The primary goal of the HMI is to connect the operators to the automated machine. The device presents an interface that sends commands to the automated systems and manually controls the machine’s operations in response to unplanned events. The HMI designer should develop an interface that enables operators to effectively and efficiently monitor autonomous operations. These include an arrangement of features and layout considerations, style, color, tactile feedback, and ergonomics. This results in optimal operator experience, which is crucial for customer satisfaction with the system. The display information can include graphs, charts, dashboards, and alarms to facilitate monitoring of work progress and operational feedback. Understanding the operators’ goals, tasks, workflow, and cognitive processes is crucial for determining the optimal input requirements for the task. It can also significantly impact HMI design considerations. In a mining context, a comprehensive understanding of (1) mining methods, (2) geological formations, (3) mine design and planning, (4) ventilation, and (5) underground mining standards, regulations, and workflows are essential factors for effective HMI design. For instance, in the automated roof bolting process, the first task is for the robotic arm to grasp the drill steel and position it for drilling. The manufacturer must understand the manual machine’s workflow to effectively design the automated machine’s HMI. According to Rockwell Automation [84], for an effective HMI, the interface should be designed in such a way that the operator would understand the mode of operations and how information is related. It should facilitate operators’ rapid comprehension, demonstrate the necessary information for the task, and reduce visual complexity.
In order to enhance clarity and avoid visual clutter and unnecessary information, grouping and presenting only essential data in a well-organized format is crucial. This approach increases reliability and assists operators in effectively monitoring operation status and task execution success. The touchpad should display only task-specific information relevant to the operator’s current activity. Presenting extraneous data can lead to increased response times and a higher likelihood of errors. To improve the HMI, it is essential to translate data effectively into visual feedback and highlight vital information that requires the operator’s attention. Also, organizing similar information provides operators with visual clues, values, and the state of the automated equipment. The interface layout should allow operators to rearrange and organize controls based on anticipated usage, and grouping related controls together logically. This can be accomplished by strategically placing related items close to each other, using lines, or employing background shading. These factors were used as the guidelines for developing an effective HMI for the automated roof bolting machine.

4.2. Designing HMI for an Automated Roof Bolting Machine

The HMI design for the automated roof bolting machine is tailored to the specific tasks involved in roof bolting operations. The aim is to develop an interface that allows the operator to communicate with the roof bolter machine, the robotic arm, and other components for autonomous operations. To address the safety concerns, the device is engineered to require operator approval for each task before execution to ensure control and reduce the risk of unplanned events. This helps to control the automated machine and allows the operator to make sound decisions during the bolting process. This automated roof bolter machine utilizes a programmable MD4 touchscreen display, developed by Parker Hannifin, as its HMI device. The HMI interface includes all functional buttons required to perform the roof bolting cycle in automatic and manual modes. Additionally, the HMI features buttons linked to each component of the roof bolting machine. Parker [85] describes the MD4 touchscreen display as a rugged mechanical device with no moving parts which is completely sealed. The screen is made with bonded glass, improving readability, minimizing light refraction, and preventing condensation by eliminating the air gap between the glass and the liquid crystal display (LCD). The MD4 HMI system provides a comprehensive set of visualization software and hardware solutions, including features like graphics, alarm systems, and networking modules.
The HMI is linked to the expansion module mounted in the PLC, which links all components through the iQAN system. The iQAN system has software that integrates all of the components using the controller area network (CAN) signals. A CAN bus is installed to support communication through the CAN protocol. The HMI’s configuration and design are performed using the iQAN software, allowing users to configure the interface by setting the preferences of the measuring system required for digital input and output (I/O) based on the task. The HMI controls the automated roof bolting machine by transmitting CAN messages to the PLC to initiate task execution. When a function is initiated from the HMI touchscreen, a CAN message is generated from the PLC and sent for immediate execution of the task as demonstrated in Figure 4. The HMI establishes a seamless connection between the operator and the machine, allowing manual control of the system’s operation. The EtherNet/IP protocol is employed for message exchange between components of the automated roof bolter machine. The iQAN interface communicates with the programming panel, ABB, and PLC through the Anybus CAN-to-Ethernet/IP interface. The Anybus communicator is installed to ensure seamless communication by converting commands from the iQAN touchpad into a format that the robot controller can interpret, enabling the robot to execute those commands. As a protocol converter, it facilitates communication between the robot and the hydraulic bolter, which operates on a different protocol.
Traditionally, the roof bolting process is performed manually where humans operate the machine in close proximity to accomplish the bolting process. The designed HMI for this study allows the machine to be operated in both manual and automatic modes. Each of the subroutine tasks of the bolting process is assigned connected to a corresponding button created on the HMI using iQAN software, allowing the same function to be executed when uploaded to the MD4 via the PLC. That is, the HMI contains all functional buttons necessary to complete roof bolting processes in both automatic and manual modes. These buttons include the joystick for controlling the movement of hydraulic parts such as the up and down movement of the drill head, opening and closing of clamps, and rotational movement of the drill head in clockwise and counterclockwise directions. To address a safety concern, the HMI should be designed so that the operator must activate the hydraulic components from the menu page before starting the bolting operations. When the hydraulic components are activated, a red light will continuously flash to alert the operator that the hydraulic roof bolter machine is in operation.
During the design, the manufacturer deploying the autonomous equipment should consider incorporating human-centered analysis within the broader systems engineering process. The manufacturer implementing autonomous systems must simultaneously invest in both personnel and technology. According to the Global Mining Guideline Group [86], equipment manufacturers and mining companies should conduct regular technical trainings, workshops, and seminars, as part of new skill requirements to empower the mine workers. Training the current workforce in fundamental digital skills, such as programming and coding, and advanced cognitive abilities like information processing and creative thinking, can foster operator acceptance of automated technologies. Additionally, it can help alleviate boredom caused by simplifying complex tasks and address behavioral changes in response to different levels of system automation.
Apart from upskilling existing miners, the integration of automated machines in underground mining operations should account for interactions and potentially provide feedback based on the commands issued to direct the machines to perform specific tasks. This is imperative as decisions made regarding the autonomous system and interface complexity may influence personnel behavior, training requirements, and the anticipated number of people required for system operation and maintenance.

5. Contributions and Limitations of Automation in Mineral Extraction Industry

5.1. Contributions of Automation to Mining Operations

The introduction of automation in mining operations offers numerous benefits to the mining sector. In the last decade, the global mineral extractive industry, particularly in the United States, has experienced significant increase in the implementation of automated and digital technology that positive impacts productivity and safety. Since then, there have been ongoing attempts and trials to automate the entire mining operations from the exploration phase to processing. However, the increasing demand for mining products, particularly critical minerals essential for industrial sustainability and economic growth, has led to the escalating application of automated technologies in the mining value chain. The challenging operational condition in underground mines encourages the deployment of smart and digital technologies as solutions to inherent mining issues in the areas of costs, productivity, and safety. Moreover, the introduction of automated technologies in the mining process has potentially reduced health and safety risks associated with mineral extraction by distancing operators from hazardous situations. The following are the noticeable benefits of automation to the global mining industry.

5.1.1. Improved Health and Safety in Mining Operations

The mining industry is considered hazardous due to its safety records. Since the introduction of autonomous technologies into the mining industry, there has been improvement in the safety of mine workers in both open pit and underground mining operations. The hazardous work conditions of mining operations are increasingly deterring mineworkers, as deeper and more remote mineral deposits expose them to numerous occupational dangers. Extraction of these deposits is increasingly challenging due to the presence of toxic gases, particulate matter, high temperatures, inadequate ventilation, and adverse mining conditions. The implementation of smart technologies in mining has significantly reduced the exposure of personnel to occupational hazards. However, the statistical trends and injury records from various mining operations indicated that the safety of miners has tremendously improved as mines are seeing a decrease in mine accident and injury records. For example, automation in mining reduces emissions, exemplified by the 13 km Takraf belt conveyor at Codelco’s Chuquicamata mine in Chile, which eliminates 120 haul trucks and cuts CO2 emissions by 70% [87]. Ultimately, implementing autonomous technologies in the mining value chain enables humans to be moved away from the hazardous areas and allows smart technologies to autonomously perform hazardous and repetitive tasks. This has created changes in roles and workflows as human operators will supervise the automated systems in a safe place as mining progresses.

5.1.2. Reduction in Overall Mining Operating Costs

Like other sectors, the mining industry continually seeks ways to lower expenses and increase profits. Mining operations are capital-intensive due to high initial costs, ongoing operational expenses, and long-term investment horizons. The industry is facing the challenge of being more productive to increase profitability due to rising mining production costs, which threatens the sustainability of mining operations. To maintain their competitive edge in today’s intense market, the leading mining companies are left with no choice but to leverage technological advancements in their operations, especially in the loading and haulage operations, known as the most extravagant part of a surface mining project [88]. A significant section in mining projects to reduce operational costs is fleet management. The development of autonomous mining equipment loading, hauling, and dumping vehicles has substantially reduced labor costs and improved productivity, making the mining industry more competitive in the long term. Generally, human operators drive large haul trucks to move material from the pit to be unloaded for further processing. In today’s operation, mining companies find autonomous trucks valuable for hauling applications. The benefits of autonomous haulage trucks in mining include reductions in fuel consumption and maintenance costs, and increased lifespans of the tires. An example of this is the Roy Hill iron mining site which owns the world’s single largest AHS [89]. According to Motion Metrics [90], the autonomous fleet outperforms the manned fleet by an average of 14% and has 13% lower operating costs.

5.1.3. Increased Productivity

The fundamental goal of implementing autonomous technology in mining operations is to remove personnel from exposure to occupational risks and hazards. The integration has not only improved the safety of miners but also increased operational efficiency and productivity. Mining companies are enjoying the benefits of implementing autonomous technologies in their operations as there have been significant improvements in overall mine productivity. The mining industry is gradually undergoing significant changes as their operations’ autonomous machines can enable mines to operate for longer hours, at the highest level of productivity [91]. Integrating automated technologies in the mining value chain allows for operation in extreme conditions, consequently reducing idle time and human errors due to fatigue. These are the core of innovative solutions that drive productivity in mining operations.

5.1.4. Real-Time Data Acquisition for Sound Decision Making

Automation enables the availability of real-time data for sound decision-making in mining operations. Mining companies generate volumes of data from various equipment. Unfortunately, little of this data is used in decision-making. Due to advancements in current technologies, machine learning technologies, data analytic computer codes, and mining operations are now leveraging these tools to make real-time and sound decisions that enhance safety and improve productivity. These systems can generate important data to predict and give insight into future events. The data would allow the miners to be aware of their surroundings and anticipate potential risks before they occur. For instance, the application of computer codes and simulation modeling approaches in mining operations can predict the stability conditions of an excavation based on input parameters. Additionally, advanced analytics can analyze data trends, identify opportunities, and pinpoint operational bottlenecks that may lead to outages or the closure of mining operations.

5.1.5. Remote Monitoring and Control of Equipment

One of the major benefits of automation is the control of mining machines remotely, especially when working in complex and challenging mining conditions. Also, gaining access to deposits is becoming difficult because of poisonous gases, dust, high temperatures, poor ventilation, and complex and challenging mining conditions. For instance, elevated temperatures in underground mines will not only affect miners’ physical and mental health by reducing their work efficiency, causing accidents, but also trigger the thermodynamic effect of rock masses and produce unexpected mine disasters [92]. Currently, many mines face challenges accessing high-quality ore deposits due to safety issues, costs, and risks associated with developing new mines. For example, high-temperature hazards are prevalent in western South African mines, where temperatures can reach up to 50 °C at a depth of 3300 m [26,93]. Additionally, high water pressure hazards, such as water inrush incidents, have been observed in deep mines like the Shaling Gold Mine in Laizhou, China, where water pressure can reach 11 MPa at a depth of 1425 m [92]. The capability of automated technologies to remotely monitor and control mine equipment promotes cost reduction, increased efficiency, and access to new and profitable reserves [94]. This means there is a need to implement autonomous equipment that can be remotely controlled and perform tasks with improved human-like accuracy.

5.2. Challenges and Limitations of Automation in the Mining Industry

In the last few decades, the mining industry has witnessed a surge in the deployment of new automated and digital systems such as autonomous vehicles, installation of positioning and navigation systems, AI, computer codes and simulation modeling, integration of industrial robot and cobot technologies, automated drills, and increases in smart sensors. The introduction of automation has significantly influenced the critical areas of the mining industry: productivity, cost savings, and health and safety. Despite the numerous advantages of automated technologies in the mineral extractive industry, some challenges affect the acceptability and implementation of autonomous systems in mining operations. A few of these challenges are discussed below.

5.2.1. Technological and Technical Challenges

Technical issues, such as system or machine failures and faults, can make mining companies skeptical about introducing automated technologies in their operations. One of the notable limitations of automation in mining industries is the failure of a system to respond to commands which may be attributed to faulty components. For instance, in robotics, the sources of errors can be ascribed to robot mechanical issues such as faulty electronics or sensors, failed controllers due to programming bugs, and loose parts. These errors can lead to the robot working at an uncontrollable speed or the robot failing to stop. This may jeopardize the safety of personnel operating in proximity to the system. Furthermore, another notable drawback in deploying autonomous equipment in underground mining operations is the unavailability of satellites in subsurface mines making it difficult to accurately position and navigate the equipment. However, there has been some research and attempts to construct a system that can assist in tracking and locating mine machinery in underground mines. Ideally, fully autonomous equipment should be able to present information regarding its immediate environment. However, due to technological advancements, alternative smart sensors are installed to provide accurate maps of the roads. An example of this is the SLAM-based LiDAR 3D scanner. The device is a suitable replacement where GNSS signals are not available. It enables the georeferencing of each measurement or point within the generated point clouds, combining these points to form a 3D representation of the target object. A LiDAR-based SLAM system creates a 3D map of its surroundings by using lasers to measure the distance between the scanner and the target object. One drawback of using this equipment is the poor visibility in underground mining environments, caused by dust and inadequately lit roadways. Other technological obstacles hindering the application of autonomous equipment include a lack of skilled workers to operate the autonomous technologies, undulating or uneven mine roadways, lack of adequate training and skills, and poorly designed human–machine interfaces that can sabotage the system.

5.2.2. Challenging Geology and Harsh Working Conditions

According to Günther et al. [95], the working conditions in underground mines are heavily influenced by factors such as the mine’s depth, the type of mineral being extracted, the surrounding geological and rock formations, the mining technologies employed, the current operational status of the mine, and various other site-specific parameters unique to each mine. These challenges not only pose risks to operators but can also be harmful to digital systems. Specifically in underground mining environments, high temperatures and humidity can affect the durability of the equipment. In addition to high temperatures underground, the presence of water, whether dripping or stagnant, combined with acidic mine water can cause corrosion or electrical short circuits in mining machines. Also, multiple levels and intricate networks within the mine may require robust communication systems and sophisticated algorithms to maintain continuous operation and coordination of the autonomous equipment. A complex and irregular mine layout may require more advanced navigation and mapping technologies to ensure the autonomous equipment can operate effectively. Uneven and undulating roadways in underground mines can make it difficult to navigate autonomous equipment. Notably, the lack of natural lighting results in absolute darkness without artificial illumination in underground mines. This presents challenges, particularly for detection algorithms that depend on optical properties and must operate effectively under highly variable lighting conditions. The effect of complexities in orebody formation can cause a continuous change in the mine planning, layout, and design. The design of a mine is a significant factor that determines the mine equipment selection which can influence the overhead cost of the mining project. However, an irregular mine layout could impact operating costs as the existing autonomous system may not be suitable for new mining methods.

5.2.3. Economical and Cyclical Mining Commodity Price

Fluctuating mining commodity prices significantly impact the economic viability of investing in mining operations. Prices for critical minerals can be volatile, as shown in Figure 5. The mining market is experiencing cyclical commodity pricing which can impact the deployment or investing in automated systems. When prices are low, mining companies may be hesitant to make large capital investments. The commodity price is often regarded as the primary factor determining a mine’s profitability, as it forms the basis for the revenue generated from each unit of the produced commodity. This volatility hurts the ability to plan and implement long-run capital investments in new technologies. This is one of the reasons why mining companies are skeptical about deploying autonomous systems in the mining value chain as long-term return on investment (ROI) is doubtful.
There is a need to carefully evaluate the existing investment and anticipated gains to justify investing in new technologies such as the cost of purchasing autonomous systems, sensors, software, and other related infrastructure is excessively high. However, considering the nature of the mining industry, the industry is conservative as existing mining operations may need substantial modifications or replacements to be compatible with new autonomous systems. The fear of being replaced by automated systems contributes to sabotaging the success of the implementation. However, several researchers have predicted that the deployment of autonomous equipment has the potential to result in workforce redundancies and job displacement. In the opinion of these authors, integrating smart technologies would create new, high-paying jobs that specialize in a new set of skills. That is, it may lead to a shift in skills and knowledge, such as computer coding, data analytics, critical thinking skills, IT, and system engineering skills.

5.2.4. Regulatory and Legal Issues

In the mining industry, there is no generally established legal framework or guideline governing the behavior of autonomous systems and the deployment of autonomous technologies in mining operations. Proper legal documents should be created to highlight the various conditions required for deploying autonomous systems, addressing compliance with safety, liability issues in the event of accidents, adherence to environmental protection standards, and labor laws related to worker displacement and training. This would protect the interest of the mining company in the event that the autonomous device acts contrary to the law as a result of poor development or manufacturer defects. If the issue of responsibility and the appropriate penalties for technology failures can be addressed, the performance of autonomous systems will improve. In most cases involving robotics, responsibility is attributed to the designers or manufacturers, as autonomous systems cannot be held accountable for their actions, including failures or accidents. That is, the manufacturer of the technology should be responsible for ensuring safety throughout the entire intended lifecycle of the product, including gathering and evaluating operational data to reveal potential risks.
Additionally, privacy remains a critical issue that requires attention. Integrating automated systems results in a significant increase in the deployment of sensors, generating large volumes of data. To comply with legal regulations, it is essential to collect and process data according to guidelines that specify the scope of personal data collection and the appropriate anonymity measures in specific instances. When clear standards and protocols are established, data can be exchanged between systems developed by different manufacturers. Marshall et al. [97] noted that legal concerns and insurance-related challenges could impede the adoption and implementation of autonomous equipment in the mining industry. Conducting a comprehensive risk analysis for autonomous machines presents significant challenges. Legal issues in mining may involve intellectual property rights related to technology, data privacy, and environmental considerations. Thus, there is a need to establish regulations that govern the relationship between manufacturers and mining companies, thereby enhancing contract law with a regulatory framework.

6. Conclusions

Global industry is quickly advancing toward adopting autonomous systems, as operations in engineering and industrial fields increasingly shift toward intelligent, unmanned methods to carry out hazardous tasks. Like other industries, the increasing use of automation technologies is slowly transforming the mineral extractive industry. The mining industry is gradually transforming, with operations becoming increasingly autonomous through integrating digital technologies aimed at enhancing miner safety and to boost productivity. The increasing demand for mining products driven by the increase in the manufacturing sector motivates the mining industry to adopt technological innovations that save costs, increase productivity, and improve worker safety. The advancement of technologies and innovations has led to the development of autonomous systems. The fundamental goal of deploying automated technologies in the mining value chain is to minimize human intervention during the mining process. The need for safety improvements and an increases in mining productivity encourages the implementation of these smart technologies and digital systems to enhance the safety of miners. Apart from sensor integration, recent technological advances in AI, data analytics, and other soft computing technologies have become indispensable tools that are helping the mining industry predict and address the potential challenges of mining operations. However, AI, data investigation, and visualization methods can potentially predict and consequently prevent potentially dangerous situations or eliminate the need for human work within hazardous environments, such as hauling, loading, triggering explosives, installing roof supports, or removing toxic gases and dust. Despite the positive impacts of integrated autonomous systems, mining companies remain cautious about deploying these smart technologies due to several challenges. Concerns remain about the quality control of the equipment and the interaction between humans and the equipment in the timely management of wider societal implications. For a system to function safely, the objective must consider human abilities and limitations. Like in other industries, it has always been argued that designers of automated systems must consider the human factor by designing and implementing user-friendly systems. One of the primary factors preventing the complete removal of humans from these systems is the belief that humans possess greater flexibility, adaptability, and creativity than automation, making them better equipped to respond to changing or unforeseen conditions. While manufacturers of automated systems cannot predict all potential scenarios in a complex environment, the prevailing approach is to rely on human operators to apply their experience and judgment when utilizing automation.
This study highlights the transformative potential of autonomous systems in the mining industry, focusing on improving safety, productivity, and cost efficiency. Integrating automation into mining operations should not aim to eliminate human labor but rather to complement it by reducing workloads and improving safety and efficiency. The sustained success of autonomous systems will depend on human involvement, thoughtful design, and iterative implementation that balances technology with the unique strengths of human operators. This balance ensures that automation is an enabler rather than a substitute, fostering broader acceptance and long-term sustainability in mining practices. The findings emphasize that while automation reduces human exposure to hazardous environments, human operators remain crucial due to their adaptability, creativity, and ability to manage unforeseen situations. The key findings of the study are summarized as follows:
(1)
Human–machine interaction (HMI): developing effective HMIs is critical to fostering trust and ensuring seamless communication between operators and autonomous systems. Alarms and alert mechanisms must be intuitive, enabling swift and efficient responses during system failures or emergencies.
(2)
Human factors in automation: successful deployment of autonomous systems requires designs that account for human capabilities and limitations, including the ability to override and manage automated systems when necessary.
(3)
Training and trust: operators must be adequately trained to interact with and adapt to automated technologies, ensuring human oversight and operational efficiency.

Author Contributions

Conceptualization and manuscript writing, P.K.; manuscript planning and writing, N.O.O.; manuscript editing, K.K.; manuscript editing, D.D.O. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

Author Daniel Dare Omole was employed by the company Midgate Global Resources. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Schafrik, S.; Kolapo, P.; Agioutantis, Z. Development of an automated roof bolting machines for underground coal mines. In Proceedings of the 32nd SOMP Annual Meeting and Conference, Windhoek, Namibia, 12–14 September 2022; The Southern Africa Institute of Mining and Metallurgy: Johannesburg, South Africa, 2022. [Google Scholar]
  2. Ralston, J.; Reid, D.; Dunn, M.; Hainsworth, D. Logwall automation: Delivering enabling technology to achieve safer and more productive underground mining. Int. J. Min. Sci. Technol. 2015, 25, 865–876. [Google Scholar] [CrossRef]
  3. Javaid, M.; Haleem, A.; Singh, R.P.; Rab, S.; Suman, R. Significant applications of Cobots in the field of manufacturing. Cogn. Robot. 2022, 2, 222–233. [Google Scholar] [CrossRef]
  4. Paredes, D.; Fleming-Muñoz, D. Automation and robotics in mining: Jobs, income and inequality implications. Extr. Ind. Soc. 2021, 8, 189–193. [Google Scholar] [CrossRef]
  5. Lopes, L.; Miklovicz, T.; Bakker, E.; Milosevic, Z. The benefits and challenges of robotics in the mineral raw materials sector—An overview. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, 1–5 October 2018. [Google Scholar] [CrossRef]
  6. Sam-Aggrey, H. Assessment of the Impacts of New Mining Technologies: Recommendations on the way Forward. WIT Trans. Ecol. Environ. 2020, 245, 177–187. [Google Scholar]
  7. Rowduru, S.; Kumar, N.; Kumar, A. A critical review on automation of steering mechanism of load haul dump machine. Proc. Inst. Mech. Eng. Part I J. Syst. Control. Eng. 2019, 234, 160–182. [Google Scholar] [CrossRef]
  8. Zimroz, R.; Hutter, M.; Mistry, M.; Stefaniak, P.; Walas, K.; Wodecki, J. Why Should Inspection Robots be used in Deep Underground Mines? In Proceedings of the 27th International Symposium on Mine Planning and Equipment Selection—MPES; Springer: Berlin/Heidelberg, Germany, 2019. [Google Scholar]
  9. Larsson, J.; Broxvall, M.; Saffiotti, A. A Navigation System for Automated Loaders in Underground Mines. In Field and Service Robotics; Corke, P., Sukkariah, S., Eds.; Springer: Berlin/Heidelberg, Germany, 2006; pp. 129–140. [Google Scholar]
  10. Larsson, J.; Broxvall, M.; Saffiotti, A. Laser Based Intersection Detection for Reactive Navigation in an Underground Mine. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Nice, France, 22–26 September 2008; pp. 2222–2227. [Google Scholar]
  11. Gu, Q.; Li, L.; Bai, G.; Li, K.; Meng, Y.; Ma, F. Longitudinal and Lateral Trajectory Planning for the Typical Duty Cycle of Autonomous Load Haul Dump. IEEE Access 2019, 7, 126679–126695. [Google Scholar] [CrossRef]
  12. Rocha, F.; Garcia, G.; Pereira, R.; Faria, H.; Silva, T.; Andrade, R.; Barbosa, E.; Almeida, A.; Cruz, E.; Andrade, W.; et al. ROSI: A Robotic System for Harsh Outdoor Industrial Inspection—System Design and Applications. J. Intell. Robot. Syst. 2021, 103, 1–22. [Google Scholar] [CrossRef]
  13. Szrek, J.; Wodecki, J.; Błazej, R.; Zimroz, R. An Inspection Robot for Belt Conveyor Maintenance in Underground Mine-Infrared Thermography for Overheated Idlers Detection. Appl. Sci. 2020, 10, 4984. [Google Scholar] [CrossRef]
  14. Gaber, T.; El Jazouli, Y.; Eldesouky, E.; Ali, A. Autonomous Haulage Systems in the Mining Industry: Cybersecurity, Communication and Safety Issues and Challenges. Electronics 2021, 10, 1357. [Google Scholar] [CrossRef]
  15. Chaowasakoo, P.; Leelasukseree, C.; Wongsurawat, W. Introducing GPS in fleet management of a mine: Impact on hauling cycle time and hauling capacity. Int. J. Technol. Intell. Plan. 2015, 10, 49–66. [Google Scholar] [CrossRef]
  16. Thrun, S.; Thayer, S.; Whittaker, W.; Baker, C.; Burgard, W.; Ferguson, D.; Hahnel, D.; Montemerlo, D.; Morris, A.; Omohundro, Z.; et al. Autonomous exploration and mapping of abandoned mines. IEEE Robot. Autom. Mag. 2004, 11, 79–91. [Google Scholar] [CrossRef]
  17. Nüchter, A. 3D Robotic Mapping: The Simultaneous Localization and Mapping Problem with Six Degrees of Freedom; Springer: Berlin/Heidelberg, Germany, 2009. [Google Scholar]
  18. Lösch, R.; Grehl, S.; Donner, M.; Buhl, C.; Jung, B. Design of an Autonomous Robot for Mapping, Navigation, and Manipulation in Underground Mines. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, 1–5 October 2018. [Google Scholar]
  19. Evans, S. IOT World Today. 2023. Available online: https://www.iotworldtoday.com/robotics/mining-robot-designed-to-automate-blasting-process#close-modal (accessed on 10 December 2024).
  20. Bonchis, A.; Duff, E.; Roberts, J.; Bosse, M. Robotic Explosive Charging in Mining and Construction Applications. IEEE Trans. Autom. Sci. Eng. 2014, 11, 245–250. [Google Scholar] [CrossRef]
  21. Peng, S.; Du, F.; Cheng, J.; Li, Y. Longwall mining automation horizon control: Coal seam gradient identification using piecewise linear fitting. Int. J. Min. Sci. Technol. 2019, 29, 151–159. [Google Scholar] [CrossRef]
  22. Billingsley, J.; Brett, P. Machine Vision and Mechatronics in Practice; Springer: Berlin/Heidelberg, Germany, 2015. [Google Scholar]
  23. Liebherr. LIPOS—Liebherr Positioning System. 2023. Available online: https://www.liebherr.com/en/usa/products/construction-machines/deep-foundation/digital-solutions/lipos/lipos.html (accessed on 12 October 2023).
  24. Li, J.; Zhan, K. Intelligent Mining Technology for an Underground Metal Mine Based on Unmanned Equipment. Engineering 2018, 4, 381–391. [Google Scholar] [CrossRef]
  25. Yang, X.; Wang, R.; Wang, H.; Yang, Y. A novel method for measuring pose of hydraulic supports relative to inspection robot using LiDAR. Measurement 2020, 154, 107452. [Google Scholar] [CrossRef]
  26. He, M.C.; Xu, M. Research and development of HEMS cooling system and heat-harm control in deep mine. Chin. J. Rock Mech. Eng. 2008, 27, 1353–1361. (In Chinese) [Google Scholar]
  27. Jeon, P.R.; Moon, J.H.; Ogunsola, N.O.; Lee, S.H.; Ling JL, J.; You, S.; Park, Y.K. Recent advances and future prospects of thermochemical biofuel conversion processes with machine learning. Chem. Eng. J. 2023, 471, 144503. [Google Scholar] [CrossRef]
  28. Ogunsola, N.O.; Oh, S.S.; Jeon, P.R.; Ling, J.L.J.; Park, H.J.; Park, H.S.; Lee, S.H. Progresses and Challenges of Machine Learning Approaches in Thermochemical Processes for Bioenergy: A Review. Korean J. Chem. Eng. 2024, 41, 1923–1953. [Google Scholar] [CrossRef]
  29. Lee, S.; Oh, H.J. Application of artificial neural network for mineral potential mapping. In Artificial Neural Networks—Application; Artificial Neural Networks-Applications: Shanghai, China, 2011; pp. 67–104. [Google Scholar]
  30. Zhang, N.; Zhou, K. Mineral prospectivity mapping with weights of evidence and fuzzy logic methods. J. Intell. Fuzzy Syst. 2015, 29, 2639–2651. [Google Scholar] [CrossRef]
  31. Kashani, S.; Abedi, M.; Norouzi, G. Fuzzy logic mineral potential mapping for copper exploration using multi-disciplinary geo-datasets, a case study in seridune deposit, Iran. Earth Sci. Inform. 2016, 9, 167–181. [Google Scholar] [CrossRef]
  32. Tabaei, M.; Esfahani, M.; Rasekh, P.; Esna-ashari, A. Mineral prospectivity mapping in GIS using fuzzy logic integration in Khondab area, western Markazi province, Iran. Q. J. Tethys 2017, 5, 367–379. [Google Scholar]
  33. Tessema, A. Mineral Systems Analysis and Artificial Neural Network Modeling of Chromite Prospectivity in the Western Limb of the Bushveld Complex, South Africa. Nat. Resour. Res. 2017, 26, 465–488. [Google Scholar] [CrossRef]
  34. Sun, T.; Chen, F.; Zhong, L.; Liu, W.; Wang, Y. GIS-based mineral prospectivity mapping using machine learning methods: A case study from Tongling ore district, eastern China. Ore Geol. Rev. 2019, 109, 26–49. [Google Scholar] [CrossRef]
  35. Yeomans, C.M.; Shail, R.K.; Grebby, S.; Nykänen, V.; Middleton, M.; Lusty, P.A. A machine learning approach to tungsten prospectivity modelling using knowledge-driven feature extraction and model confidence. Geosci. Front. 2020, 11, 2067–2081. [Google Scholar] [CrossRef]
  36. Farahbakhsh, E.; Maughan, J.; Müller, R.D. Prospectivity modelling of critical mineral deposits using a generative adversarial network with oversampling and positive-unlabelled bagging. Ore Geol. Rev. 2023, 162, 105665. [Google Scholar] [CrossRef]
  37. Zhao, H.; Zhang, Y.; Xu, Y.; Shao, Y.; Chen, X.; Hao, J.; Wang, X. Machine learning model for deep exploration: Utilizing short wavelength infrared (SWIR) of hydrothermal alteration minerals in the Qianchen gold deposit, Jiaodong Peninsula, Eastern China. Ore Geol. Rev. 2024, 168, 106060. [Google Scholar] [CrossRef]
  38. Mining. Goldcorp Partners with IBM to Hunt for Exploration Targets at Red Lake. 2017. Available online: http://www.mining.com/goldcorp-partners-ibm-hunt-exploration-targets-red-lake/ (accessed on 1 November 2024).
  39. Wu, X.; Zhou, Y. Reserve estimation using neural network techniques. Comput. Geosci. 1993, 19, 567–575. [Google Scholar] [CrossRef]
  40. Al-Alawi, S.; Tawo, E. Application of Artificial Neural Networks in Mineral Resource Evaluation. J. King Saud Univ. Eng. Sci. 1998, 10, 127–138. [Google Scholar] [CrossRef]
  41. Jalloh, A.B.; Kyuro, S.; Jalloh, Y.; Barrie, A.K. Integrating artificial neural networks and geostatistics for optimum 3D geological block modeling in mineral reserve estimation: A case study. Int. J. Min. Sci. Technol. 2016, 26, 581–585. [Google Scholar] [CrossRef]
  42. Afeni, T.B.; Lawal, A.I.; Adeyemi, R.A. Re-examination of Itakpe iron ore deposit for reserve estimation using geostatistics and artificial neural network techniques. Arab. J. Geosci. 2020, 13, 657. [Google Scholar] [CrossRef]
  43. Kaplan, U.E.; Topal, E. A new ore grade estimation using combine machine learning algorithms. Minerals 2020, 10, 847. [Google Scholar] [CrossRef]
  44. Yasrebi, A.B.; Hezarkhani, A.; Afzal, P.; Karami, R.; Tehrani, M.E.; Borumandnia, A. Application of an ordinary kriging–artificial neural network for elemental distribution in Kahang porphyry deposit, Central Iran. Arab. J. Geosci. 2020, 13, 748. [Google Scholar] [CrossRef]
  45. Zhang, S.E.; Nwaila, G.T.; Tolmay, L.; Frimmel, H.E.; Bourdeau, J.E. Integration of machine learning algorithms with Gompertz Curves and Kriging to estimate resources in gold deposits. Nat. Resour. Res. 2021, 30, 39–56. [Google Scholar] [CrossRef]
  46. Atalay, F. Estimation of Fe Grade at an Ore Deposit Using Extreme Gradient Boosting Trees (XGBoost). In Mining, Metallurgy & Exploration; Springer Nature: Berlin/Heidelberg, Germany, 2024; pp. 1–10. [Google Scholar]
  47. Ajak, A.D.; Lilford, E.; Topal, E. Application of predictive data mining to create mine plan flexibility in the face of geological uncertainty. Resour. Policy 2018, 55, 62–79. [Google Scholar] [CrossRef]
  48. Nourali, H.; Osanloo, M. A regression-tree-based model for mining capital cost estimation. Int. J. Min. Reclam. Environ. 2020, 34, 88–100. [Google Scholar] [CrossRef]
  49. Zhang, H.; Nguyen, H.; Bui, X.N.; Nguyen-Thoi, T.; Bui, T.T.; Nguyen, N.; Vu, D.A.; Mahesh, V.; Moayedi, H. Developing a novel artificial intelligence model to estimate the capital cost of mining projects using deep neural network-based ant colony optimization algorithm. Resour. Policy 2020, 66, 101604. [Google Scholar] [CrossRef]
  50. Noriega, R.; Pourrahimian, Y. A systematic review of artificial intelligence and data-driven approaches in strategic open-pit mine planning. Resour. Policy 2022, 77, 102727. [Google Scholar] [CrossRef]
  51. Zheng, X.; Nguyen, H.; Bui, X.N. Exploring the relation between production factors, ore grades, and life of mine for forecasting mining capital cost through a novel cascade forward neural network-based salp swarm optimization model. Resour. Policy 2021, 71, 102300. [Google Scholar] [CrossRef]
  52. Nobahar, P.; Pourrahimian, Y.; Mollaei Koshki, F. Optimum fleet selection using machine learning algorithms-Case study: Zenouz Kaolin mine. Mining 2022, 2, 528–541. [Google Scholar] [CrossRef]
  53. Ogunsola, N.O.; Lawal, A.I.; Saliu, M.A. Variations of physico-mechanical, mineralogical, and geochemical properties of marble under the influence of weathering. Mining of mineral deposits. Min. Miner. Depos. 2019, 13, 95–102. [Google Scholar] [CrossRef]
  54. Lawal, A.I.; Oniyide, G.O.; Kwon, S.; Onifade, M.; Köken, E.; Ogunsola, N.O. Prediction of mechanical properties of coal from non-destructive properties: A comparative application of MARS, ANN, and GA. Nat. Resour. Res. 2021, 30, 4547–4563. [Google Scholar] [CrossRef]
  55. Ogunsola, N.O.; Lawal, A.I.; Kim, G.; Kim, H.; Cho, S. Prediction of Mode-I Fracture Toughness of the ISRM-Suggested Semi-Circular Bend Rock Specimen Using ANN and Optimized ANN Models. In Rock Mechanics and Rock Engineering; Springer Nature: Berlin/Heidelberg, Germany, 2024; pp. 1–26. [Google Scholar]
  56. Lawal, A.I.; Hosseini, S.; Kim, M.; Ogunsola, N.O.; Kwon, S. Prediction of factor of safety of slopes using stochastically modified ANN and classical methods: A rigorous statistical model selection approach. Nat. Hazards 2024, 120, 2035–2056. [Google Scholar] [CrossRef]
  57. Armaghani, D.J.; Mamou, A.; Maraveas, C.; Roussis, P.C.; Siorikis, V.G.; Skentou, A.D.; Asteris, P.G. Predicting the unconfined compressive strength of granite using only two non-destructive test indexes. Geomech. Eng. 2021, 25, 317–330. [Google Scholar]
  58. Skentou, A.D.; Bardhan, A.; Mamou, A.; Lemonis, M.E.; Kumar, G.; Samui, P.; Asteris, P.G. Closed-form equation for estimating unconfined compressive strength of granite from three non-destructive tests using soft computing models. Rock Mech. Eng. 2023, 56, 487–514. [Google Scholar] [CrossRef]
  59. Zhao, J.; Li, D.; Jiang, J.; Luo, P. Uniaxial Compressive Strength Prediction for Rock Material in Deep Mine Using Boosting-Based Machine Learning Methods and Optimization Algorithms. CMES-Comput. Model. Eng. Sci. 2024, 140, 275–304. [Google Scholar] [CrossRef]
  60. Mahmoodzadeh, A.; Mohammadi, M.; Ghafoor Salim, S.; Farid Hama Ali, H.; Hashim Ibrahim, H.; Nariman Abdulhamid, S.; Rashidi, S. Machine learning techniques to predict rock strength parameters. Rock Mech. Rock Eng. 2022, 55, 1721. [Google Scholar] [CrossRef]
  61. Frank, A.; Kolapo, P.; Ogunsola, N.; Munemo, P.; Akinola, A. Application of Improved Blasting Techniques in Open Pit Mining for Maximum Productivity: A Case of Oakyam Quarry Limited, Ogun State, Nigeria. Science 2022, 10, 12–23. [Google Scholar]
  62. Ogunsola, N.O.; Shin, C.; Kim, Y.K.; Lawal, A.I.; Cho, S. Development of an ANN-based closed-form equation for the prediction of airblast overpressure induced by construction rock excavation blasting in urban areas. Appl. Sci. 2022, 12, 12654. [Google Scholar] [CrossRef]
  63. Ogunsola, N.; Pureun, J.; Cho, S.; Kim, Y. An artificial neural network-based model for blast-induced ground vibration prediction in urban area. In Rock Dynamics: Prospect and Report; CRC Press: Boca Raton, FL, USA, 2023; pp. 229–234. [Google Scholar]
  64. Komadja, G.C.; Rana, A.; Glodji, L.A.; Anye, V.; Jadaun, G.; Onwualu, P.A.; Sawmliana, C. Assessing ground vibration caused by rock blasting in surface mines using machine-learning approaches: A comparison of CART, SVR, and MARS. Sustainability 2022, 14, 11060. [Google Scholar] [CrossRef]
  65. Abd Elwahab, A.; Topal, E.; Jang, H.D. Review of machine learning application in mine blasting. Arab. J. Geosci. 2023, 16, 133. [Google Scholar] [CrossRef]
  66. Dumakor-Dupey, N.K.; Arya, S.; Jha, A. Advances in blast-induced impact prediction. Minerals 2021, 11, 601. [Google Scholar] [CrossRef]
  67. Amiri, M.; Bakhshandeh Amnieh, H.; Hasanipanah, M.; Mohammad Khanli, L. A new combination of artificial neural network and K-nearest neighbors models to predict blast-induced ground vibration and air-overpressure. Eng. Comput. 2016, 32, 631–644. [Google Scholar] [CrossRef]
  68. Amoako, R.; Jha, A.; Zhong, S. Rock fragmentation prediction using an artificial neural network and support vector regression hybrid approach. Mining 2022, 2, 233–247. [Google Scholar] [CrossRef]
  69. Samadi, M.; Bakhtavar, E.; Hu, G.; Hewage, K.; Sadiq, R. Assessment of mine post-closure strategies by integrating evidential reasoning and fuzzy cognitive maps: Toward sustainable mining policy. Resour. Policy 2023, 83, 103719. [Google Scholar] [CrossRef]
  70. Maxwell, A.E.; Warner, T.A.; Strager MP CJ, F.; Sharp, A.L. Assessing machine-learning algorithms and image-and lidar-derived variables for GEOBIA classification of mining and mine reclamation. Int. J. Remote Sens. 2015, 36, 954–978. [Google Scholar] [CrossRef]
  71. Li, X.; Chen, W.; Cheng, X.; Wang, L. A Comparison of Machine Learning Algorithms for Mapping of Complex Surface-Mined and Agricultural Landscapes Using ZiYuan-3 Stereo Satellite Imagery. Remote Sens. 2016, 8, 514. [Google Scholar] [CrossRef]
  72. Bui, X.N.; Nguyen, H.; Choi, Y.; Nguyen-Thoi, T.; Zhou, J.; Dou, J. Prediction of slope failure in open-pit mines using a novel hybrid artificial intelligence model based on decision tree and evolution algorithm. Sci. Rep. 2020, 10, 9939. [Google Scholar] [CrossRef]
  73. McCoy, J.T.; Auret, L. Machine learning applications in minerals processing: A review. Miner. Eng. 2019, 132, 95–109. [Google Scholar] [CrossRef]
  74. Zarie, M.; Jahedsaravani, A.; Massinaei, M. Flotation froth image classification using convolutional neural networks. Minerals Engineering 2020, 155, 106443. [Google Scholar] [CrossRef]
  75. Gomez-Flores, A.; Ilyas, S.; Heyes, G.W.; Kim, H. A critical review of artificial intelligence in mineral concentration. Miner. Eng. 2022, 189, 107884. [Google Scholar] [CrossRef]
  76. Bendaouia, A.; Qassimi, S.; Boussetta, A.; Benzakour, I.; Benhayoun, A.; Amar, O.; Hasidi, O. Hybrid features extraction for the online mineral grades determination in the flotation froth using Deep Learning. Eng. Appl. Artif. Intell. 2024, 129, 107680. [Google Scholar] [CrossRef]
  77. Hosseini, S.H.; Samanipour, M. Prediction of final concentrate grade using artificial neural networks from Gol-E-Gohar iron ore plant. Am. J. Min. Metall. 2015, 3, 58–62. [Google Scholar]
  78. Jahedsaravani, A.; Marhaban, M.H.; Massinaei, M. Application of statistical and intelligent techniques for modeling metallurgical performance of a batch flotation process. Chem. Eng. Commun. 2016, 203, 151–160. [Google Scholar] [CrossRef]
  79. Kolapo, P. The Methodology for Integrating Robotic Systems in Underground Mining Machines. Ph.D. Thesis, University of Kentucky, Lexington, KY, USA, 2023. [Google Scholar]
  80. Kolapo, P.; Onifade, M.; Akinseye, O.; Amwaama, M.; Khadija, O.; Abdulsalam, J.; Ismail, L.; Aladejare, E. An Overview of the Use of Rockbolts as Support Tools in Mining Operations. Geotech. Geol. Eng. 2021, 40, 1637–1661. [Google Scholar]
  81. Zhang, H.; Chan, S.-H.; Zhong, J.; Li, J.; Kolapo, P.; Koenig, S.; Agioutantis, Z.; Schafrik, S.; Nikolaidis, S. Multi-robot geometric task-and-motion planning for collaborative manipulation tasks. Auton. Robot. 2023, 47, 1537–1558. [Google Scholar] [CrossRef]
  82. Kolapo, P.; Schafrik, S.; Zhang, H.; Nikolaidiz, S.; Agioutantis, Z. A concept for integrating robotic systems in an underground roof support. J. Ind. Saf. 2024, 1, 100002. [Google Scholar]
  83. Gruhn, P. Human Machine Interface (HMI) Design: The Good, The Bad, and The Ugly (and What Makes Them so). In Proceedings of the 66th Annual Instrumentation Symposium for the Process Industries, College Station, TX, USA, 25–27 January 2011. [Google Scholar]
  84. Rockwell Automation. Rockwell Automation Process HMI Style Guide; Rockwell Automation: Milwaukee, WI, USA, 2019. [Google Scholar]
  85. Parker. IQAN-MD4 Display Modules; Parker Hannifin Corporation: Mayfield Heights, OH, USA, 2019. [Google Scholar]
  86. Global Mining Guidelines Group. System Safety for Autonomous Mining; Canadian Institute of Mining, Metallurgy and Petroleum: Montreal, QC, Canada, 2021. [Google Scholar]
  87. Heidi Vella. Exploring the Environmental Benefits of Autonomous Underground Mining Equipment. 2021. Available online: https://www.mining-technology.com/features/environmental-benefits-of-autonomous-underground-mining-equipment (accessed on 19 December 2024).
  88. Hazrathosseini, A.; Afrapoli, A. Maximizing Mining Operations: Unlocking the Crucial Role of Intelligent Fleet Management Systems in Surface Mining’s Value chain. Mining 2024, 4, 7–20. [Google Scholar] [CrossRef]
  89. Moore, P. International Mining. 2023. Available online: https://im-mining.com/2023/01/12/roll-out-of-autonomous-haulage-system-at-roy-hill-creates-worlds-largest-single-autonomous-mine (accessed on 30 August 2024).
  90. Motion Metrics. Weir Motion Metrics. 2019. Available online: https://www.motionmetrics.com/nulla-aliquet-auctor-leo-sit (accessed on 21 April 2023).
  91. World Economic Forum (WEF). Digital Transformation Initiative: Maximizing the Return on Digital Investments; (White Paper); World Economic Forum: Geneva, Switzerland, 2017; p. 36. [Google Scholar]
  92. Li, P.; Cai, M.-F. Challenges and new insights for exploitation of deep underground metal mineral resources. Trans. Nonferrous Met. Soc. China 2021, 31, 3478–3505. [Google Scholar] [CrossRef]
  93. He, M.C. Application of HEMS cooling technology in deep mine heat hazard control. Min. Sci. Technol. 2009, 19, 269–275. [Google Scholar] [CrossRef]
  94. Keenan, J.; Kemp, D.; Owen, J. Corporate responsibility and the social risk of new mining technologies. Corp. Soc. Responsib. Environ. Manag. 2019, 26, 752–760. [Google Scholar] [CrossRef]
  95. Günther, F.; Mischo, H.; Lösch, R.; Grehl, S.; Güth, F. Increased safety in deep mining with IoT and autonomous robots. In Mining Goes Digital; Mueller, C., Assibey-Bonsu, W., Baafi, E., Dauber, C., Doran, C., Jerzy Jaszczuk, M., Nagovitsyn, O., Eds.; Taylor & Francis Group: London, UK, 2019; pp. 603–611. [Google Scholar]
  96. Govreau, J. Constraints Mount to Slow Mining Project Growth in 25 January 2024. Eng. Min. J. 2024, 225, 16–23.98. [Google Scholar]
  97. Marshall, J.; Bonchis, A.; Neobot, E.; Scheding, S. Robotics in Mining. In Springer Handbook of Robotic; Springer: Berlin/Heidelberg, Germany, 2016; pp. 1549–1576. [Google Scholar]
Figure 1. Autonomous drilling rig connected to GNSS for real time positioning, navigation, and timing functions.
Figure 1. Autonomous drilling rig connected to GNSS for real time positioning, navigation, and timing functions.
Mining 05 00005 g001
Figure 2. Laboratory testing of an ABB robot grabbing a roof bolt for installation (adapted from Kolapo [79]).
Figure 2. Laboratory testing of an ABB robot grabbing a roof bolt for installation (adapted from Kolapo [79]).
Mining 05 00005 g002
Figure 3. The human–machine interface to autonomous roof bolter (adapted from Kolapo [79]).
Figure 3. The human–machine interface to autonomous roof bolter (adapted from Kolapo [79]).
Mining 05 00005 g003
Figure 4. The developed HMI for an autonomous bolting process.
Figure 4. The developed HMI for an autonomous bolting process.
Mining 05 00005 g004
Figure 5. Global mining projects by commodity (adapted from Govreau [96]).
Figure 5. Global mining projects by commodity (adapted from Govreau [96]).
Mining 05 00005 g005
Table 1. Applications of innovative technologies in mining operations.
Table 1. Applications of innovative technologies in mining operations.
Autonomous EquipmentArea of Usage in Mining OperationsReferences
Selected Automated Mining OperationsAutonomous drilling[23,24]
Explosive charging robots[19,20]
Loading, hauling and Dumping[9,10,11,24]
Conveyor belt inspection and monitoring robots[12,13]
LiDAR and Positioning[16,17,18,25]
Automated Longwall Mining[21,22,26]
GNSS Applications[14,15]
Table 2. Applications of ML methods in mineral prospecting and exploration.
Table 2. Applications of ML methods in mineral prospecting and exploration.
ReferenceMineral Exploration and ProspectingML Model(s)Model Performance(s)
Mineral Resource MappingMine Cost EstimationMine PlanningMine Reserve and Grade Estimation
Lee and Oh [29] ANNAccuracy > 70%
Tessema [33] ANN *, fuzzy-WofEMSE = 0.0937
SSE = 170.477
Sun et al. [34] ANN, SVM, RF *MSE = 0.039, Accuracy = 96.03%, Kappa = 92.06%
Afeni et al. [42] ANNSE = 0.015, R2 = 0.819, MAE = 2.023
Atalay [46] XGBoostRMSE = 0.69, MAE = 0.65
Jalloh et al. [41] ANNR2 = 0.8807, MSE = 0.2087
Nourali and Osanloo [48] RTRMSE = 219.36, MAE = 178.5
Zhang et al. [49] ANN, DNN, ACO-DNN *R2 = 0.991, MAPE = 0.072, VAF = 99.052
Zheng et al. [51] CFNN, SalpSO-CFNN *R2 = 0.980, MAE = 179.567, RMSE = 248.401
Ajak et al. [47] KNN, DT *, SVM, RF, LR, NBAccuracy = 72%, Probability = 78.6%
Nobahar et al. [52] KNN, DT, LR, RF, GB *Accuracy = 83%
Note: *: best performing model; WofE: weights-of-evidence, SVM: support vector machine, RF: random forest, XGBoost: extreme gradient boosting, RT: regression trees, DNN: deep neural network, ACO-DNN: ant colony optimization-deep neural network, CFNN: cascade forward neural network, SalpSO-CFNN: Salp swarm optimization-cascade forward neural network, KNN: k-nearest neighbor, DT: decision trees, LR: linear regression, NB: Naïve Bayes, GB: gradient boosting; MSE: mean square error; SSE: sum of squared errors; RMSE: root mean square error; MAE: mean absolute error; MAPE: mean absolute percentage error; VAF: variable accounted for; R2: coefficient of determination.
Table 3. Applications of ML methods in mineral excavation.
Table 3. Applications of ML methods in mineral excavation.
ReferenceMineral ExcavationML Model(s)Model Performance(s)
Geomechanical TestingDrilling and BlastingSlope StabilityMine Reclamation and Closure
Ogunsola et al. [55] ANN, ANN-GOA *, ANN-SSA, ANN-AOAR = 0.98498, MSE = 0.0036, VAF = 97.02%
Lawal et al. [54] ANN *, MARS, GA, RMSEUCS = 0.1248, RMSETS = 0.0332, RMSESS = 0.0639, RMSEYM = 2.91108
Skentou et al. [58] ANN *, ANN-ICA, ANN-PSOR = 0.9607 RMSE = 14.8272
Lawal et al. [56] ANN, MVO-ANN *, SSA-ANNRMSE = 0.35083 MAE = 0.263396
Komadja et al. [64] CART, SVR, MARS *RMSEBIGV = 0.227, R2BIGV = 0.951
Amiri et al. [67] ANN, ANN-KNN *R2BIGV = 0.88
R2AOP = 0.95
Amoako et al. [68] ANN, SVRMSEFRAG = 0.0031
Ogunsola et al. [63] ANNMAEBIGV = 0.1185, MSEBIGV = 0.0316
Maxwell et al. [70] SVM *, RF, CART, KNNAccuracy = 86.6%
Li et al. [71] ANN, SVM *, RFAccuracy = 87.34%
Bui et al. [72] ANN, SVR, M5Rules-GA *, ANN-PSO, ANN-ICARMSE = 0.024, R2 = 0.983, VAF = 98.26
Note: *: best performing model; ANN-GOA: artificial neural network-grasshopper optimization algorithm, ANN-SSA: artificial neural network-Salp swarm algorithm, ANN-AOA: artificial neural network-arithmetic optimization algorithm, GA: genetic algorithm, MARS: multivariate adaptive regression spline, ANN-ICA: artificial neural network-imperialist competitive algorithm, ANN-PSO: artificial neural network-particle swarm optimization, MVO-ANN: multiverse optimization-artificial neural network, SSA-ANN: Salp swarm algorithm-artificial neural network, CART: classification and regression trees; R: correlation coefficient.
Table 4. Applications of ML methods in mineral processing.
Table 4. Applications of ML methods in mineral processing.
ReferenceMineral ProcessingML Model(s)Model Performance(s)
Hosseini et al. [77]Final concentrate gradeANNR2 = 0.98
Bendaouia et al. [76]Floatation frothANN, RF *, GB, SVM, DT, LRRMSE = 5.40, MAE = 4.58
Zarie et al. [74]Floatation frothANN, CNN *Accuracy = 93.1%
Jahedsaravani et al. [78]Batch floatation of copper sulfideANN *, ANFISRCu = 0.77, GCu = 0.84, Rm = 0.88, Rw = 0.87
Note: *: best performing model; RCu: Cu recovery (%), GCu: concentrate grade (%), Rm: mass recovery (%), Rw: water recovery (%), ANFIS: adaptive neuro-fuzzy inference system, CNN: convolutional neural network.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kolapo, P.; Ogunsola, N.O.; Komolafe, K.; Omole, D.D. Envisioning Human–Machine Relationship Towards Mining of the Future: An Overview. Mining 2025, 5, 5. https://doi.org/10.3390/mining5010005

AMA Style

Kolapo P, Ogunsola NO, Komolafe K, Omole DD. Envisioning Human–Machine Relationship Towards Mining of the Future: An Overview. Mining. 2025; 5(1):5. https://doi.org/10.3390/mining5010005

Chicago/Turabian Style

Kolapo, Peter, Nafiu Olanrewaju Ogunsola, Kayode Komolafe, and Dare Daniel Omole. 2025. "Envisioning Human–Machine Relationship Towards Mining of the Future: An Overview" Mining 5, no. 1: 5. https://doi.org/10.3390/mining5010005

APA Style

Kolapo, P., Ogunsola, N. O., Komolafe, K., & Omole, D. D. (2025). Envisioning Human–Machine Relationship Towards Mining of the Future: An Overview. Mining, 5(1), 5. https://doi.org/10.3390/mining5010005

Article Metrics

Back to TopTop