Autonomous Robotic Systems in Underground Mining Environments

Autonomous robotic systems in underground mining environments are transforming the extraction and monitoring of subsurface resources by reducing the need for humans to operate in hazardous conditions. These systems integrate advanced sensor arrays—such as LiDAR, stereo and thermal cameras, inertial measurement units and gas detectors—with robust algorithms for simultaneous localisation and mapping (SLAM), path planning and obstacle avoidance in GNSS-denied, feature-sparse tunnels. Ruggedised and explosion-proof hardware ensures reliable operation under conditions of high humidity, dust and variable temperatures. Applications span continuous inspection of conveyor belts, structural integrity assessment of support beams, search-and-rescue tasks and environmental hazard detection. By automating these functions, underground platforms not only minimise exposure to rock falls, noxious gases and mechanical failures but also enable round-the-clock operation, higher productivity and real-time situational awareness. Ongoing advances in 3D mapping, loop-closure recognition and teleoperation are paving the way towards fully autonomous subsurface excavations and maintenance routines on a global scale.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent developments in underground navigation and mapping have achieved significant improvements in accuracy and efficiency. A graph-based SLAM framework utilising Generalised Iterative Closest Point (GICP) registration merges consecutive-frame and loop-closure constraints with roadway-plane extraction to suppress noise and deliver sub-decimetre localisation drift in complex tunnel networks, all without inertial sensors. The AMICOS project combines infrared thermography with advanced image-processing algorithms on an unmanned ground vehicle to support post-accident search-and-rescue operations; by fusing thermal signatures and object-recognition techniques, the system can detect trapped individuals and assess gas or heat hazards for rapid remote intervention. An enhanced SLAM algorithm, LeGO-LOAM-SC, integrates a light-weight LiDAR odometry pipeline with a scan context global descriptor to improve loop detection and graph optimisation. Trials in underground parking structures and mine simulation corridors demonstrate pronounced reductions in trajectory error, lower computational demands and real-time performance on embedded platforms.

Autonomous Robotic Systems in Underground Mining Environments publication trend

The graph below shows the total number of articles in autonomous robotic systems in underground mining environments across all publications each year (not limited to Nature Index journals).

Technical terms

SLAM: Simultaneous localisation and mapping; concurrently builds a map of the environment and estimates the robot’s pose within it.

LiDAR: Light detection and ranging sensor technology that emits laser pulses to measure distances and generate three-dimensional point clouds.

UGV: Unmanned ground vehicle designed for autonomous or remote-controlled operation in confined underground spaces.

Scan context: A global descriptor capturing the structural layout of LiDAR scans to enable efficient loop-closure detection in SLAM.

References

1. Robust GICP-Based 3D LiDAR SLAM for Underground Mining Environment. Sensors (2019).
2. Application of the Infrared Thermography and Unmanned Ground Vehicle for Rescue Action Support in Underground Mine—The AMICOS Project. Remote Sensing (2020).
3. LeGO-LOAM-SC: An Improved Simultaneous Localization and Mapping Method Fusing LeGO-LOAM and Scan Context for Underground Coalmine. Sensors (2022).
4. A Mobile Robot-Based System for Automatic Inspection of Belt Conveyors in Mining Industry. Energies (2022).