Computation & Automation Mining Laboratory (CAML)
The Computation & Automation Mining Laboratory (CAML) at New Mexico Tech was established to advance research, education, and training in underground mine safety, automation, and emergency response. The laboratory addresses a major challenge in mining research: limited access to operational underground mines for large-scale experimentation because of safety, regulatory, and production constraints. To provide a realistic yet controlled research environment, CAML integrates custom-built physical infrastructure, immersive virtual environments, advanced sensing systems, and data-driven technologies.
A central component of CAML is a custom-built room-and-pillar mine simulation rig designed to reproduce underground mine geometries and emergency conditions, including fire and evacuation scenarios. The rig supports systematic evaluation of evacuation strategies, human–system interaction, safety-system performance, and decision-making under stressful and time-critical conditions. CAML also incorporates omnidirectional treadmills and immersive Virtual Reality (VR) environments to enable physically realistic walking and navigation through simulated underground mines. Integrated with the laboratory’s simulation databases, these systems support studies of human behavior, stress response, spatial awareness, evacuation performance, and rescue decision-making in realistic emergency scenarios.
A 4,200-square-foot bunker-style Mine Safety Laboratory provides a physical test environment for underground safety, robotics, and autonomous-system research. The multi-level facility incorporates variable surface roughness and room-and-pillar configurations representative of underground mine environments. It supports controlled evaluation of Unmanned Ground Vehicles (UGVs), Unmanned Aerial Vehicles (UAVs), sensing platforms, and navigation algorithms under confined and safety-critical conditions. A digital twin of the bunker-style facility provides a synchronized virtual representation of the laboratory geometry, environmental conditions, and sensor systems. This physical–virtual integration enables scalable experimentation in autonomous navigation, human–robot interaction, sensor evaluation, and safety-system performance. It also allows rapid testing of operational and emergency scenarios, supporting training and decision-support research without requiring repeated physical reconfiguration of the laboratory.
Virtual and Augmented Reality are also central research capabilities. Unity-based environments are integrated with CAD models, haptic interfaces, and omnidirectional treadmills to support mine navigation, rescue training, emergency response, and human–machine interaction studies. Research also examines physiological and cognitive responses to immersive environments, including cybersickness and other factors that can affect VR effectiveness in safety-critical applications. Together, these resources provide an integrated experimental environment in which physical facilities, autonomous platforms, sensing technologies, digital twins, and immersive simulation can be evaluated within a unified framework. The laboratory serves as an interdisciplinary testbed for advancing underground mine safety, autonomous monitoring, human–machine interaction, and emergency-response technologies.
Funded by NIOSH (CDC), this project develops a semi-autonomous robotic system for rapid reconnaissance, environmental monitoring, and trapped-miner localization during underground mine emergencies. The platform combines an Unmanned Ground Vehicle (UGV), an aerial drone, deployable wireless communication nodes, and distributed environmental sensors to support search-and-rescue operations in hazardous, GPS-denied environments.
The UGV serves as the primary mobile platform and carries a drone for extended exploration and data transmission. Deployable sensor packages inspired by ostrich eggs monitor critical environmental conditions, including carbon dioxide, carbon monoxide, oxygen, dust, humidity, and temperature, providing real-time information for navigation and rescue decision-making. LiDAR and depth cameras support autonomous navigation in dark and visually degraded underground environments, while backscatter-based communication technologies are investigated to improve information exchange between robotic platforms, rescue teams, and trapped miners.
The project also explores bio-inspired precision landing methods that enable the drone to autonomously return to and land on the UGV under challenging underground conditions. Together, these technologies form an integrated robotic framework for improving situational awareness, communication, and operational efficiency during underground mine search-and-rescue missions.
Research in the laboratory integrates acoustic and elastic-wave sensing, seismic imaging, and machine learning to support early hazard detection and post-disaster rescue decision-making. Ongoing work focuses on multi-modal sensing frameworks that combine microseismic monitoring, distributed acoustic sensing (DAS), and geophone arrays to identify early signs of instability, classify underground events, and support real-time situational awareness. These sensing systems are coupled with machine learning models for hazard recognition, event characterization, and decision support in dynamic underground environments. Complementing this effort, a laboratory-scale ultrasonic seismic imaging and rock physics platform is being developed to enable high-resolution tomographic imaging of rock structures and controlled integration with robotic motion, supporting advances in autonomous inspection and subsurface characterization.