Algoryx in research collaboration to develop autonomous machines for risky mining environments

In a new research project, the Robotics and AI team at Luleå University of Technology is joining forces with mining equipment manufacturer Epiroc and Algoryx to develop advanced autonomous navigation capabilities and sensor technology for mining vehicles at great depths. 

The global mining industry is facing increasing demand for metallic commodities, especially rare earth metals, which play a crucial role in the transition from fossil to renewable energy sources. 

The majority of newly discovered ore deposits are located far down in the ground, at depths of more than 300 metres, and in Sweden there are now plans to mine at depths of more than 2,000 metres. However, mining at these depths involves increased risks and significantly higher costs for creating a safe and good working environment in the mine.

“We are doing so much more than autonomy, essentially, we are accelerating machines to a new level of technology”.

George Nikolakopoulos, Professor of Robotics and AI, Luleå University of Technology,

Autonomous mining technology Is seen as a future solution to these challenges. But today’s autonomous underground mining machines are still insufficiently capable of reading their surroundings, as well as planning routes and movements autonomously.

Making machines better at perceiving their surroundings

In a new research project called ReNAM, Algoryx is joining Epiroc and Luleå University of Technology in an effort to solve the challenges of safe and effective deep mining. The aim of the project is to develop new sensor technology and autonomous control for an Epiroc mining vehicle. 

George Nikolakopoulos, Professor of Robotics and AI at Luleå University of Technology, characterizes the project as a groundbreaking effort: 

“We are doing so much more than autonomy, essentially, we are accelerating machines to a new level of technology. The overall quality of Swedish mines, efficiency and safety will be unparalleled”

From digital twins in simulation to real machines in a physical mine

The ReNam project has received funding from the Swedish innovation agency Vinnova and will run from 2023 to the end of 2025. The goal of the project is to develop and validate two use cases that can lay a foundation for future robotics research, and contribute to Epiroc’s future development of commercial products with fully developed machine autonomy.

“Through this partnership, we are contributing to a new generation of machines with improved autonomy capabilities. This is an exciting addition to the ongoing development of our automation offering, which helps to further improve safety and productivity for our customers” says Åsa Gabrielsson, VP of Research & Development at Epiroc’s Underground division.

In the first phase of the ReNAM project, realistic digital twins of mining machines and surroundings are modeled in a physics-based simulation environment provided by Algoryx Simulation.

A concept illustration of a typical scenario that will be used to develop autonomous environment perception and path planning capabilities for mining machines. Image credit: Luleå Institute of Technology.

Once the sensor- and AI-equipped virtual mining machines have been developed to the point where they can solve tasks safely and reliably in a digitally simulated environment, the next step will be to transfer the developed control systems to Epiroc’s physical mining machine and have it perform the same tasks in a real-life test mine.

“Through realistic simulations and with safety in mind, where the AI can be trained on the whole system, the machine that is, with its environment and real forces, contacts and behaviors, the outcome will be superior to most projects we see in the world. In order to be successful, collaboration between universities, high-tech, and industrial companies is extremely important”, says Urban Wikman, Business Director at Algoryx.

“In order to be successful, collaboration between universities, high-tech, and industrial companies is extremely important.”

Urban Wikman, Business Director, Algoryx