Smart Mining Operations -- AI powered swarm robotics

Challenge

This project represented mining operations conducted by AI-powered swarm robotic systems with three types of robots:

• Excavators: Robots with two arms (chisel or saw configurations). An excavator with a chisel and a saw can operate independently; other configurations require teamwork.
• Collectors: Two configurations — either they carry ore themselves (acting as collectors and haulers), or they team up with a separate hauler.
• Surveyors: Used for mapping purposes, with operational decisions based on distance to other robots and scanning frequency.

The challenge was to study the operational behavior of these robots for a given layout and understand how different setups impact performance metrics. The mine layout configuration also plays an important role in the coordination strategy.

Additionally, conventional equipment (loaders and haulers) is used to transport material from underground stockpiles up to a processing circuit.

Solution

A model was built in AnyLogic using the material handling library, the fluid library, and nearly 30 different agents for the full model architecture.

Automatic Layout Generation

One of the most important features was building the mine layout automatically, as the swarming strategy would change depending on the configuration of stopes, corridors, and movement space. The layout was built entirely in an Excel file that AnyLogic reads to construct the mine configuration before simulation start.

Expert System for Robot Coordination

As this was the first step toward smarter swarming behavior with AI, an expert system with behavioral rules was built for the team work strategy. This was integrated with AnyLogic's material handling library obstacle avoidance, achieving performance results similar to what would be expected with AI.

Outcome

The simulation was used to understand the performance of different robot combinations for different layouts, which was very relevant as the operational results were unknown and unknowable without simulation. As some robots didn't even exist yet, it was also used to determine what robots to build first, as some combinations proved to work better with certain robot types.

Project Features

• Industry: Mining, Robotics
• Model: Discrete Events
• Duration: 1 month