Automated Planning and Coordination of Autonomous Haulers in Underground Mines
| Reference number | |
| Coordinator | Örebro universitet - Akademin för naturvetenskap och teknik |
| Funding from Vinnova | SEK 4 095 000 |
| Project duration | May 2020 - April 2023 |
| Status | Completed |
| Venture | The strategic innovation programme for Swedish mining and metal producing industry - SIP Swedish Mining Innovation |
| Call | Towards a sustainable development in the mining and metal extraction industry |
Purpose and goal
AI methods have been developed to be able to plan and coordinate a fleet of autonomous LHD machines for material handling in a mining environment. These tools consist of planning and coordination approaches. Planning is to select which task is to be carried out and by which machine inorder to transport as much material as possible. Coordination is used to manage how the machines are allowed to run when several machines are in the same area. The methods are designed to be easily integrated into existing systems as few assumptions are made regarding how the machines are controlled.
Expected results and effects
Software has been developed that can be demonstrated in a simulator where a mining environment with multiple machines, unloading points and loading points can be simulated. One objective was a demonstration corresponding to TRL level 7; “system prototype demonstration in operational environment”. In the project, only tests were run in simulation and not in a real environment, However, extra emphasis has been placed on the simulation being correct and close to reality and the simulation provides opportunity to test other scenarios, e.g. having more machines running.
Planned approach and implementation
Parts of the tools that have been developed regarding coordination was initially based on an existing method developed at Örebro University. This method was used to test how the planning worked inorder to find the next assignment aiming to achieve an efficient flow. During the course of the project, flaws were also found in the simulation environment that were iteratively improved. A system to better estimate delays in a fleet has also been developed. The simulation and visualization is partly based on previous work combined with improvements.