A System for Ore Tracking in Autonomous Mining

摘要

A probabilistic system for autonomous tracking of excavated ore in mining would be highly valuable.rnCurrently, mine sites can have a vast array of sensors; examples include excavator load cells, bucketrnvolume estimators and haul truck strut pressures which can be used to gather information about thernexcavated ore at different locations. These sensors can be utilised effi ciently to increase the accuracyrnof mine estimates by introducing a real-time ore tracking system. This requires the introduction of arnprobabilistic ore tracking framework to consistently fuse information from any sensor input. Currentrnmodelling practices are manually intensive, deterministic (no recognition of the quality of data) andrngenerally not a complete end-to-end system. This paper will provide a novel method for trackingrnextensive ore properties (mass, volume) and intensive ore properties such as chemical compositionrn(eg Fe per cent, SiO2 per cent, Al2O3 per cent). This method is proved to ensure that correctrncorrelations between states are conserved as material is transferred, combined and/or separated atrndifferent locations with any confi guration of sensor inputs. This ensures that consistency of estimatesrnis maintained and other benefi ts such as real-time probabilistic reconciliation can be implemented.rnA method for combining material of different mass and chemical compositions is also developed tornenable accurate and effi cient operation of the system. Representative experiments on a small-scalernand a larger fi eld trial which mimic the mining processes of excavation, haul and unload to ROMrnStockpile will be used to verify the approach.