MSc Robotics Final Project by Johanna Amunyela

I need to plan on how to  benchmark indicators or measures in order to perform  the indexing on the mining rate. External and Internal Research External  Research Benchmark against other mining institutions. literature review Internal Research Geological experts on the team  Process Managers  Drillers  Navigation Officers to do: develop data collection methods   Optimisation problems  What are the factors that are affecting the performance? Mechanical Factor  Geological Factor  Human Factor  Keywords defined 100*100 meter block = 10000 meters squared Debligon  Quadrant  Move time  Presentation Time  Overburden  Footprint  Area Covered  Actual Area Effective Area Predetermined rate Effective rate  Permutations 

Automation of decision support systems  CNN -Convolutional Neural Network RNN -Recurrent Neural network LSTM - Long Short Term Memory  Deep Reinforcement Learning  Adversarial Learning  Pattern recognition  Vertix classification  NETTETION Model Algebra and differential calculus  literature review  Self-Attentive Attributed Network Embedding Through Adversarial Learning Deep Learning in Mining and Mineral Processing Operations: A Review Reference  https://blog.prototypr.io/mining-companies-using-ai-machine-learning-and-robots-e6dcdebaccc3 https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8970684

  Goal: from fluctuating mining rate performance to reducing its variability,  it stabilises frequent peak performance. process state improvement  Overall objectives:   • Maximising throughput  • Process stability  • Increased mining rate stability  • Decreasing equipment wear • Improving energy efficiency • Predicting failure  • Increase delivery on targets   

To do            Information Gathering             System States Descriptions             Possible Decitions             Quality Measure             Rewards/ Expected Performance  Findings  Project title:  Investigating improving the drill ship's mining rate with an autonomous mining path predicting Machine Learning Model. Project description: using the historic and current mining time series data and environmental parameters to create a Machine learning model than will perform a few days/ a week's mining rate performance forecast. Key areas: -Mining Rate = Area covered                             Time -Overlaps                ...