Towards Autonomous Excavation of Fragmented Rock : Experiments , Modelling , Identification and Control by Joshua

Towards Autonomous Excavation of Fragmented Rock: Experiments, Modelling, Identification and Control Joshua Alexander Marshall Master of Science (Engineering) Department of Mechanical Engineering Queen’s University August, 2001 Increased competition and recent globalization in the minerals industry has resulted in further demands for advanced mining equipment technology. The autonomous excavation problem for fragmented rock entails the design of a control system capable of regulating the complicated bucket-rock interactions that occur during excavation operations. A notable amount of previous work has been performed to address the excavation automation problem, although this work has mainly focused on the excavation of granular material such as soil rather than on rock. Moreover, none has resulted in widespread industry adoption of automated rock loading technologies. This thesis revisits the autonomous excavation problem for fragmented rock from a fresh perspective, and in particular, focuses on the problem of autonomous excavation using load-haul-dump (LHD) underground mining machines. First, an extensive review of the state-of-the-art in autonomous excavation is provided. Then, results of pioneering full-scale experimental studies are provided. These studies were carried out with the intent of identifying the evolution of machine parameters during freespace motions of the employed LHD mechanism, and during a selection of excavation trials conducted by skilled operators in fragmented rock typical of an underground hard-rock mining scenario. It was discovered that information contained within actuating cylinder pressure signals might offer a means for identification of the bucket-rock interaction status. Having reviewed the conventional techniques for development of robot dynamical equations of motion, the results are presented of modelling the LHD loader mechanism