<title>Underground mining and construction robotics</title>

Abstract

The benefits of sensor based computer control in the international industry are many. Consequently, the industry expends considerable effort researching and developing automated systems. These efforts span the total industry including exploration, extraction, processing, and refining. All of the major segments of the industry: underground and surface, industrial minerals, metal, non-metal, and coal have participated in these developments. The result is a large body of information that is impossible to cover completely in a conference paper. To provide interesting information to the conference audience, this paper focuses on efforts to automate mobile production equipment. Most of the efforts apply to international underground metal and coal mines. Mobile and construction equipment presently utilize mechanization and remote control. For example, remote controlled equipment, where an operator is stationed some distance from the machine, but within line-of-sight, is found on many continuous machines in underground coal mines and LHD'S in underground metal mines. Teleoperated equipment where operators can be stationed further away, beyond-line-of-sight, is currently being tried at mines in the USA, Canada, Australia, and Sweden. Research projects currently focus on intelligent analysis of data from sensors to produce control algorithms which will be termed mining robotics to distinguish it from simpler closed ioop control termed automation. The reason current research focuses on intelligent analysis of sensor data to produce control algorithms is that takes place in the geological environment where conditions are highly variable and unpredictable. As a result, systems must have substantial cognitive abilities to recognize and deal with these unpredictable variations. Robotics has been described as the intelligent connection between perception and action. It may provide the greatest benefits in productivity and safety since it incorporate artificial-intelligence-based algorithms that show promise for adapting to the dynamic environment. Many mines have purchased the latest available technology from equipment manufacturers. These large, expensive, advanced machines have high production capabilities and are very reliable. For these machines to provide the highest cost effectiveness, they should be fully utilized; however this is not always the case. For example, underground coal continuous miners are only utilized 17% of the shift (Suboleski and King, 1990, and King and Suboleski, 1991). These problems occur in surface and underground mines, but underground has the most opportunity for improvement. In addition, maintaining an artificial environment 348 / SPIE Vol. 1613 Mobile Robots VI (1991) 0-8 194-0750-X/92/$4.00 in an underground mine that is conducive to optimal work performance is expensive. By its nature, underground can be hot or humid, in addition, the equipment produces dust, noise, fumes, and other hazards. Explosive or toxic gases and the constant danger of rock falls add to the increasing list of health and safety concerns that have initiated volumes of federal and state regulations for maintaining safe work places for humans in underground mines. These regulations cause tremendous capital and operating expenditures that prevent many mineral deposits from being mined profitably. For example, in both underground coal and metal mines, miners excavate many more entries or drifts than necessary to produce and remove the ore or coal in order to provide a safe environment for humans. These difficulties will increase as mines deplete nearsurface reserves that have the best working conditions. As a result, some areas are not mined because of the expense of providing an adequate environment for human operators. A good example is narrow ore veins in metal mines that would normally require cut and fill stoping methods to provide adequate ground support. With teleoperated or robotic equipment, less expensive open stoping practices can be used. Another example is automated longwall systems that can be operated more freely in a bi-directional cutting mode because they eliminate respirable dust exposure on the face.