The Bureau of Mines is becoming increasingly involved in the areas of automation and robotics for the U.S. mining industry to increase both mining productivity and safety. Both robotization of existing mining equipment and future integrated robotic and/or automated mining systems are being investigated. One of the primary concerns associated with automated underground coal mining is the development of an appropriate coal-rock interface detection (cid) sensing system to determine when a mining machine is mining coal or starting to mine adjacent strata (e.g., roof or floor). This paper describes two cid methods associated with vibrations generated by mining machines and how adaptive signal discrimination (asd) technology is being used to process these signals. In these cid methods, strata (roof, coal seam, floor) or mining machine vibrations are monitored for complex signals generated, which vary according to the type of geologic material being cut and the type of mining machine being used. These signals are then analyzed using sophisticated state-of- the-art asd technology. It is this advanced signal analysis that distinguishes this approach from those of prior research. The asd system is initially "trained" using a database of features extracted from known signals measured under conditions of interest (e.g., "cutting coal," "starting to cut roof," etc.). Subsequently, it uses this database to determine if new, unknown signals belong to a given condition.