The first step in automating machining systems was the introduction of computer numerically controlled (CNC) machine tools. The primary function of CNC is to automatically execute a sequence of multiaxis motions according to a part geometry. However, safe, optimal, and accurate machining processes are generally planned by manufacturing engineers based on their experience and understanding of the process. It is difficult to predict vibration, tool wear and breakage, thermal deformation of the machine tools, and similar process-based events by using off-line theoretical models. In addition to engineering the process plans before actual machining, the machine tools are instrumented with vibration, temperature, displacement, force, vision, and laser sensors to improve the productivity and reliability of the cutting operations on-line. The sensors must have reliable frequency bandwidth, have a good signal-to-noise ratio, and provide signals with reliable correlation to the state of the process. They must also be practical for installation on machine tools. The measured sensor signals are processed by real-time monitoring and control algorithms, and the corrective actions are taken by the CNC accordingly. The corrective actions may be manipulation of spindle speed, feed, tool offsets, compensation of machine tool positions, feed stop, and tool change depending on the process monitoring and control application. Such a sensor-assisted cutting is called intelligent machining in the literature [16, 17]. The architecture of CNC must be organized in such a way that it allows real-time manipulation of the machine tool's operating conditions.