Knowledge-based techniques in instrumentation

Opportunities for the application of knowledge-based techniques in Fourier transform infrared (FTIR) spectrometry are examined. Three issues are considered: (1) the new functionalities that can be implemented by artificial intelligence (AI) techniques; (2) the kind of AI techniques that can be used in this specific problem; and (3) how to merge AI techniques with existing, relatively small instruments, such as FTIR systems. The ADVICE facility developed to support the experiment design phase of the FTIR operation is described. Spectrum interpretation is discussed.<>