查看文章

Model-based engineering becomes more and more important in industrial practice. System identification is a vital technology for producing the necessary models, and has been an active area of research and applications in the automatic control community during half a century. At the same time, increasing demands require the area to constantly develop and sharpen its tools. This paper deals with how system identification does that by amalgamating concepts, features and methods from other fields. It describes encounters with four areas in systems theory and engineering: Networked Systems, Particle Filtering Techniques, Sparsity and Compressed Sensing, and Machine Learning. The impacts on System Identification methodology by these encounters are described and illustrated.