To interpret sensor signals like images, image sequences, or continuou
s speech the representation and use of task-specific knowledge is nece
ssary. The paper presents a framework for the representation of declar
ative and procedural knowledge using a suitable definition of a semant
ic network. Based on that formalism a problem-independent control algo
rithm for the interpretation of sensor signals is presented. It provid
es both data-driven and model-driven control structures which can easi
ly be combined to perform any mixed strategy. An explanation facility
is available which makes the development of complex knowledge bases ea
sier and increases the acceptance of such a knowledge-based analysis s
ystem.