SVM-based approach for instrument fault accomodation in automotive systems

The paper deals with the use of support vector machines (SVMs) in software-based instrument fault accommodation schemes. A performance comparisons between SVMs and artificial neural networks (ANNs) is also reported. As an example, a real case study on an automotive system is presented. The ANNs and SVMs regression capability are employed to accommodate faults that could occur on main sensors involved in the engine operating. The obtained results prove the good behaviour of both tools. Similar performances have been achieved in terms of accuracy.