Experiments in multiobjective fuzzy control of hybrid automotive engines

Hybrid electric vehicles (HEV) generate the power required to drive the vehicle and to operate on-board accessories by a combination of internal-combustion devices and electrical generators. The power train controllers for HEV regulate and coordinate the operation of power sources and power reservoirs, such as flywheels or batteries, to assure efficient operation that complies with various operational constraints. We present the result of the application of fuzzy-logic techniques to various aspects of this control problem focusing on questions related to the tracking of multiple reference signals while satisfying other elastic operational constraints. Our discussion encompasses theoretical developments leading to methods for the rapid derivation of generalized control surfaces and practical methods for the approximation and refinement, by clustering and learning techniques, of control surfaces or of existing controllers. Our learning techniques are generalizations of previously reported neuro-fuzzy techniques for the automated generation of fuzzy-control rules