Adaptive Neuro-fuzzy Control of Active Vehicle Suspension Based on H2/H∞ Synthesis

This paper deals with a road-type-adaptive strategy for an active suspension control system to provide more satisfactory suspension performance. H2-synthesis is adopted when the road irregularities are modeled as impulses or white noise. Since this design ensures the minimum RMS of performance outputs for above road types, it will have a diminished performance for other road profiles. On the other hand, H∞-controller minimizes the RMS of performance outputs for worst case road irregularities and therefore cares conservatively for all road profiles. To exploit the benefits of both controllers to obtain a more satisfactory performance, an automatic switching between the controllers based on the road type is recommended. To this goal, through manual switching, the I/O data of the controllers are gathered. Training these data to an ANFIS network completes the task. Simulation results employing this ANFIS controller show considerable performance enhancements compared to either of H2 or H∞ controllers.