Multimodel Control
Dumitru Popescu, Amira Gharbi, Dan Ştefănoiu, Pierre Borne · 2017
Automatic control is based on the concept of models that represent the dynamic behavior of a system. Given the difficulty of the task, a comprehensive approach was proposed, based on multiple models that were defined around different operating points. This multimodel is a representation that can be obtained either from an initial nonlinear model or directly from input-output data. To achieve the same level of performance at different operating points for the nonlinear feature, this chapter proposes a solution that involves using the notion of multimodel command based on a linguistic approach. Multimodel control is achieved by separately calculating the command for each of the models or submodels. Multimodel command is particularly suited to the trajectory tracking problem in a nonlinear system. When all models are of the same order and defined in the state space with a similar state vector, the Takagi-Sugeno approach, which is also based on fuzzy logic, is often used.