Flight Control Reconfiguration Based on Online Physical Model Identification and Nonlinear Dynamic Inversion

摘要: There are many control approaches possible in order to achieve fault tolerant flight control. An important aspect of these algorithms is that they should not only be robust, but even adaptive some way, adapt the faulty situation, see Ref. [1] and [5] literature. In category algorithms, a distinction made between indirect direct Indirect involves two stages. First, an estimate plant model generated online. Once available, it used generate controller parameters. Instead estimating model, algorithm estimates parameters directly controller. This can done via main approaches: output error input error. Of both categories mentioned here, preferable due its flexibility property being based. categories, there also subversions, namely reference (MRAC) self-tuning (STC). former, one relies on works minimizing tracking (such as concept sliding mode control). With feasible three goals, trim value adjustment for inputs, decoupling inputs outputs closed loop pilot commands, [1]. Self-tuning focuses adapting (PID) gains by making use estimated parameter values known more flexible, [21]. Currently, much research performed field control, where adaptation extensive than tuning PID gains. One new possibilities predictive (AMPC), which interesting thanks nature deal with (input) inequality constraints. These constraints good representation actuator faults. It noted have been already successful applications MPC [10] [14]. alternative nonlinear approach discussed this chapter, allows develop reconfigurable routine placing emphasis physical models, thus producing internal physically interpretable at any time.