Abstract:
In this thesis, we considered the tracking control of a class of multi-input multioutput (MIMO) nonlinear uncertain systems with faults. At the beginning, general
ideas of adaptive fault-tolerant with the best researchers who worked in this field
and the used techniques are presented. Based on these techniques we proposed three
adaptive control schemes. The first scheme is based on backstepping and fuzzy logic
systems. This scheme deals with the occurrence of actuator and sensor faults at the same
time which the performances are tested on a dynamic model of a quadrotor system. The
second integrates Nussbaum-type functions to overcome the control gain sign problem
with time-varying and state-dependent actuator faults, the performances are tested on
the dynamic of two-inverted pendulums system. The third control scheme is based on an
optimal technique who is called Particle Swarm Optimization PSO. This optimal technique
is used to circumvent the problem of the adaptive parameters and the initial parameters
of the used fuzzy systems, a simulation stage was applied on two links-robot manipulators
to prove the accuracy and the effectiveness of the proposed approach. The analysis of
stability and robustness for all the proposed control schemes are performed by using the
Lyapunov synthesis method with the help of Barbalat’s Lemma, and the simulation results
are given to highlight its performance
