Decentralized Adaptive Fuzzy Finite-Time Event-Triggered Control for Interconnected Nonlinear Systems Subject to Input Saturation

Decentralized event‐triggered output feedback adaptive neural network control for a class of MIMO uncertain strict‐feedback nonlinear systems with input saturation

“…The 𝜎 parameters are selected as 𝜎 τi = 2 and 𝜎 𝜆 i = 3 and 𝜖 i = 0.1, one should note that the sigma parameters need to be selected positive to prevent the drift of the parameters. The feedback gain as well as the observer-gain vectors are chosen on the basis of the desired poles location using the Ackerman method, respectively, as follows: k i = [6,11,6] and L i = [100,100, 100]. The low-pass filter is chosen to be a stable transfer function as T i (p) =…”

“…These terms were considered to have a specific mathematical form and were bounded by either known or unknown nonlinear functions to simplify the stability analysis. This was evident in various references such as References 2,4,5,7,8,10,11,13,20,[23][24][25] As far as we know, there are no findings on the implementation of event-triggered NNs adaptive control for LS-SFNLS with input saturation without using recursive design, while incorporating high approximation units.…”

“…• In contrast to References 2,4,5,7,8,10,11,13,20,[23][24][25], where restrictive assumptions regarding interconnection terms were imposed, in the proposed approach, the interconnection terms are only assumed reasonably bounded without any additional restrictive conditions.…”

“…To avoid the serious performance degradation or even the instability of the closed loop of LS‐SFNS, input saturation was considered in the controller design using $$$ \tanh \left(\cdotp \right) $$$ approximation and following transformational, 9 dynamic programming 10 and backstepping 11 approaches. Input hysteresis nonlinearity was dealt with in Reference 12 by using the backstepping approach.…”

“…One major limitation of the previously discussed literature on event‐triggered based approaches for LS‐SFNLS is the unrealistic assumption made regarding the control gains of interconnected strict‐feedback nonlinear systems. This assumption was made to simplify the design and analysis of the controllers, with the control gains being considered only as unity in References 5–8,11,13,17,19, and 20, known constants in References 4 and 15 and known nonlinear functions in Reference 12, which excessively hard‐limit the application of the developed approaches. In addition, the event‐triggered based approaches also face the well‐known issue of complexity explosion, which results from the repetitive derivation of virtual controllers through back‐stepping recursive design.…”